A STUDY INVESTIGATING THE DETERMINANTS OF CONSUMER BUYER BEHAVIOUR RELATING TO THE PURCHASE OF ORGANIC FOOD PRODUCTS IN URBAN CHINA Jue Chen Thesis submitted in fulfillment of the requirements of the degree of Doctor of Philosophy Faculty of Business and Enterprise Swinburne University of Technology 2012 ABSTRACT In the last two decades, organic food has developed as one of the fastest growing areas in the world and has assumed greater global awareness. Organic food is perceived as being more nutritious, healthier, safer and environmentally friendly. China has experienced dramatic economic growth, and is developing an affluent urban middle class which is attracted to the organic food market. As it is a highly populated country with great potential in production and consumption of organic food, the understanding of Chinese organic food consumers’ behaviour is critical for the global organic food market. Current literature examining consumers’ awareness of organic food has been well developed in North America and Western Europe. Hence the aim of this research is to investigate the various determinants and elements which influence consumers’ purchase behaviour of organic food products in urban China. A five stages conceptual framework was proposed for analysing consumer purchase decision making towards organic food in urban China. The five stages are influencing stage, cognitive/affective stage, evaluation of alternatives stage, behavioural intentions stage and purchase stage. The emphasis was placed on the influencing aspects and dimensions of this process in the Chinese context. The four constructs – namely, product, regulatory, lifestyle and ethnocentrism – were positioned in the influencing stage. The research design is predominantly quantitative in nature. The data was collected in China in two stages. Stage one involved 204 valid online surveys which were collected for the purpose of the pilot study. Stage two, the main study, involved 964 paper-based surveys which were collected in four major cities. The data analyses were performed using exploratory factor analysis, confirmatory factor analysis and structural equation modelling. The findings of this study have revealed that the influencing stage’s product-related, regulatory and lifestyle constructs directly or indirectly influence urban Chinese consumers’ beliefs/attitudes, pre-purchase evaluation and behavioural/purchase intention. The cognitive/affective stage of which beliefs and attitudes were the main component was found to be a significant predictor of pre-purchase evaluation. In the i evaluation of alternatives stage, pre-purchase evaluation was found to have a highly significant effect on behavioural/purchase intentions. This study is one of the very few associated with consumer buyer behaviour of organic food in urban China, and it was the first national organic food consumption survey in China. It has provided valuable insights into current studies of global and Chinese consumer behaviour towards organic food. Beneficiaries of this research study include stakeholders in China and globally such as consumers, vendors both local and international, and government agencies. ii ACKNOWLEDGEMENTS Thank you To those people who have encouraged me and inspired me during my PhD journey… From my homeland China, and now to my new home Australia. It has been a migrant’s dream… I can never thank enough what life has given to me… iii As a Chinese migrant to Australia, it has always been my dream to connect my homeland – China where I was born, where I grew up and where I worked for many years, to my new home – Australia. This study intends to make my dream come true. First of all, I would like to take this opportunity to thank my principal supervisor Associate Professor Antonio Lobo who provided me with tremendous support and invaluable guidance. Tony’s rigorous scholarship toward research influenced me throughout this study and will continually influence my future academic life. I must say I am very lucky to have had the best supervisor who always gave me whatever support whenever I needed it. Thanks also go to my associate supervisor Dr Bruno Mascitelli whose hard-working and unwavering passion for the academic world has inspired me during my academic journey. Without his continued courage and support, I would not have finished this study. Both Tony and Bruno are not only my supervisors during this thesis, they are my academic mentors who provided invaluable advice and support and also encouraged me to pursue, continue and complete this long PhD journey. Thanks also go to my third supervisor Professor Chongguang Li (from China), who gave me tremendous support and guidance during the data collection in China. Secondly, thanks go to people working in the organic industry who provided invaluable advice and inspiration during this study – German Organic Services GmbH director Mr Udo Censkowsky, Chinese National Professional Committee of Organic Health Industry Deputy Secretary-General Mr Mika Yuan (袁晓东), China ‘Organic farm’ Deputy General Manager Ms XiaoXue Li (李晓雪) and Australia Pure Harvest manager Ms Linda Liu – and to colleagues from the Victoria Organic Industry Committee Women’s Leadership group and especially to my industry mentor, VOICe chair Ms Liz Clay. I’d love to borrow her motto: live and breathe organic. Thirdly, I would like to acknowledge Swinburne University of Technology for awarding me with a scholarship (Swinburne University Postgraduate Research Award, SUPRA) that enabled me to complete this thesis. English is my second language and Ms Nancy Moncrieff provided invaluable assistance in improving my writing skills. Associate Professor Denny Meyer and Ms Natalia Rajendran assisted me with the statistical side of this research. I also wish to express my appreciation to Mr David iv Hudson for his proofreading of this thesis during the final stages of the manuscript. Many of my PhD colleagues in BA 701 inspired and energised me to continue the long journey. We had a great time, had so much fun and shared so much emotion and joy. Finally, I’d like to acknowledge the love, support and understanding of my family, my husband Frank and my son Mark. Particular thanks go to my parents, Shuzhen and Weimin who came to Australia to help me look after the family during the final year of the completion of this thesis to enable me to fully concentrate on finishing this study. I cannot believe that I have finished this journey. I still enjoy doing research and I feel as though it has just begun … v DECLARATION This thesis contains no material which has been accepted for the award to the candidate of any other degree or diploma in any university or other institution. To the best of my knowledge, this thesis contains no material previously published or written by another person, except where due reference is made in the text. Jue Chen Melbourne, Australia vi TABLE OF CONTENTS CHAPTER ONE: INTRODUCTION AND BACKGROUND ............................................. 1 1.1 Introduction .................................................................................................................... 1 1.2 Background of Chinese economic development ............................................................ 3 1.3 Evolution of the global organic food scenario ............................................................... 5 1.3.1 A brief history of the organic concept ................................................................... 5 1.3.2 A contemporary overview of the global organic food market ............................... 7 1.3.3 The nature and meaning of the term ‘organic food’ .............................................. 8 1.4 Changes in the Chinese economy ................................................................................ 10 1.4.1 China’s new urban consumers ............................................................................. 10 1.4.2 The organic industry in China .............................................................................. 12 1.5 Research problem and questions .................................................................................. 16 1.6 Academic and practical justification for this research ................................................. 18 1.6.1 Understanding Chinese culture ............................................................................ 18 1.6.2 Theoretical academic contribution ....................................................................... 18 1.6.3 Business implications and beneficiaries of this research ..................................... 19 1.7 Research methodology ................................................................................................. 20 1.8 Outline of the research ................................................................................................. 21 1.9 Chapter summary ......................................................................................................... 24 CHAPTER TWO: LITERATURE REVIEW ...................................................................... 25 2.1 Introduction .................................................................................................................. 25 2.2 Understanding organic food consumers ....................................................................... 26 2.2.1 Is organic food healthier? ..................................................................................... 26 2.2.2 Consumer awareness and knowledge of organic food ......................................... 27 2.2.3 Differences between buyers and non-buyers of organic food .............................. 28 2.3 Demographic profiles of organic food consumers ....................................................... 29 2.3.1 Gender .................................................................................................................. 30 2.3.2 Age ....................................................................................................................... 31 2.3.3 Education level ..................................................................................................... 31 2.3.4 Income level ......................................................................................................... 32 2.3.5 Role of the family................................................................................................. 32 2.4 Purchase motivations ................................................................................................... 33 2.4.1 Food attributes ...................................................................................................... 34 2.4.2 Trends and fashion ............................................................................................... 35 2.4.3 Health consciousness ........................................................................................... 35 2.4.4 Organic food information and knowledge ........................................................... 36 2.4.5 Impact on food safety issues ................................................................................ 37 2.4.6 Environmental concerns ....................................................................................... 38 2.4.7 Animal rights and social justice ........................................................................... 38 2.4.8 Country of origin .................................................................................................. 39 2.5 Barriers to organic food consumption .......................................................................... 40 2.5.1 Price...................................................................................................................... 40 2.5.2 Willingness to pay ................................................................................................ 41 2.5.3 Labelling and certification ................................................................................... 42 2.5.4 Lack of availability .............................................................................................. 43 2.6 Pre-purchase intention .................................................................................................. 44 vii 2.6.1 Attitudes ............................................................................................................... 44 2.6.2 Values................................................................................................................... 45 2.6.3 Beliefs .................................................................................................................. 46 2.6.4 Purchase intention ................................................................................................ 47 2.7 Theories associated with previous studies ................................................................... 48 2.7.1 The Consumer Decision Process (CDP) model applied to food consumption .... 50 2.7.2 Attitude theory and measurement ........................................................................ 62 2.7.3 Integration of three models: CDP, TPB and Hierarchy of Effects ....................... 67 2.8 Summary of previous studies ....................................................................................... 70 2.9 Limitation of previous studies...................................................................................... 71 2.10 Chapter summary ..................................................................................................... 73 CHAPTER THREE: DEVELOPMENT OF THE CONCEPTUAL FRAMEWORK AND RELATED HYPOTHESES ....................................................................................... 74 3.1 Introduction .................................................................................................................. 74 3.2 Development of the proposed conceptual framework ................................................. 75 3.3 Influencing stage .......................................................................................................... 80 3.3.1 The product construct ........................................................................................... 80 3.3.2 The regulatory construct ...................................................................................... 83 3.3.3 The lifestyle construct .......................................................................................... 87 3.3.4 The ethnocentrism construct ................................................................................ 90 3.4 Cognitive/affective stage .............................................................................................. 93 3.5 Evaluation of alternatives stage ................................................................................... 95 3.6 Behavioural/purchase intention stage .......................................................................... 96 3.7 Demographic variables................................................................................................. 97 3.8 Chapter summary ....................................................................................................... 102 CHAPTER FOUR: METHODOLOGY ............................................................................. 103 4.1 Introduction ................................................................................................................ 103 4.2 Research design .......................................................................................................... 104 4.2.1 Research paradigm ............................................................................................. 104 4.2.2 Research context ................................................................................................ 105 4.2.3 Unit of analysis .................................................................................................. 106 4.2.4 Refining the unit of analysis .............................................................................. 106 4.3 Design of the survey instrument ................................................................................ 107 4.3.1 Focus groups ...................................................................................................... 107 4.3.2 Operationalising the constructs .......................................................................... 108 4.3.3 Description of the survey instrument ................................................................. 116 4.3.4 Scaling and measurement................................................................................... 117 4.3.5 Validity and reliability of survey instrument ..................................................... 118 4.4 Data Collection........................................................................................................... 121 4.4.1 Ethics approval ................................................................................................... 121 4.4.2 Sampling ............................................................................................................ 122 4.4.3 Stage one: pilot study (online survey) ................................................................ 123 4.4.4 Stage two: main data collection (paper-based survey) ....................................... 124 4.4.5 Selection of cities in China for stage two: paper-based data collection ............. 125 4.5 Chapter summary ....................................................................................................... 128 viii CHAPTER FIVE: ANALYSES AND RESULTS OF THE PILOT STUDY (STUDY ONE) .................................................................................................................. 129 5.1 Introduction ................................................................................................................ 129 5.2 The objectives of the data analysis in the pilot study ................................................ 129 5.3 Choosing the right statistical method ......................................................................... 130 5.4 Identifying factors and latent variables ...................................................................... 132 5.4.1 Results of the EFA for the product-related construct......................................... 133 5.4.2 EFA results for the regulatory construct ............................................................ 135 5.4.3 EFA results for the lifestyle construct ................................................................ 136 5.4.4 EFA results for the ethnocentrism construct ...................................................... 137 5.4.5 EFA results for the beliefs and attitudes construct............................................. 138 5.4.6 EFA results for the pre-purchase evaluation construct ...................................... 138 5.4.7 EFA results for the behavioural/purchase intention construct ........................... 139 5.4.8 Consolidated details of the ten factors ............................................................... 140 5.5 Chapter summary ........................................................................................................ 140 CHAPTER SIX: ANALYSES AND RESULTS OF THE MAIN STUDY (STUDY TWO).................................................................................................................. 142 6.1 Chapter overview ....................................................................................................... 142 6.2 Data screening ............................................................................................................ 143 6.2.1 Screening and cleaning missing data ................................................................. 143 6.2.2 Assessing reliability and normality .................................................................... 144 6.2.3 Outliers ............................................................................................................... 145 6.2.4 Assessing normality ........................................................................................... 146 6.3 Profile of respondents ............................................................................................... 146 6.4 Confirmatory factor analysis and measurement models ............................................ 153 6.4.1 Confirmatory factor analysis .............................................................................. 153 6.4.2 Maximum Likelihood ........................................................................................ 153 6.4.3 Measurement model fit indices .......................................................................... 154 6.4.4 Results of CFA ................................................................................................... 157 6.4.5 Summary of the results of CFA ......................................................................... 172 6.5 Full structural equation model ................................................................................... 172 6.5.1 Convergent validity ............................................................................................ 173 6.5.2 Discriminant validity.......................................................................................... 173 6.5.3 Establishing a full structural model.................................................................... 177 6.5.4 The output of the final model ............................................................................. 178 6.5.5 Assessing the reliability of the final model ........................................................ 182 6.5.6 Hypotheses tests ................................................................................................. 182 6.5.7 Discussion relating to the hypotheses tests of SEM........................................... 185 6.6 Analysis of additional data ......................................................................................... 189 6.6.1 Analysis of demographic control variables as shown in the conceptual framework .. ……………………………………………………………………………189 6.6.2 Important attributes relating to purchase of organic food .................................. 193 6.6.3 Usage pattern of organic food ............................................................................ 195 6.6.4 Important attributes versus usage pattern ........................................................... 196 6.6.5 Comparative analysis of responses from the four selected cities ....................... 199 6.7 Chapter summary ....................................................................................................... 206 ix CHAPTER SEVEN: DISCUSSION OF FINDINGS, RECOMMENDATIONS AND CONCLUSION .................................................................................................................. 207 7.1 Introduction ................................................................................................................ 207 7.2 Implications of the best-fit SEM model ..................................................................... 208 7.3 Discussion of findings relating to research question one ........................................... 212 7.3.1 Product ............................................................................................................... 212 7.3.2 Regulatory .......................................................................................................... 216 7.3.3 Lifestyle.............................................................................................................. 218 7.3.4 Ethnocentrism .................................................................................................... 221 7.3.5 Pre-purchase evaluation and behavioural/purchase intention ............................ 224 7.4 Discussion related to research question two .............................................................. 225 7.4.1 Gender ................................................................................................................ 225 7.4.2 Age ..................................................................................................................... 226 7.4.3 Education and Income levels ............................................................................. 227 7.4.4 Usage patterns .................................................................................................... 228 7.5 Contribution of this study .......................................................................................... 230 7.5.1 Theoretical contributions ................................................................................... 231 7.5.2 Business implications ......................................................................................... 233 7.6 Limitations of this study ............................................................................................ 241 7.7 Future research directions .......................................................................................... 242 7.8 Concluding remarks ................................................................................................... 244 REFERENCES................................................................................................................... 246 APPENDICES ................................................................................................................... 283 Appendix 1: Questionnaires ............................................................................................... 283 Appendix 2: Summary of key authors from selected studies in different continents and countries ............................................................................................................................. 298 Appendix 3: Ethics approval .............................................................................................. 300 Appendix 4: Detailed analysis of the Pilot study(study 1) ................................................. 302 Appendix 5: Detailed analysis of the main study (study 2) ............................................... 317 Appendix 6: Discriminant validity ..................................................................................... 328 Appendix 7: The output of the final model ........................................................................ 329 Appendix 8: Hypotheses seven .......................................................................................... 334 Appendix 9: Referred Journal, Book and Conference Publications .................................. 349 x LIST OF FIGURES Figure 1.1 Labels of food categories in China Figure 1.2 Research design and sequence Figure 1.3 Roadmap to the structure of the thesis Figure 2.1 Structure of Chapter Two Figure 2.2 The Consumer Decision Process Model Figure 2.3 Conceptual framework for analysing consumer decisionmaking towards the purchase of meat Figure 2.4 Framework of factors which affect organic consumers’ attitudes and purchase decisions Figure 2.5 The Theory of Planned Behaviour Figure 3.1 Structure of Chapter Three Figure 3.2 Proposed conceptual framework for the purchase intention of organic foods in urban China Figure 4.1 Structure of Chapter Four Figure 4.2 Map showing the geographic dispersion of the four selected cities Figure 5.1 Structure of Chapter Five Figure 6.1 Structure of Chapter Six Figure 6.2 Recognition of organic food logos Figure 6.3 First thing in mind when thinking about organic food Figure 6.4 Types of organic food recently purchased Figure 6.5 Usage of the most recently purchased organic food Figure 6.6 Willing to pay extra for organic food Figure 6.7 Country of origin of imported organic food Figure 6.8 Choice of distribution channel Figure 6.9 Frequency of purchase of organic food products Figure 6.10 Monthly family income levels Figure 6.11a A two factor CFA model of the product-related construct Figure 6.11b A re-specified one factor CFA model for the productrelated construct Figure 6.12a A one factor CFA model of the regulatory construct Figure 6.12b A re-specified one factor CFA model for the regulatory construct Figure 6.13a Initial three factor model for the lifestyle construct Figure 6.13b A re-specified three factor model for the lifestyle construct Figure 6.13c Higher-order CFA for the lifestyle construct Figure 6.14 CFA for ethnocentrism measures Figure 6.15a One factor CFA model for the beliefs/attitudes construct xi 13 20 22 26 51 57 61 64 75 79 103 126 129 143 146 147 148 149 149 150 151 151 152 157 158 160 161 162 163 164 166 167 Figure 6.15b A re-specified CFA for the beliefs/attitudes construct Figure 6.16a One factor CFA model for the pre-purchase evaluation construct Figure 6.16b A re-specified CFA for the pre-purchase evaluation construct Figure 6.17 CFA for behavioural/purchase intention construct Figure 6.18 Endogenous constructs of the influencing stage of the conceptual model Figure 6.19 Exogenous construct of the cognitive/affective, evaluation of alternatives and behavioural/purchase intention stages of the conceptual model Figure 6.20 Final best-fit model of consumer purchase intention of organic food in urban China Figure 6.21 Simplified final best-fit model Figure 7.1 Structure of Chapter Seven Figure 7.2 Final best-fit model for the purchase intention of organic foods in Urban China xii 168 169 170 171 174 176 181 183 208 232 LIST OF TABLES Table 2.1 Summary of key attributes measured with associated theory in studies relating to organic products Table 3.1 List of all research hypotheses Table 4.1 Components of product-related construct and associated measures Table 4.2 Components of regulatory construct and associated measures 48-49 Table 4.3 Components of lifestyle construct and associated measures Table 4.4 Components of ethnocentrism construct and associated measures Table 4.5 Attributes of beliefs/attitudes construct and associated measures Table 4.6 Attributes of pre-purchase evaluation and behavioural/ purchase intention constructs and associated measures Table 4.7 Important attributes when purchasing organic food Table 5.1 Pattern and structure matrix for EFA of the product-related construct and associated measures Table 5.2 Factor loadings for EFA of the regulatory construct and associated measures Table 5.3 Pattern and structure matrix for EFA of the lifestyle construct and associated measures Table 5.4 Result for EFA of the ethnocentrism construct and associated measures Table 5.5 Result for EFA of the beliefs and attitudes construct and associated measures Table 5.6 Result for EFA of the pre-purchase evaluation construct and associated measures Table 5.7 Result for EFA of the behavioural/purchase intentions construct and associated measures Table 5.8 Summary of the ten factors Table 6.1 Progressive reliability scores for the ten factors Table 6.2 Model fit indicators adopted in this study Table 6.3 Regression weights for the product-related construct Table 6.4 Regression weights for regulatory measures Table 6.5 Regression weights for lifestyle measures Table 6.6 Regression weights for ethnocentrism measures Table 6.7 Regression weights for beliefs/attitudes measures Table 6.8 Regression weights for pre-purchase evaluation measures Table 6.9: Regression weights for behavioural/purchase intention Table 6.10 Summary of reliability of the measurement models 111 112 xiii 101 109 110 113 114 115-116 134 135 136 137 138 139 140 140 145 156 159 161 165 166 168 170 171 172 Table 6.11 Factor pattern and structure coefficients for the endogenous constructs of the influencing stage Table 6.12 Factor pattern and structure coefficients for the exogenous constructs of beliefs/attitudes, evaluation of alternatives and behavioural/ purchase intention Table 6.13 Summary of techniques used to test hypotheses 175 Table 6.14 Summary of reliability of the final model scales Table 6.15 Summary of hypotheses tests (H1a- H6) Table 6.16 Independent samples t test Table 6.17 ANOVA (purchase intentions, age groups) Table 6.18 ANOVA (purchase intentions, education levels) Table 6.19 ANOVA (purchase intentions, income levels) Table 6.20 Top five important attributes Table 6.21 Five least important attributes Table 6.22 Means of important attributes as applicable to the usage pattern Table 6.23 Summary of hypotheses testing Table 6.24 Goodness-of-fit indices showing comparative analysis of data from the four cities Table 6.25 Goodness-of-fit indices for data from all four cities Table 6.26 Standardised regression weights (four cities) Table 6.27 Evaluated hypotheses test for Shanghai Table 7.1 Action plans for the decision makers 182 184 190 191 191 193 194 194 196 xiv 177 178 198-199 201 202 203 205 210-211 LIST OF ABBREVIATIONS ACO ANOVA CDP CFA CGFDC CPC DV EFA GM GDP IFOAM IV KMO LOHAS OFDC SEM SEZ PRC TPB TRA Australian Certificated Organic (logo) analysis of variance Consumer Decision Process confirmatory factor analysis China Green Food Development Center Communist Party of China dependent variable exploratory factor analysis genetically modified gross domestic product International Federation of Organic Agriculture Movements independent variable Kaiser-Meyer-Olkin Lifestyle of Health and Sustainability Organic Food Development Center structural equation modelling Special Economic Zone People’s Republic of China Theory of Planned Behaviour Theory of Reasoned Action xv CHAPTER ONE: INTRODUCTION AND BACKGROUND 1.1 Introduction Over the last decade, there has been a significant increase in interest in organic foods. New food technology, growing health awareness, busier lifestyle and global environmental issues have triggered an awareness of organic foods in many developed countries. Consumers have started to question what they eat and the food production processes, as well as question whether food is safe (IFOAM 2012). Organic food has developed as one of the fastest growing areas of the world food market. From being a niche market, organics have evolved as mainstream products. Organic food studies have drawn the attention of scholars and government policy makers, and many studies have been undertaken in developed countries. In some developing markets, such as China, as a result of its dramatic and recent economic development, a series of food scandals have damaged the food industry and confidence in some sectors of the industry (Euromonitor International 2010). This in turn has raised awareness and sensitivity towards organic food products. China’s economic development has created a large middle class attuned to western trends and lifestyle adaptation. Organics have emerged as an important sector of food and, while predominantly understood and consumed in western countries, it is relatively new in countries such as China (Yin et al. 2010). It is evident from this research that few studies related to Chinese consumers buying organic food have been carried out. The primary objective of this study is to investigate the determinants of consumer purchase behaviour relating to organic food in urban cities of mainland China. The purpose of this first chapter is to introduce the topic and to provide an overview of this thesis as well as the rationale for undertaking this study. It provides the background to the research and explains the nature and use of the term ‘organic food’. As will become clearer, the term is not understood or agreed upon by most people in China. The research problem and associated research questions have been elicited from gaps in the extant literature. A conceptual model is developed and hypotheses related to its various constructs are 1 formulated, then the justification for the research is presented. The later sections of this chapter briefly describe the research methodology employed and the chapter concludes with an outline of the structure of the thesis. Research into organic food products has failed to provide convincing scientific evidence demonstrating their health merits (Brennan, Gallagher & McEachern 2003; Magkos, Arvaniti & Zampelas 2006). Conventional food outlets such as supermarket chains are adamant that the differences between conventional and organic foods are either nonexistent or so small as to be insignificant, which has created confusion among consumers when it comes to understanding the health benefits of organic food. This research does not aim at making any judgement on this debate; however, it investigates related perceptions of existing and potential Chinese consumers of organic foods. Previous studies have shown that consumers’ concerns about health and the environment are the most important factors that influence their purchase of organic food (Chen 2007; Chryssochoidis & Krystallis 2005; Millock, Wier & Andersen 2004; Padel & Foster 2005; Radman 2005; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006). Organic foods are perceived as being more nutritious and healthier, safer, environmentally friendly, containing few chemical residues and tasting better than conventional food (Harper & Makatouni 2002; Krystallis & Chryssohoidis 2005; Sanjuán et al. 2003). As a result, consumers are willing to pay a premium price for them (Aguirre 2007; Botonaki et al. 2006; Krystallis & Chryssohoidis 2005; Ureña, Bernabéu & Olmeda 2008). Even in newly emerging markets, consumers are willing to pay more for higher quality and for food certified as being ‘safe’ (Grannis, Hine & Thilmany 2001). The range of organic food products not only includes fruits, vegetables and meat, but also extends to grains, nuts, dairy produce and many other processed products (Aarset et al. 2004; International Trade Centre 2011). Reasons for the non-purchase of organic food include consumers’ perceptions of it being expensive, its lack of perceived value, limited availability, lack of choice, low profile distribution channels and even unsatisfactory appearance (Bhaskaran et al. 2006; Chang, Zepeda & Griffith 2005; Essoussi & Zahaf 2008; Radman 2005; Stobbelaar et al. 2007). Misleading labelling and certification have also been identified as 2 barriers to purchase (Aarset et al. 2004; Brennan, Gallagher & McEachern 2003; Essoussi & Zahaf 2008; Krystallis, Fotopoulos & Zotos 2006). China, the geographical focus of this study, has experienced dramatic economic growth in the last 20 years. Its organic food sector has grown faster than the worldwide average (IFOAM 2006) and the market has been developing at an annual double digit rate over the past decade (US Department of Agriculture 2008). Official figures for the exact sales of organic food products are unavailable, although they had an estimated market value of US$1.1 billion in 2008 (International Trade Centre 2011). According to one source, China was the fourth largest organic market in the world (Zhou 2008b). It is developing an affluent urban middle class and also an expatriate community who are interested in purchasing organic food (Euromonitor International 2008b). Large international retailers such as Wal-Mart (US) and Carrefour (France) have significantly increased their sales of organic food products in China, even though these are relatively more expensive for consumers (Baer 2007; Sanders 2006). 1.2 Background of Chinese economic development Before discussing the nature of the organic industry in China, it is desirable to provide a background of the evolution of modern China and especially the economic reforms which have defined it in its modern context. China’s history is a long and complex one. Modern China – the People’s Republic of China (PRC) – was founded by the Communist Party of China (CPC) in 1949. From 1949 to 1979, the PRC’s development was slow and quite unbalanced. The programs under the banner of the Great Leap Forward of the late 1950s failed miserably as an economic and social campaign by the CPC. The economic decisions between 1958 to 1961, to rapidly transform the country from an agrarian economy into a modern communist society through agriculturisation, industrialisation and collectivisation, were patently unsuccessful. Private farming was prohibited, and those engaged in it were labelled as counter-revolutionaries and persecuted. The Great Leap Forward ended in catastrophe, resulting in millions of deaths, and was deemed a very expensive disaster in modern Chinese history (Chesneaux 1979; Tian & Dong 2011). 3 Further complicating Chinese economic development was the Cultural Revolution (19661976). According to the then leadership of the CPC, this was a political movement designed to further advance ‘socialism in the country’ by removing capitalist elements from society and imposing Maoist orthodoxy within the party. The movement brought chaos and largescale damage to China economically and socially. Millions of people were persecuted, suffering a wide range of abuses in the violent factional struggles that ensued across the country. A large segment of the population was forcibly displaced, most notably the transfer of urban youth to rural regions during the Down to the Countryside Movement (Chesneaux 1979; Slavicek 2010; Tian & Dong 2011). Food was scarce and many people were at starvation point. In 1978 the new CPC leader Deng Xiao Ping introduced and shaped the economic reforms soon after Mao’s death.1 These reforms have been a transitionary path to a market based economy, under Deng’s famous slogan ‘As long as we increase production, we can even revert to individual enterprise; it hardly matters whether a good cat is black or white, as long as it catches mice, it is a good cat’ (Chesneaux 1979, p. 115). China started to adapt an open door policy, coinciding with a shift in thinking from a centralised planned economy to a so-called ‘socialist market economy with Chinese characteristics’. Over time, China started to reintegrate within the global economy (Fairbank & Goldman 1998). One of the first and most conspicuous signs of change in the early 1980s was the welcomed array of new food and everyday items of consumption. The expansion of the free market, relaxed price controls and proliferation of retail outlets boosted the availability and range of highquality foods (Croll 2006). By the 1980s, economic reform had fundamentally changed Chinese consumers’ experience, shopping and the dynamics of food retailing. Food coupons and ration books to purchase food and many other consumer goods eventually were phased out from people’s daily lives (Veeck 2000). In 1997, the party also announced the phasing out of state-owned industries and, except for sensitive and key industries, most state enterprises were to be sold off (Fairbank & Goldman 1998). Consumer products, including fast food outlets, gradually opened up and the Chinese retail market began 1 Mao Zedong (Mao Tse-tung), commonly referred to as Chairman Mao (26 Dec. 1893 – 9 Sept. 1976), was the founder of the PRC in 1949 and ruled until his death in 1976. 4 opening its doors to products manufactured by foreign companies. Consumers were becoming more aware of foreign products (Kahal 2001), and now had a wider selection of foreign and domestic products. China’s inclusion in the World Trade Organization (WTO) in 2001 allowed it to take the next step to invigorate its economic reforms. It was a major milestone in its economic transition, and led the country to embrace global standards of manufacturing in many areas (Croll 2006; Keane 2007). There has been a sense of expectation that China would play by the market rules and its emergence would create unprecedented opportunities for the middle classes, a social demographic segment comprising between 48 million and 90-100 million people (Keane 2007). By staging the 2008 Beijing Olympic Games, China displayed its economic and political prowess to the world (Xu 2008). In the last decade China has experienced dramatic economic growth with an average GDP growth of about 10%, and is now the single largest contributor to global economic growth and the second largest economy after the US (DFAT 2012). In 2010, the value of foreign imports reached US$1,398 billion, up by 38.7% on the previous year (National Bureau of Statistics of China 2011). By 2015, China is envisaged to become the third biggest consumer market after the US and Japan (Atsmon et al. 2009). The chances of it becoming the biggest economy in the world, once thought impossible, are within reach. Most importantly, China has the potential to offer increased purchasing power to the largest population in the world (Wu 1999). As this study involves investigation of organic food consumers’ perceptions, both within China and the rest of the world, it is fitting to introduce the related concept and industry of organic foods. As organic food products result from the use of organic agriculture (Guido et al. 2010), a brief history of organic agriculture will bring rich context to this study. 1.3 Evolution of the global organic food scenario 1.3.1 A brief history of the organic concept As noted in previous studies, ‘The origins of modern organic agriculture are intertwined with the birth of today’s industrially based agriculture’ (Kristiansen 2006, p. 4). Organic agriculture was practised initially in response to the industrialisation of agriculture 5 (Torjusen et al. 2001). It was developed as a form of farming compatible with natural systems and codified in the work of Rudolf Steiner and Sir Albert Howard in the 1920s (Courville 2006), evolving as an alternative to the major initiatives taking place in agricultural development. In 1943 Howard published a forward thinking endorsement of organic agriculture, An Agricultural Testament (Heckman 2005; Howard 1943), advocating that the UK preserve the cycle of life and adopt permanent agriculture systems using urban food waste and sewage to enhance fertile soil. The term ‘organic farming’ was first used by Lord Northbourne in his 1940 book, Look to the Land. It was later championed by US businessman and publisher Jerome Rodale who was the first to use the term ‘organic’ in food production in his publication Organic Gardening and Farming. This sought to promote a healthy and active lifestyle that emphasised organically grown foods (Heckman 2005; US Department of Agriculture 2012c). The first organic farming society was established in Australia in October 1944: the Australian Organic Farming and Gardening Society (Paull 2008b). In 1946, Lady Eve Balfour was inspired by Howard to set up the Soil Association of the UK (Institute of Food Science and Technology 2009). The publication of Silent Spring (Carson 1962) in the US saw a time of significant change and upheaval, which witnessed the start of both modern organic and environmental movements. The Soil Association established the UK’s first set of Organic Food Standards in 1974 (Institute of Food Science and Technology 2009). Prior to industrialisation, fertilisers and chemicals were not used and all food was grown ‘organically’. However, today there is a high degree of difficulty in determining whether a farm’s production is or is not organic without organic certification (Halpin 2004). Scientists became increasingly aware of organic agriculture in the 1980s, including those who did not support alternative agricultural systems (Kristiansen 2006). Organic products are now a rapidly growing sector in most developed agricultural economies, becoming one of the most convincing answers to the increasing environmental and food quality issues (Sottoni et al. 2002). In the 1960s, organic production was very much outside the mainstream of the agrifood business, and few took it seriously. This brought various organic stakeholders together to form the International Federation of Organic Agriculture Movements (IFOAM) in 1972 (Courville 2006). This stands out today as the only global organic non-government 6 organisation (IFOAM 2012; Kristiansen 2006), with more than 750 member organisations in 116 countries. The IFOAM organic guarantee system accredits national organic food certification bodies to endorse genuine organic food products with the IFOAM seal (IFOAM 2012, 2011a). However, as organic food labels are designed by certification bodies within each country, there are no standard global labels and this obviously creates ambiguity and problems with international scrutiny and consistency (Aarset et al. 2004). For example, since 1993 the European Union (EU) has regulated standards for organic food products. In the UK, all organic food producers are required to be registered on the UK Register of Organic Food Standards (UKROF) or one of its approved certification bodies (Byng 1993). As organic food production has increased in volume and value, governments have increasingly taken an interest in regulating the industry. Tensions continually arise between private accreditation bodies accredited by IFOAM and public bodies controlled by regional and national governments, creating significant threats to the development of organic agriculture and trade (Courville 2006). 1.3.2 A contemporary overview of the global organic food market Organic food and beverages, from being an insignificant niche market, have become part of the mainstream since the mid-1990s (Lockie, Halpin & Pearson 2006), with global sales reaching $US27.8 billion in 2004. An estimated annual growth rate of organic product sales reached 16% to 20% in the European market between 2002 and 2005 (Willer & Yussefi 2006). Rising disposable income, growth of health awareness and busier lifestyles have driven a burgeoning world demand for organic and dietetic food and beverages (2005). Although the organic food market is comparatively small, as little as 1% of farming in most OECD countries, it is regarded as one of the biggest growth markets in the food industry (Hughner et al. 2007), as well as in some developing countries (Kortbech-Olesen 1998). Consumers especially in industrialised countries have an increased awareness in organic food (Wier & Calverley 2002), with consumption increasing significantly across the world (Chen 2007; Makatouni 2002; Sottoni et al. 2002; Wier & Calverley 2002). According to the latest report from the Research Institute of Organic Agriculture in Switzerland, Germany and Austria, organic agriculture is practised in 160 countries and 37.2 million 7 hectares of agricultural land are managed organically by 1.8 million farmers. These figures show that global organic products continue to grow even in the face of the global economic crisis. The total sales of organic products reached nearly $US55 billion in 2010 (IFOAM 2011b). The countries with the greatest levels of allocated organic farmland areas are Australia 11.8 million hectares, Argentina 3.1 million hectares, China 2.3 million hectares and the US 1.6 million hectares (Willer & Yussefi 2006). 1.3.3 The nature and meaning of the term ‘organic food’ The term ‘organic’ is derived from the Greek word bios, meaning life or way of living (Essoussi & Zahaf 2008). It has many different meanings, interpretations and definitions (Aarset et al. 2004; Brennan, Gallagher & McEachern 2003; Chryssochoidis 2000). In 1995 the US National Organic Standards Board adopted the following definition: ‘Organic’ is a labeling term that denotes products produced under the authority of the Organic Foods Production Act. The principal guidelines for organic production are to use materials and practices that enhance the ecological balance of natural systems and that integrate the parts of the farming system into an ecological whole (US Department of Agriculture 2012b). Australia’s National Standard for Organic and Bio-Dynamic Produce defines ‘organic’ as: the application of practices that emphasise the: use of renewable resources; and conservation of energy, soil and water; and recognition of livestock welfare needs; and environmental maintenance and enhancement, while producing optimum quantities of produce without the use of artificial fertiliser or synthetic chemicals (Organic Federation of Australia 2005, p. 8). The Food and Agriculture Organization of the United Nations (FAO) (2002, p. 1) defines organic products as those which are ‘certified as having been produced through clearly defined organic production methods’. In support of this definition, ‘organic foods are minimally processed to maintain the integrity of the food without artificial ingredients, preservatives or irradiation’ (Essoussi & Zahaf 2008, p. 96). 8 Equally as significant is the link between organic food production and organic farming. EU regulations on organic farming entail ‘significant restrictions on the use of fertilisers and pesticides which may have detrimental effects on the environment or result in the presence of residues in agricultural produce’ (Roddy, Cowan & Hutchinson 1994, p. 2). Organic agriculture has been developed as an alternative form of farming compatible with natural systems, and the concept of organic farming is based on a holistic viewpoint (KortbechOlesen 1998). Organic agriculture is different from other techniques of farming, such as conventional agriculture which is an industrialised system using chemical synthetic substances for fertilisers and insecticide to maximise the volume of production and profit (Guido et al. 2010). Vindigni, Janssen and Jager (2002) argue that ‘organic’ often refers to a claim associated with the process, not a claim associated with the product. According to IFOAM (2009b): Organic agriculture is a production system that sustains the health of soils, ecosystems and people. It relies on ecological processes, biodiversity and cycles adapted to local conditions, rather than the use of inputs with adverse effects. Organic agriculture combines tradition, innovation and science to benefit the shared environment and promote fair relationships and a good quality of life for all involved. Global consumers require a clearer understanding of what organic food actually means, as confusion prevails amongst regulatory bodies. Fotopoulos and Krystallis (2002b) report that consumers are confused by the use of different terminologies. Even if they have heard of the term ‘organic’, nearly half could not give an accurate definition of it (Krystallis, Petrovici & Arvanitoyannis 2006). Various contrasting ‘organic’ definitions are often coupled and confused with terms such as ‘biological’, ‘ecological’, ‘natural’, ‘alternative’, ‘unsprayed’, ‘free of pesticides’, ‘without artificial additives’, ‘without chemicals’, ‘environmentally produced’, ‘green’ and ‘sustainable’ (Aarset et al. 2004; Essoussi & Zahaf 2008; Hutchins & Greenhalgh 1995; McEachern & McClean 2002; Schifferstein & Oude Ophuis 1998). Others argue that organic food should never be defined as pesticidefree (Institute of Food Science and Technology 2009). It can be seen that the dilemma 9 associated with the term ‘organic’ has caused worldwide confusion in the minds of consumers and researchers alike. For the purposes of this study, the definition adopted is taken from the Australian organic market report 2010: ‘Organic is a labelling term that denotes products that have been produced in accordance with organic production standards and certified by a duly constituted certification body or authority’ (Kristiansen et al. 2010, p. 8). 1.4 Changes in the Chinese economy 1.4.1 China’s new urban consumers For more than two decades, economic reforms have lifted hundreds of millions of people out of poverty, giving many the opportunities to benefit from a rise in income levels and living standards. The growth and reforms have brought on fundamental structural shifts in China’s economy. The country has become a manufacturing powerhouse, combining technologically sophisticated factories with energetic, intelligent and low-cost labour (Austrade 2011; McGregor 2005). According to the Chinese Academy of Social Sciences 2011 report, urbanisation is at the centre of Chinese economic reform and development and will be one of the most important driving forces for future economic growth. China will also accelerate the pace of its urbanisation in the next 20 years, from the current 47% to more than 70% by 2030 (Xing 2011). The global consulting firm McKinsey predicts that urban households in China will make up one of the largest consumer markets in the world by 2025 (Farrell, Gersch & Stephenson 2006). The number of wealthy households continually grows, and in 2010 China held the world’s fourth-largest concentration of wealthy people. Chinese consumers account for more than 20% of the global luxury market (Atsmon & Dixit 2009). As former Australian Prime Minister Kevin Rudd commented, ‘within 30 years, China has transformed itself from an impoverished, isolated and mostly agrarian economy to the increasingly wealthy, internationalised and urban economy we see today’ (Callick 2011, p. 1). The steady growth in income levels and shrinking family size (Farndon 2008) has led to newer lifestyles emerging in the reformed China (Wei 1997). Consumers’ product choices 10 in its large cities match those in most developed economies (Piron 2006). A recent trend of lifestyle emanating from the US is called LOHAS (Lifestyle of Health and Sustainability). There has been an increase in the number of people seeking to emulate this lifestyle trend in Shanghai. In China’s large cities, consuming organic food, growing fruits and vegetables, and touring the local farms has become popular. In Chinese this trend is called ‘Nong Jia Le’ which means ‘vacation in the countryside’. Since June 2008, the government has banned free plastic bags in supermarkets, and the media has raised awareness about environmental and sustainability issues, forcing more consumers to bring their own shopping bags to supermarkets (US Department of Agriculture 2008). The rising sense of ‘enjoying life’ and ‘quality of life’ is beginning to drive up spending, not only among consumers in key cities but also in the massive second and third tier cities. China also demonstrates big differences between its cities and regions in terms of wealth and size. Although the richer cities such as Beijing, Shanghai and Guangzhou are the logical first point of entry for companies targeting high-end consumers, there are several second tier cities such as Dalian and Qiangdao in the north, Ningbo, Hangzhou and Nanjing in the Yangtze River Delta and Xiamen in the south (Smith 2007). Certainly in big cities like Beijing, Shanghai, Guangzhou and Shenzhen, the large middle class considerably influences the rest of the population with their lifestyle adaptations (Zhou 2008a). With the increase in the pace of home and working life, health issues are becoming important, especially among office workers who lack fresh air and facilities for exercise. Healthy products and supplements are popular among these people, being middle-aged and young adults with high incomes and a taste for the good life. They pay more attention to the enjoyment of cultural products and a sense of wellbeing. These groups make up the elites in society, and are part of a big consumer group (Euromonitor International 2008b). In some Chinese cities, the general population’s obesity level is approximately 20%, similar to that in the US and Europe (Kluger 2010). The recent increases in income levels have raised issues regarding food safety and food nutrition in China (Peng 2006). Concerns about the safety of food following recent food safety scandals as well as the rising purchasing power of consumers have driven growth in 11 sales of organic food (Euromonitor International 2010). It has been reported that Chinese consumers are losing confidence in locally produced milk after a series of scandals involving food safety (Euromonitor International 2011). The 2008 incident of baby milk formula contamination involving melamine-tainted dairy products has increased concerns about domestically produced food products. This scandal prompted well-off consumers to turn to imported milk as the alternative (US Department of Agriculture 2008). The melamine crisis was a wake-up call to the entire Chinese society, causing issues relating to food safety to become top priority on the agenda of the state councils. 1.4.2 The organic industry in China Chinese organic agriculture originally came into being over a thousand years ago, based on traditional sustainable farming practices (Gao et al. 2009). The record of sustainable farming dates back 4,000 years, involving organic practices such as crop rotation, composting and diversified production. The ‘green revolution’ in China occurred in the 1980s, much later than the rest of world, due to the economic reform being introduced at that time (International Trade Centre 2011). 1.4.2.1 Debunking the organic food myth in China The term ‘organic food’ has many different connotations and interpretations, and in China is sometimes used interchangeably with the term ‘green food’. Organic products are often referred to as being ‘eco-products’, suitable for ‘green’ consumers who are ecologically or environmentally aware and concerned with health and quality of life issues (Fotopoulos & Krystallis 2002b). Generally speaking, in many western countries, ‘green’ connotes a sense of environment associated with fears of global warming, and saving the planet by reducing carbon emissions (Bonini & Oppenheim 2008; Kalafatis et al. 1999; Okun 2007). Concerns about climate change are another reason to go ‘green’ (Bonini & Oppenheim 2008): ‘The green consumer is generally defined as one who adopts environmentally-friendly behaviour and/or who purchases green products over the standard alternatives’ (Shamdasani, ChonLin & Richmond 1993, p. 488). Other stereotypical images of organic consumers are ‘greenies’, ‘health nuts’ or ‘yuppies’, more interested in fashion than anything else, hence 12 consumption of organic food reflects a ‘greening lifestyle’ (Lockie et al. 2002). According to Tanner and Kast (2003, p. 885): Green food products are domestically cultivated rather than imported from foreign countries; they are organically rather than conventionally grown; they are seasonal and fresh rather than frozen; they are not wrapped and support fair trade. Many developed countries have set stringent standards to ensure quality of their organic food products. A number of developing countries including China have started to establish national organic product standards and regulations (Food and Agriculture Organization of the United Nations 2002). In China three categories of food are deemed to be safe, ecological and environmentally friendly: ‘non-polluted food’ (or hazard-free food), ‘green food’ and ‘organic food’. These three categories can be explained by means of a pyramid where the lowest level is nonpolluted food, and the highest is organic food (Gao et al. 2009; Luo 2007; Paull 2008a; Zhang 2007). The term ‘organic food’ has been influenced by the demand of the international market, while ‘green food’ and ‘non-polluted food’ labels have been introduced for domestic consumption (Gao et al. 2009). Figure 1.1 illustrates the labels for the three categories. Figure 1.1 Labels of food categories in China Non-Polluted Food This logo has only Chinese script. Green Food This logo has both Chinese and English scripts. Most Chinese consumers are aware of this logo. Organic Food This logo has both Chinese and English scripts. Only a few Chinese consumers are aware of this logo. 13 Source: Gao et al. 2009; Luo 2007; Paull 2008a; Yin et al. 2010; Zhang 2007 Although only a very small and almost insignificant percentage of the population consume organic food products, China has gone well beyond being an adopter of the global organic concepts, and is now an active organic food innovator by promoting the concept of ‘green food’ (Gao et al. 2009; Paull 2008a). Most Chinese consumers are familiar with the concepts of ‘green’ and ‘non-polluted’ food, but many are not as familiar with the term ‘organic food’ (Paull 2008a; Zhang 2007). A survey by Li, Cheng and Ren (2005) suggests that 92.2% of respondents were aware of ‘green food’, 69.6% knew about ‘non-polluted food’, but only 37.2% were aware of ‘organic food’. The ‘green food’ concept appears to be well understood and recognised in China (Paull 2008a). However, the concept’s interpretation in China is different from that understood by consumers in western countries, and will be further elaborated in later chapters. 1.4.2.2 Organic food standards in China With increasing affluence derived from the rapid economic growth, certain sections of Chinese society and government have begun to realise the hazardous impact of environmental deterioration on national wellbeing (Chan 2001; Peng 2006). The Chinese market for green and organic food is still in its early stage. Standards are low, inspections are weak, and regulations are not adequately policed. In recent years, the government has increased warnings against sub-standard food products, and consumers have become cautious about food safety issues (Ho, Vermeer & Zhao 2006). The General Standard for the Labelling of Food (GB7718-94) (Chinese new food labelling law) has been in effect since February 1995 and closely follows standards recommended by the Food and Agriculture Organization of the United Nations/World Health Organization’s Food Law Committee. It requires that imported food labels are primarily in English or other foreign languages, and the required content in simplified Chinese characters (Zhao et al. 2000). The Chinese national organic logo has been mandatory for all Chinese organic products since 2005, and this has made understanding organic labels more straightforward for consumers (Euromonitor International 2010). 14 The China Green Food Development Center (CGFDC) was founded in 1992 to meet domestic consumers’ demand and the level of development of China’s agricultural production and international competition. It is a specialised agency for the development and management of green food under the supervision of the Ministry of Agriculture. This agency advocates: Green food products include edible produce and processed products produced in a sustainable environment using quality standards governing quality control, pollution, safety, quality and certification (CGFDC 2009). The CGFDC has instituted two standards, ‘A’ and ‘AA’. ‘A’ represents a transitional level between conventional and organic food, allowing restricted use of chemical fertilisers and pesticides, while the ‘AA’ symbolises full organic status and matches all the international organic food standards. A total of 98% of green food produced in China belongs to standard ‘A’ (Baer 2007; Sanders 2006). However, the ‘AA’ standard is not accepted by international markets as organic food (Zhou et al. 2004). The Organic Food Development Center (OFDC), founded in 1994, is the oldest and largest research, inspection and certification organisation. It is the only organic food certification body in China which has affiliations to IFOAM. Its mission is to ensure human health and protect the ecological environment through promoting sustainable agriculture (Organic Food Development Center 2009). The Ministry of Agriculture oversees certification schemes in relation to non-polluted and green food. Organic food certification is jointly supervised by the ministry and the State Environment Protection Agency (Paull 2008a; Zhang 2007). In July 2002, premier Wen Jia Bao emphasised that the government must regulate the certification and accreditation of non-polluted, green and organic food, and accelerate support for modern production techniques (Gao et al. 2009). In recent years, the government has established a legislative framework in line with international organic institutions, with administrative measures for organic product certification and rules for implementing the certification of organic food. The Certification 15 and Accreditation Administration of the PRC is a peak body set up by the State Council to perform the functions of administrative operation, unified management, supervision and comprehensive coordination of certifications and accreditations (Wang 2010). 1.5 Research problem and questions Research investigating consumers’ awareness and perceptions of organic food is well developed in North America and Western Europe (Bonti-Ankomah & Yiridoe 2006). On the other hand, studies of the Chinese market indicate that little is known or understood about consumers’ attitudes towards organic produce or about their related purchasing behaviour (Yin, Wu & Chen 2008; Yin et al. 2010), and the general awareness of organic food products is relatively low (Li, Cheng & Ren 2005; Yin, Wu & Chen 2008; Yin et al. 2010). As a heavily populated country with great potential in production and consumption of organic food, the development of China’s organic food market is likely to affect the global market of organic food (Yin et al. 2010). The overarching research problem and objective for this study is to ascertain the framework, conditioning and new understanding of consumers’ purchase behaviour in relation to organic food in China. To date, there is no theoretical framework or model that attempts to evaluate the overall organic food purchasing behaviour of mainland Chinese consumers. Organic food in China is a relatively new development both in its consumption and related marketing strategies. Recent food scandals have produced scepticism towards both the food industry and the government. It has not passed without notice that much Chinese consumer behaviour is influenced by western fashion and lifestyle trends. This has also influenced organic food consumption, being a sector which is influenced by health, food safety, lifestyle and government regulations. These, on examination, appear to be similar to the consumer motivation for purchase of organic foods in western societies. Factors affecting consumer choice of organic food in China have not been analysed on a national basis. To further investigate this topic, the following research questions are posed as result of gaps in the literature: 16 1. What are the factors that influence urban Chinese consumers’ beliefs and attitudes, pre-purchase evaluation, and behavioural/purchase intention towards the purchase of organic foods? 2. Evaluate the role of demographic variables and usage patterns in the behavioural /purchase intentions of urban Chinese consumers towards organic food. Hence the primary research aim is to investigate the various determinants and elements which influence consumers’ purchase behaviour of organic food products in urban China. The main research objective details are as follows: The first objective is to investigate the dimensions that influence the personal attitudes of urban Chinese consumers towards the purchase of organic food. The second objective is to examine the impact of urban Chinese consumers’ personal attitudes on their pre-purchase evaluation and purchase behavioural intentions associated with organic food. The third objective is to examine the influence of demographic segmentation variables on the behaviour/purchase intention of organic food. The fourth objective is to develop and empirically test a conceptual model which incorporates all the elements of the above three objectives. This research study also intends to develop a theoretical model which would explain the purchase intention and decision-making processes about organic food in this respect. The following hypotheses have been elicited from the conceptual framework: H1: Influencing dimensions have positive and direct relationships with consumers beliefs/ attitudes, pre-purchase evaluation and behavioural/purchase intention of organic food. H2: Consumers’ beliefs/attitudes have direct and positive relationships with pre-purchase evaluation. H3: Pre-purchase evaluation has a direct and positive relationship with behavioural/ purchase intention. H4: Demographic variables influence behavioural/purchase intention of organic food. A comprehensive country-wide survey was conducted in order to quantify and explicate the impact of various determinants influencing consumers’ purchase intention relating to organic food in urban China. While the aim of this study is to investigate the buyer 17 behaviour of urban Chinese consumers, its findings can be used by various stakeholders in China for the marketing and promotion of organic food products. 1.6 Academic and practical justification for this research 1.6.1 Understanding Chinese culture The author of this study is of Chinese origin and has conducted research in both English and Chinese. Coming originally from mainland China, possessing a bilingual capability and having both Australian and Chinese industry background as well as a deep understanding of the Chinese people and their culture is of value for this study. These advantages enable the researcher to scrutinise and analyse both Chinese and English secondary and primary data, and to address issues arising during data collection. This benefit of having access to both languages is acknowledged by many scholars. Sun (2003) emphasises that there are distinct advantages if the researcher is a native Chinese with a Chinese cultural background who is also fluent in both Chinese and English. Bednall and Kanuk (in Sun 2003) claim that there are nine basic differences in terms of crosscultural consumer behaviour research that could undermine the research qualities, if the researcher does not belong to the culture and cannot communicate appropriately in its language. 1.6.2 Theoretical academic contribution The unique theoretical model used in this study is a combination and integration of three theoretical models: the theory of Consumer Decision Process (CDP) (Blackwell, Miniard & Engel 2006), the Hierarchy of Effects model (Lavidge & Steiner 1961; Barry & Howard 1990) and the Theory of Planned Behaviour (TPB) (Ajzen 1991). These models have been widely utilised within food and organic food studies (Bredahl, Grunert & Frewer 1998; Drichoutis, Lazaridis & Nayga Jr 2007). Organic foods are perceived as being more nutritious and healthier, safer, environmentally friendly, containing fewer chemical residues and tasting better than normal food (Krystallis, 18 Fotopoulos & Zotos 2006). Consumers’ purchase of organic food is based on subjective experiences and perception (Hughner et al. 2007). Those with positive attitudes towards organic food are more likely to form positive intentions to purchase it (Chen 2007; Honkanen, Verplanken & Olsen 2006). Demographic and psychographic variables demonstrate significant influences on consumers’ decision-making in the purchase of organic foods (Chang, Zepeda & Griffith 2005; Lea & Worsley 2005). This research is designed to provide a better understanding of Chinese consumers’ purchase intention towards organic food. It extends and integrates theory and findings from several related areas, and empirically validates theory and practice. 1.6.3 Business implications and beneficiaries of this research This study is primarily designed for an Australian audience and stakeholders. Australia is an active participant in organic farming and has one of the largest outlays of land in the world for organic production. Its organic food industry has been strongly influenced by the rapid growth of overseas demand (Willer & Yussefi 2006). With China being a potential huge market, the study’s findings would certainly interest both public and private enterprises in Australia. Investigating Chinese consumers’ attitudes towards locally grown and imported organic food are essential for Australian and international players who are interested in entering the Chinese market. Given the background and research questions delineated in the previous sections, this study provides valuable insights into the global consumer behaviour related to organic foods. It also provides a better understanding of Chinese consumers’ attitudes towards the purchase of organic food products and builds a new body of knowledge necessary to market such products in China. The findings would be useful to public policy makers interested in identifying strategies aimed at increasing the demand for organic food. They will also have important implications for marketers of organic food in China. Beneficiaries of this research include stakeholders in China and globally, such as consumers, vendors both local and international and government agencies. 19 1.7 Research methodology In order to effectively address the research problem and questions stated in Section 1.5, it was decided to employ a quantitative research design. This was done in two stages (Study One and Study Two) as shown in Figure 1.2. Figure 1.2 Research design and sequence Source: structure adapted and modified from Mathews (2009) Data was collected in China in two stages. Firstly, online data was collected for the purpose of the pilot study (Study One) in December 2009. Subsequently, paper-based data (Study Two) was obtained from four major Chinese cities in May 2010. The survey instrument was developed using validated measures from previous literature. It was then pre-tested using two focus groups in China. The English version of the survey instrument was translated into Chinese by the author. All items of the survey were translated back into English to verify the accuracy of translation. The back-translation technique is the most widely employed method in cross-cultural studies (Usunier 2000). The translated versions were also cross-checked by three other bilingual researchers, thus ensuring content and face validity of the survey instrument. The online survey in Study One was made available via the university’s Opinio platform. This Opinio software enables data collection, production and reporting of a survey ensuring anonymity, confidentiality and privacy to the participant. The web link to the online survey 20 was sent to selected Chinese food outlets for onward submission to their customers. The sampling frame was drawn up using retail food outlets’ customer databases. Participants were randomly selected from the sampling frame. Care was taken to include a proportionate number of prospective customers. A series of exploratory factor analyses (EFA) were performed on the pilot study data to identify the factors and associated latent variables. After the online data was analysed, the survey instrument was slightly modified for use in the paper-based survey, which comprised Study Two. Previous studies in the organic food area indicated that data collected in food outlets of retail chains is an efficient method (Drichoutis, Lazaridis & Nayga Jr 2007). The paper-based survey questionnaires were administrated at major supermarkets in four selected first and second tier cities, namely, Beijing, Shanghai, Shenzhen and Chengdu. These cities are geographically dispersed being located in the north, south, east and west respectively of China, and are economically and politically prominent and the main engines of China’s phenomenal economic development. Hence these consumers are more affluent, and are more likely to be aware of organic foods. All the participating supermarkets had dedicated organic food sections. A series of confirmatory factor analyses (CFA) were then conducted on data from the paper-based surveys to validate the findings from the EFA of the pilot study. The structural model was tested and presented as the final stage, and then a number of hypotheses were tested. Further analysis using invariance testing was performed to investigate whether there were significant differences in the data obtained from the four cities. 1.8 Outline of the research This thesis comprises a total of seven chapters. Figure 1.3 depicts the roadmap to the structure of the thesis as suggested by Perry (1995). 21 Figure 1.3 Roadmap to the structure of the thesis Source: developed from Perry (1995) for this research Chapter One provides an overview of the thesis, introduces the research background, and explains the nature and meaning of the term ‘organic food’. It further identifies the research 22 problem and teases out the research questions in this study. Additionally, justification and a brief research methodology have been developed. Chapter Two provides a context of previous studies and literature. It examines previous studies relating to organic food products in the global sense, discusses and identifies the gaps in the literature, noting the very few studies undertaken within a Chinese context. The chapter also explores the extensive literature surrounding related theories. CDP, Hierarchy of Effects model and TPB are the most important theories used to investigate consumers’ buyer behaviour of organic food. Chapter Three develops a unique conceptual framework and theoretical model. Various constructs have been introduced which influence consumers’ purchase behaviour of organic food products. This chapter elicits a series of relevant hypotheses intended to address the research questions. Chapter Four develops an appropriate research design and describes details of the methodology utilised in the study. Two stages of data collection are employed, online and paper-based surveys. The survey instrument originally designed in English was translated into Chinese. Chapter Five presents the analysis and findings of Study One, the pilot study, which is based on 200 valid online responses. A series of EFAs were performed on the data of the pilot study to identify the factors and associated latent variables. Various reliability and validity tests were conducted. Chapter Six presents the analyses and results of the main study (Study Two), which is based on 1000 paper-based responses. The analysis of this data is completed using a series of CFAs to build measurement models, which in turn are used in a structural equation model, and finally hypotheses tests are conducted. A comparative study of the four selected cities which are geographically representative was conducted by the use of invariance tests. Chapter Seven discusses the findings in relation to the theoretical model. Results of individual hypotheses tests are discussed to provide meaningful insights and answers to the research questions. Both practical and theoretical findings and recommendations are suggested to various stakeholders. Finally, limitations are articulated and future research directions are proposed. 23 1.9 Chapter summary This chapter has provided a brief background of the global organic market. Organic food products are perceived as being more nutritious, healthier, safer and environmentally friendly. China has experienced dramatic economic growth in the last 20 years, and in more recent times the Chinese organic food sector has grown faster than the worldwide average. China has gone well beyond being an adopter of the world organic concepts, and is now an active organic innovator. Current research examining consumers’ awareness of organic food has been well developed in North America and Western Europe, but very few studies of this nature have been undertaken in China. Little is known or understood about Chinese consumers’ attitudes towards organic food products or their associated purchasing behaviour. Thus, this research seeks to explore the determinants of Chinese consumers’ buyer behaviour towards organic food. In order to evaluate previous research in this area, a thorough literature review is performed which is explained in the next chapter. 24 CHAPTER TWO: LITERATURE REVIEW 2.1 Introduction The purpose of this chapter is to document and analyse previous studies related to organic food and to provide an overview of consumers’ purchase behaviour of organic food in the global sense. It also seeks to review relevant literature of theories associated with this theme. Given that this study focuses on organic food, the chapter addresses relevant literature from well-developed and mature markets such as Europe, North America and Australia. This is undertaken by unpackaging the elements and determinants which influence consumers’ purchase behaviour of organic food (Section 2.2). Categories which play a role in the decision-making include demographic profiles of consumers (Section 2.3), purchasing motivations (Section 2.4), barriers to organic food consumption (Section 2.5) and purchase intention (Section 2.6). This chapter also seeks to identify themes that reflect the various rationales used by consumers when deciding to purchase organic food and explore possible theoretical approaches to explain their buyer behaviour (Section 2.7). It further evaluates and summarises previous studies (Section 2.8) and finally examines the limitation of previous studies (Section 2.9). Figure 2.1 diagrammatically illustrates the structure of Chapter Two. 25 Figure 2.1 Structure of Chapter Two Source: developed from Perry (1995) for this research 2.2 Understanding organic food consumers 2.2.1 Is organic food healthier? Although this study references industry experts who strongly argue that there are many scientific studies demonstrating the benefits of organic food (Leu 2009), this is not universally agreed on by researchers in the field. Even though organic food is perceived as healthier and safer, scientific evidence for such claims is scarce and, as is later 26 demonstrated, ‘organic’ does not automatically equal ‘safe’. It is debatable whether organic food production and its marketing is an opportunity or just hype (Bhaskaran et al. 2006). There have been very few scientific studies which compare the differences between organic food and conventional food in terms of their nutrient composition or their biological effects on animal or human subjects. Even small variations of levels of certain micronutrients have been reputed to have unlikely health implications for consumers (Williams 2002). A critical review (Brennan, Gallagher & McEachern 2003) of literature on UK consumers’ interest in organic food makes the claim that although consumers believe that organic food is healthier, more nutritious and tastes better, these beliefs are not scientifically demonstrated. In addition, food watchdogs warn that organic products may not be quite as healthy and environmentally friendly as consumers perceive. According to Magkos, Arvaniti and Zampelas (2003, 2006), there is a popular view that organic food is pesticidefree, though it does contains lower levels of agro-chemical products than conventional products. This research concurs with findings of a 2009 study in the American Journal of Clinical Nutrition (cited in Kluger 2010) that there is no significant nutritional difference between organic and conventional products with regard to all but three of the vitamins and other food components examined. So is buying organic food more for food safety or just peace of mind? One US study (Zhao et al. 2007) demonstrates that there are no significant differences between organic and conventional vegetables in terms of ‘consumer perceived’ sensory quality. It indicates that the demand for foods that are produced under environmentally sustainable standards has been slow to take off. Consumers do not perceive these products as offering any special benefits. They distrust the claims made by organisations and are confused by the use of different terminologies such as organic, green and environmentally friendly in promotion of food products (Zhao et al. 2007). 2.2.2 Consumer awareness and knowledge of organic food Though there is some level of confusion amongst consumers in relation to organic food, at the same time there is growing consumer awareness of organic food around the world and 27 especially in Western Europe (Aarset et al. 2004; Bonti-Ankomah & Yiridoe 2006). As introduced in the previous chapter, the terms ‘organic food’ and ‘green food’ have many different connotations and interpretations. There is little doubt that this has caused confusion in the minds of consumers worldwide (Aarset et al. 2004; Krystallis, Fotopoulos & Zotos 2006). This leads many to be sceptical about organic food products (Aarset et al. 2004; Bonti-Ankomah & Yiridoe 2006; Hill & Lynchehaun 2002). Barnes, Vergunst and Topp (2009) suggest that changes in consumer knowledge have been significant enough to affect organic consumption decisions. Consumers who are more aware of organic products tend to accept paying a premium for them (Chang & Zepeda 2005; Makatouni 2002). In contrast, an Austrian study (Gotschi et al. 2010) highlights that knowledge about organic food products does not necessarily lead to positive attitudes or higher levels of self-reported purchase behaviour of organic products. This conclusion is consistent with findings from a Belgian study (Verdurme, Gellynck & Viaene 2002) which states that knowledge about organic food does not correspond with negative attitudes towards genetically modified (GM) food, and therefore does not influence the consumers’ purchasing decision of organic food. 2.2.3 Differences between buyers and non-buyers of organic food In a Greek study which focuses on wine purchasing, Fotopoulos, Krystallis and Ness (2003) identify organic wine buyers’ motives in comparison to non-buyers. Healthiness, quality, information, attractiveness and good taste are the five main motivations for organic wine purchasers. The differences between organic and non-organic wine buyers derive from their distinctive purchasing behaviour, meaning an evaluation of consumers’ cognitive structures and the consumption motives with which wine’s organic character is associated. In a further study, Krystallis, Fotopoulos and Zotos (2006) investigated the differences between buyers and non-buyers of organic food in Greece. It found that organic consumers are less price sensitive and more interested in signs of food quality, evidenced by organic labelling and food production methods. However, both organic and non-organic buyers share the same perceptions, particularly in relation to food safety, as they both give importance to country of origin, physical structure, freshness, vitamin content and 28 nutritional value. Some of these views are supported by a US study (Williams & Hammitt 2000) which indicates that approximately 91% of organic buyers reported that they would be willing to pay higher prices for organically produced food. Only 46% of organic buyers rated price as an important attribute, compared with 64% of conventional buyers. This study also revealed that only 45% of organic buyers believed that the current US food supply is safe, compared to 79% of conventional buyers. Chang and Zepeda (2005) suggest that Australian organic food buyers tend to consider the effect of food on personal health, the environment, animal welfare and other social issues, while non-organic food buyers tend to focus more on products’ sensory appeal, such as taste and appearance. It also suggests that increasing consumers’ awareness of organic products and proper certification may be effective in increasing the organic food market. In their research on Indian consumers, Chakrabarti and Baisya (2007) observe that both regular and occasional organic food buyers tend to be motivated by health and nutritional concerns. However, regular organic buyers tend to attach more importance to taste whereas occasional purchasers tend to be just curious. Shepherd, Magnusson and Sjödén (2005) suggest that health consciousness in Sweden appears to be more important than environmental concerns in the purchase of organic foods. However, this trend differs between heavy and light consumers of organic products. 2.3 Demographic profiles of organic food consumers Past studies suggest that demographic profiles of consumers significantly impact their purchase behaviour of organic food (Lea & Worsley 2005; Tsakiridou et al. 2008). Socioeconomic variables such as age, education and income are important in the decisionmaking and purchase of organic food. Consumers with higher income and education levels display a strong correlation to food safety and environment concerns (Tsakiridou et al. 2008). Generally, consumers of organic food are from higher socio-economic segments (Connor & Douglas 2001; O’Donovan & McCarthy 2002). Another study reveals that Japanese organic consumers are from better-educated, more affluent urban centres and are more concerned about environmental and food safety issues 29 (Gendall, Betteridge & Bailey 1999). Similarly in Ireland, organic food consumers tend to be highly educated and from a high income category (Roddy, Cowan & Hutchinson 1996). A demographic profile shows them to be female, aged between 30-45, with children and having higher levels of disposable income (Davies, Titterington & Cochrane 1995). In Greece, socio-economic and demographic variables also seem to influence organic food purchasing. Studies (Fotopoulos & Krystallis 2002a, b; Tsakiridou, Mattas & TzimitraKalogianni 2006) reveal that higher educational and higher income groups display a high awareness and high purchase intention for organic food, while the non-organic users exhibit lower education and income levels. 2.3.1 Gender One of the key demographic variables for the purchase of organic food is gender. Studies (Lea & Worsley 2005; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006) suggest that women have more positive attitudes towards organic food and are more likely to purchase it regularly. This could be explained partly by the fact that they do the household shopping, and take a higher level of responsibility for the family, hence are more likely to be aware of organic food (Lockie et al. 2004; Onyango, Hallman and Bellows 2007). Another study (Ureña, Bernabéu & Olmeda 2008) acknowledges that although women have more favourable attitudes towards the purchase and consumption of organic food than men, they are less willing to pay a premium for it. Women are more concerned about health, nourishment and environmental information, but they also demand more information about organic food products and seek to take advantage of lower costs where available. Conversely, Roitner-Schobesberger et al.’s (2008) study curiously reveals that men in Bangkok seem more likely than women to purchase organic food. It suggests that men in Thailand have higher educational qualifications and earn higher income. Dahm, Samonte & Shows (2009) investigated students’ attitudes in the south-eastern US. Equal numbers of males and females knew the correct definition of the term organic, recognised the organic labels and expressed positive attitudes towards organic foods, underscoring the fact that gender is not important in the decision to purchase organic food. 30 This finding varies somewhat from other studies which have identified females as having the stronger opinion and attitude towards organic food. Thus, gender influence on organic food purchase behaviour could be considered to be debatable. 2.3.2 Age There are many studies which advance the notion that younger consumers are more likely to purchase organic food compared to older consumers (Onyango, Hallman & Bellows 2007). Budget constraints may indirectly influence purchase as many older consumers have limited disposable income (Davies, Titterington & Cochrane 1995; Rimal, Moon & Balasubramanian 2005). By contrast, Roitner-Schobesberger et al. (2008) reveal that organic food consumers in Thailand tend to be older. Older consumers (aged over 51) have more positive attitudes towards organic food than younger age groups. The older consumer in Thailand tends to have health issues, partly due to perceived health vulnerability and an awareness that they are generally at a higher health risk than younger consumers (Bhaskaran & Hardley 2002). In a Swedish study, Magnusson et al. (2001; 2003) suggest that there are no significant differences in the age groups in terms of their intention to buy organic food. As such it could be inferred that the importance of age might not be as clearcut as initially understood or perceived. 2.3.3 Education level There is a strong correlation between increasing consumption of organic food and increasing levels of formal education. In Australia, consumers with a general science education background are more inclined to be consumers of organic products (Lockie et al. 2002). A European study by Margetts et al. (1997) suggested that the level of education is the strongest factor influencing perceptions of a healthy diet. In Greece, education appears to have turned the unaware consumer into a highly motivated organic supporter, and on the whole educated respondents appeared to be more likely to be aware of nutrition information (Krystallis, Fotopoulos & Zotos 2006). Several studies (Arvanitoyannis, Krystallis & Kapirti 2003; Fotopoulos & Krystallis 2002a; Tsakiridou et al. 2008) assert that consumers with higher education are more willing to pay 31 a premium price for organic food. By contrast, an earlier study in the US (Thompson & Kidwell 1998) found a negative relationship between education and willingness to pay; shoppers with graduate or professional degrees were less likely to purchase organic produce as they were less trustful of the retail outlet, the certification process and the organic value. Lea and Worsley (2005) claim that educational level produced minimal differences in terms of propensity to purchase organic food, a view that is supported by other studies (Lin, Smith & Huang 2008; Rimal, Moon & Balasubramanian 2005). As a result, it is unclear to what extent level of education can be an influencing factor in the purchase of organic food. 2.3.4 Income level Higher income and education levels have a strong correlation with food safety and environment concerns (Tsakiridou et al. 2008). According to UK studies (Rimal, Moon & Balasubramanian 2005; Tregear, Dent & McGregor 1994), the level of disposable income is positively correlated with the purchasing of organic food. As was discovered in a Greek study, most organic food consumers were higher income earners, often professionals with good education levels (Chryssochoidis & Krystallis 2005). These findings contradict an earlier Greek study (Fotopoulos & Chryssochoidis 2000) which revealed that higher income in a household was not an essential issue in the purchase of organic food. Nevertheless, Lockie et al. (2002) in their study, based on Australian data, highlight that low income earners are not necessarily less interested in consuming organic products. Here too we are confronted with contrasting data and conclusions and as such need to maintain some reservations on the importance of the level of income of households as important demographic variables for organic food purchase behaviour. 2.3.5 Role of the family The role of the family is another potential influence on the choice of organic food (Margetts et al. 1997). Hill and Lynchehaun (2002) suggest that children have a major influence on decisions to purchase of organic products. Families are often introduced to organic food with the arrival of newly born babies, given the extra concern that families have for baby safety. Organic food purchasers are more likely to be married couples aged between 35-55 32 with children (McEachern & Willock 2004). According to one study, households with young children are willing to pay a higher price for organic foods than other households (Soler, Gil & Sánchez 2002). These studies concur with previous studies which testify to the presence of children in the family and to their encouraging role in the choice of organic food consumption (Huang 1996; Thompson & Kidwell 1998). By contrast, a UK study (Padel & Foster 2005) indicates that individual health was a more motivating reason than family health, even for parents with young children. ‘Frequent’ organic shoppers generally have fewer children than average. Households without children made a clear connection with organic food as a source of ‘enjoyment’. Surprisingly, a Swedish study (Magnusson et al. 2001) downplays the effect of children, indicating a minimal difference in organic food purchase intention between families with children and those without. 2.4 Purchase motivations Some studies have highlighted the significance of country focused organic products consumption. An early study by Beharrell and MacFie (1991) observed that differences in the ranking and intensity of reasons to buy organic food appear to be country specific. A cross-cultural study (Squires, Juric & Cornwell 2001) echoes that consumers from different countries might have different attitudes towards organic food consumption. Keegan (cited in Squires, Juric & Cornwell 2001) suggests that consumers’ needs and attitudes may differ depending on the state of development of the market and/or stage of the product’s life cycle. Another comparative study (Baker, Thompson & Engelken 2004) of German and UK consumers’ purchase behaviour of organic food found great similarity in the ‘health’ and ‘enjoyment’ value of organic food. The major difference was the absence among the UK group of any connection between organic foods and the ‘environment’. Several other studies have confirmed these findings. Some studies indicate that organic food purchasing motivations can be triggered by a series of concerns such as environmental and ethical considerations, organic food sensory appeal and health concerns (Arvanitoyannis, Krystallis & Kapirti 2003; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006). Most organic food consumers believe that it is better quality 33 since it is free of pesticides and chemical residues (Botonaki et al. 2006; Tsakiridou et al. 2008). Added to health considerations, buying organic food according to another study is also motivated by animal welfare and support for the local economy (Tsakiridou et al. 2008). McEachern and McClean (2002) suggest that purchasing motivations can vary from individual to individual, industry to industry and country to country. These motivations are identified as both self-interest-centred (better tasting and safer) and altruistic (environmental and ethical). The details of these motivations are discussed as follows. 2.4.1 Food attributes A study from Norway (Torjusen et al. 2001) found that the traditional food quality aspects such as appearance, freshness and taste, named ‘observation traits’, were important to all respondents. Most were also concerned about aspects related to production and processing, meaning foods with no harmful substances and the least possible additives. Those who purchased organic foods frequently were more concerned about ethical, environmental and health issues, named ‘reflection traits’. A US study (Onyango, Hallman & Bellows 2007) suggests that organic food preferences were related to its naturalness, such as non-artificial flavours or colourings. Organic food attribute variables were found significant and positively related to its purchases, and the regularity of purchases was influenced by the particular food attributes deemed important in a consumption decision. Those respondents who viewed food attributes (naturalness) to be extremely important were 35% more likely to buy organic food. A more recent study (Wirth, Stanton & Wiley 2011) reveals that organic apple attributes like flavour, texture and price are more important than any other attributes in terms of the preference for organic food by consumers. Taste is an equally important item as quality, followed in order by price, appearance and availability (Dahm, Samonte & Shows 2009). The above studies are consistent in their findings that appearance, taste and quality are the key issues involved. 2.4.2 Trends and fashion Studies have highlighted the influence of ‘trends’ and ‘fashion’ as aspects of organic food consumption behaviour (Hughner et al. 2007; Lockie et al. 2002; Roitner-Schobesberger et 34 al. 2008). In part this has been a product of the media marketing and consumer perceptions of the fashionable nature of organic food (Hill & Lynchehaun 2002). An Australian study (Lockie et al. 2002) indicates that some consumers are interested in organic food for reasons related to fashion and green lifestyle and consider it to be ‘trendy’. A later study (Lockie et al. 2004) shows that the major determinants of increasing organic consumption are concerns with the naturalness of food, the sensory and emotional experience of eating organic foods, and those who consider themselves as ‘green consumers’. This is reflected in the growing trend of recycling and in lower levels of concern with convenience in the purchase and preparation of food. According to Wier and Calverley (2002), consumers in industrialised countries are changing their lifestyle and habits and becoming increasingly interested in nutrition and health. At the same time, life is getting busier and they tend to spend less time preparing meals. Also the availability of processed organic food is limited. For regular buyers, consumption of organic food is part of their lifestyle and they have related interests in nature, society and the environment. Consumers who attempt to pursue a healthy diet and balanced lifestyle have stronger intentions of purchasing organic food products (de Magistris & Gracia 2008; Schifferstein & Oude Ophuis 1998). 2.4.3 Health consciousness There is a prevalent belief that organic food is healthier, safer, more nutritious and of higher quality (Chen 2007; Chryssochoidis & Krystallis 2005; Millock, Wier & Andersen 2004; Padel & Foster 2005; Radman 2005; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006). Health consciousness is considered to be a major motivation for purchase and consumption (Chakrabarti & Baisya 2007; Schifferstein & Oude Ophuis 1998; Shepherd, Magnusson & Sjödén 2005; Squires, Juric & Cornwell 2001). Health concern is directly and positively associated with the buying decision-making of organic food products (Chen 2007; Chinnici, D’Amico & Peorino 2002; Lea & Worsley 2005; Lockie et al. 2002). Consumers’ interest in health seems to correspond to their interest in and desire to use organic food (Connor & Douglas 2001; Schifferstein & Oude Ophuis 1998; Verhoef 2005). 35 These studies provide evidence that consumers consider health concerns as one of the most critical reasons that influence their purchase behaviour of organic food. Consumers are willing to pay a premium price for pesticide eliminated organic food (Aguirre 2007; Botonaki et al. 2006; Krystallis & Chryssohoidis 2005; Ureña, Bernabéu & Olmeda 2008). Even in newly emerging markets such as Macedonia, consumers are willing to pay a premium price for higher quality, taste, and ‘safety’ certified foods (Grannis, Hine & Thilmany 2001). These studies evidence the fact that consumers consider health concern as one of the most important motives to influence their purchase behaviour. 2.4.4 Organic food information and knowledge An Italian study (de Magistris & Gracia 2008) suggests that consumers with knowledge that organic food is healthier and higher quality show positive attitudes towards organic food, and this influences their purchasing decision. Obviously, enhancing consumers’ organic knowledge is vital to develop the market. This study also found that more information in the organic food market certainly improves consumers’ knowledge. This is supported by an earlier Swedish study in which Tanner and Kast (2003) suggest that organic food consumers have adequate knowledge to distinguish between environmental friendly and environmental harmful products. A study in the Netherlands (Stobbelaar et al. 2007) suggests that as adolescents are tomorrow’s consumers of organic food, understanding their attitudes can create potential business opportunities. This study found that most children are aware of organic food and consider it healthy and environmentally friendly. However, their knowledge about organic food is low and their willingness to buy it is also low. The importance of knowledge of organic foods is an issue affecting purchasing decisions, since consumers without information cannot differentiate the positive attributes of organic food from conventional alternatives (Yiridoe, Bonti-Ankomah & Martin 2005). Greater information on the organic food market builds higher consumer knowledge and also positively influences their attitudes towards organic food products. 36 2.4.5 Impact on food safety issues The ‘mad cow’ disease scare in 1996 impacting UK beef and the dioxin crisis in Belgium in 1999 resulted in consumers in those countries looking for alternative healthier, safer food, with organic and natural products (McDonald 2001). According to a number of studies (McEachern & McClean 2002; Rimal, Moon & Balasubramanian 2005; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006), food safety is the one of essential criteria to affect consumer purchase of organic food. The increase of food safety awareness can also impact on attitudes towards specific food products. As economic development and income increases across many parts of the world, consumers demand higher levels of safety and quality in their food (Botonaki et al. 2006). Food safety is receiving more attention from governments and policy makers, health professionals, the food industry and the public. Recent Chinese food scandals have increased consumers’ demand for ‘safer food’. As the Ministry of Agriculture points out: ‘The food security in China will have an important bearing on world food security’ (in Paull 2008a, p. 11). The Chinese Ministry of Health (in Peng 2006) reported in 2005 that there were 256 food poisoning incidents, 9021 people fell ill from food poisoning, and 235 people died. Fu (in Peng 2006) states that the figures are much higher, with approximately 200,000 to 300,000 people being poisoned annually due to food safety issues in China. The government faces the challenge of maintaining credibility of food safety and nutrition in the world’s most populated country. Studies in Sweden (Magnusson et al. 2001; Magnusson et al. 2003) found that egotistical motives (health benefits for the individual or the family) rather than altruistic motives (benefits to society, such as environment and animal welfare) were the drivers of consumption of organic foods. A Scottish study (McEachern & McClean 2002) also portrays organic food consumers as being more self-interest-centred than altruistic. 2.4.6 Environmental concerns Organic agriculture initially developed in response to agricultural industrialisation (Torjusen et al. 2001). In general, materials and methods that enhance the ecological 37 balance of natural systems are used in organic production. Organic food is believed to be produced without pesticides, herbicides, inorganic fertilisers, antibiotics and growth hormones. Certain consumers are influenced by feelings of moral obligation and responsibility towards the environment. Thus, one motivation for buying organic food is driven by growing interest in ecological values and concern about environmental sustainability and threat of conventional agricultural production (McDonald 2001; Sanjuán et al. 2003; Tsakiridou et al. 2008). A few studies (Honkanen, Verplanken & Olsen 2006; Magnusson et al. 2003; Makatouni 2002; McEachern & Willock 2004) also demonstrate that ecological motives have the strongest impact on attitudes towards organic food, which highlights the important role of environmental issues influencing organic food consumption. However, a Greek study (Chryssochoidis & Krystallis 2005) argues that public related issues such as care for nature and the environment may not be relevant to organic food purchase in some societies. In an Indian study, the environment was not a major motivation for organic buyers (Chakrabarti & Baisya 2007). Further findings from an earlier study (Jain and Kaur 2004) indicate that consumers in India were less aware of environmental and ecological issues than those in developed countries. It would appear that, in a developing country like India, true green consumerism is in its infancy. Developing countries face other serious problems, such as overpopulation, poverty, unemployment, high level of illiteracy and disparity of wealth. 2.4.7 Animal rights and social justice Harper and Macaroni (2002) indicate that ethical concerns, especially in relation to standards of animal welfare, play a significant role in the decision to purchase organic food in UK. Moreover, such standards are perceived as a sign of the safety and healthiness of food. Another UK study found that values related to animal welfare were the second most important attributes which influenced consumers to purchase organic food. This study suggests that the lives of animals have an impact on human beings. Hence the animals’ living condition can affect human health, which then suggests that happy animals produce healthy meat (Makatouni 2002). A further study by McEachern and Willock (2004) asserts that animal welfare, especially related to animal transport and housing, influences organic 38 dairy and meat consumption. Buying organic food is seen as supporting the implementation of higher standards of animal welfare. Another study conducted in the US suggests that consumers purchasing organic food are motivated by issues of animal rights and social justice. Standards for the humane treatment of animals had the highest support among respondents who were mainly women, EuropeanAmericans, young and frequent organic food purchasers (Howard & Allen 2006). These views are seen primarily in developed countries and contrast with the attitude in developing countries where there is a greater need to contend with matters more directly related to day to day survival. 2.4.8 Country of origin Consumers seem to consider quality as being related to the geographical origin of production (Cicia, Giudice & Scarpa 2002). Studies demonstrate that country of origin is an important issue for organic food consumers. Norwegian consumers, for example, trust locally produced organic food as being safer than imported cheaper organic food. This then influences purchasing locally produced organic food and supporting local businesses (Storstad & Bjørkhaug 2003; Torjusen et al. 2001). Locally grown products were the second most popular choice by California respondents, after their first choice was standards for the humane treatment of animals (Howard & Allen 2006). Botonaki et al. (2006) propose that Greek consumers’ willingness to buy organic products is affected by their attitudes towards country of origin, and this also reflects the perceived organic food quality attributes. Greek consumers, though they might be aware of organic food, will still remain ‘traditional’ in terms of their cuisine and restaurant selection. Country of origin of a food product seems a powerful component of the promotional mix, since Greek organic buyers exhibit a very strong ethnocentric tendency (Fotopoulos & Krystallis 2002a, b). A Japanese study confirms that most respondents prefer to purchase locally produced organic food. However, if they were to purchase imported organic produce, they would consider choosing it from Australia, New Zealand, Canada, the US, Germany, England and China (Gendall, Betteridge & Bailey 1999). These exporting countries are seen as having 39 clean environments, clean water, air and soil. McEachern and Willock (2004) note that one of the obstacles preventing UK consumers from purchasing organic meat was the appearance of mainly imported products. Consumers are not keen to purchase organic meat originating outside the UK. The concept of meat safety was also central to attitudes, particularly within the context of country of origin. 2.5 Barriers to organic food consumption A range of studies have underscored the barriers which exist to purchasing organic food. An Italian study undertaken in 2002 (Chinnici, D’Amico & Peorino 2002) revealed that these ranged from purchasing location, purchasing difficulties (lack of choice), percentage of food expenditure devoted to organic products, perception of organic price, price differential and willingness to pay. Another study noted that people who live in an urban centre are more likely to purchase organic food, since they have easier access to diverse organic food products (Lockie et al. 2004). Greek studies (Krystallis, Fotopoulos & Zotos 2006; Tsakiridou et al. 2008) concur that the lack of product availability and higher prices are barriers to purchasing organic products. It was noted that there was a low level of consumer awareness and low knowledge of organic certification systems (Botonaki et al. 2006; Krystallis, Fotopoulos & Zotos 2006). 2.5.1 Price It is widely understood that organic farming processes cost more than conventional farming. One of the reasons is a greater loss of organic production to insects and disease if artificial pesticides and fertilisers are not used (Connor & Douglas 2001). Organic products tend to be more expensive than non-organic products. Several studies (O’Donovan & McCarthy 2002; Roddy, Cowan & Hutchinson 1996; Shepherd, Magnusson & Sjödén 2005; Tregear, Dent & McGregor 1994) indicate that price premiums of organic food tend to negatively affect purchase. If the organic food price was as low as the non-organic one, consumers would be willing to purchase more organic products. A study conducted by McEachern and Willock (2004) showed that barriers to purchasing organic meat were its high price and lack of perceived difference in taste. Padel and Foster (2005) show that price is not, in all 40 circumstances, an absolute barrier but only one factor in a complex decision-making process. Consumers consider price in the context of disposable income, but also as a factor of ‘value for money’ and the need to justify a premium through other gains. Less wealthy consumers are generally not interested in purchasing organic foods. It also could be argued that price was not always the main reason for the reluctance to purchase organic food (related to food choice). Conner (in Onyango, Hallman & Bellows 2007) found a connection between paying premium price and beliefs in the superiority of organic foods and their health benefits. While small communities appear to favour the local economy and thus support local organic producers, price as a factor did not seem to influence their purchase decisions. This is contradicted by Anders and Moeser (2008) who argue that organic consumers show a significant level of price sensitivity, with organic beef cuts perceived as luxury goods. Chang, Zepeda and Griffith (2005) in their overview of the Australian organic sector, highlighted that previously income and price had been important factors in the determination of purchasing of organic food but more recently this has declined in its importance. Willingness to pay for organic food is influenced by factors including food quality, safety and trust in the certification (Krystallis & Chryssohoidis 2005). 2.5.2 Willingness to pay Several studies (Botonaki et al. 2006; Cicia, Giudice & Scarpa 2002; Krystallis & Chryssohoidis 2005) indicate that consumers consider price as a signal of quality, or a form of proxy for quality. In a Canadian study, consumers were willing to pay premium prices for the products they considered ‘healthier’. It it was argued that the organic food industry could improve profitability from the introduction of organic functional foods (Larue et al. 2004). Another study (Thompson & Kidwell 1998) indicates that price premiums for organic products ranged from 40% to 174% in comparison to conventional products in the US market. Lockie, Halpin and Pearson (2006) summarised previous studies on the average national retail price premiums for certified organic food from 10% to 15% in Germany, 20% to 30% in Austria, 80% in Australia, and 10% to 100% in the US and the UK. Retail premiums of 25% to 30% are acceptable for most European consumers. 41 A Greek study (Krystallis & Chryssohoidis 2005) points out that consumers’ willingness to pay a premium on organic food across product categories was higher in relation to those foods that were purchased more frequently. Lin, Smith and Huang (2008) acknowledge that the organic price premium payment by US consumers can vary from 20% to 60% according to the product variation. The price and cost of organic food is important in the final purchasing decision. An Italian study suggests that consumers are willing to pay up to 30% more on organic foods than on corresponding conventional ones (Chinnici, D’Amico & Peorino 2002). A Spanish study (Ureña, Bernabéu & Olmeda 2008) explained that occasional consumers of organic food were willing to pay approximately 10% higher premium, while regular consumers would pay a premium of approximately 15% extra. An Irish study (O’Donovan & McCarthy 2002) demonstrates that 70% of consumers were not willing to pay more than 10% extra. Not surprisingly, consumers who enjoy the benefits of consumption and perceived value of organic food are willing to pay premium prices. Overall, willingness to pay premium prices for organic food varies in different markets. This poses a challenge to marketers involved in setting price levels. 2.5.3 Labelling and certification Consumers seek clear, accurate and reliable information about organic food, especially in labelling. Usually they will have seen a label or logo defining its organic credentials, and they express a need for more information about organic food (Giannakas 2002; Lin, Smith & Huang 2008). There is a relatively poor understanding of the legal use of the term ‘organic’ on food products, including inspection and certification systems. This can often have adverse effects on the consumption of organic food. However, trustworthy labels guarantee the authenticity of organic food products (Padel & Foster 2005; Wier & Calverley 2002). Both eco and fair trade labelling in Sweden make products more attractive (Grankvist, Lekedal & Marmendal 2007). A US study (Bellows et al. 2008) indicates that more people value organic production methods as shown through labelling. The claims of buying organics and therefore clear labelling, and placing importance on organic production systems when deciding what to eat, 42 are highly correlated. However, only about 27% of all of those who value organic farming production methods are regular organic purchasers. An earlier study in Greece found that organic food buyers who were the ‘consumption leaders’ believed that differences between organic and non-organic food products were through their trust in the ‘organic’ labels (Fotopoulos & Chryssochoidis 2000). A similar study in Greece in 2006 found that food labels appeared to have a significant influence on consumers’ purchases (Tsakiridou, Mattas & Tzimitra-Kalogianni 2006). However, almost half of organic food buyers are confused as to the differences between organic and conventional foods (Krystallis, Fotopoulos & Zotos 2006). It is believed that organic labelling and certification may play a larger role in purchase decisions. 2.5.4 Lack of availability One of the reasons for consumers not buying organic food is perceived lack of availability. A study in Italy in 2002 (Chinnici, D’Amico & Peorino 2002) suggests that location and purchasing difficulties (i.e. lack of choice) are amongst the most important components. A Danish study refers to people living in capital cities who are more likely to purchase organic food, since they have easier access to diverse products (Millock, Wier & Andersen 2004). Similarly, a study in the UK (Padel & Foster 2005) discovered that 26% of consumers in Wales did not know where to find organic products, and a further 30% found it difficult to source organic food. Likewise, O’Donovan and McCarthy’s study (2002) reports that among Irish respondents who did not purchase organic food, 28% cited lack of availability, and also 85% of nonpurchasers suggested that they would purchase organic food if it became available at their regular purchasing place. Even in the emerging market of Slovenia, the purchase of organic food is most influenced by availability in retail outlets. Clearly, the importance of organic food availability favours effective distribution activities, and it also indicates the emerging initiative of promoting local farmers markets and retail chain distribution of organic food (Kuhar & Juvancic 2010). 43 2.6 Pre-purchase intention Literature has demonstrated that consumers’ attitudes, values, beliefs and purchase intention may influence the purchase of organic food. Key findings from selected studies are introduced in the following sections. 2.6.1 Attitudes An early Danish study (Grunert & Juhl 1995) found that school teachers who had environmentally sensitive attitudes were more likely to purchase organic foods, and positive attitudes towards environmental issues were found to be positively correlated with frequency of organic food purchase. A Norwegian study (Honkanen, Verplanken and Olsen 2006) determined a strong correlation between ecological motives and attitudes, which in turn were statistically significant in explaining purchase behaviour of organic food. Wong (2004), in examining organic food purchase behaviour in New Zealand, concluded that key factors were personal based attitudes and values derived from a range of influences including attitudes towards health, ecology, society, family and politics. Attitudes are also derived from sensory effects such as taste and appearance. This supports the view which identifies value systems and their influence on consumers’ organic food purchase and consumption. These findings are in line with an Italian study (Zanoli & Naspetti 2002) in which most respondents held positive attitudes towards organic foods, even though these were expensive and not always readily available. Consumers’ purchase of organic food is based on subjective experiences and perceptions (Hughner et al. 2007). Kirk et al. (cited in Lodorfos & Dennis 2008) found that respondents had positive attitudes towards organic food, which resulted from their perception of it being better quality than conventional food. Similarly, Chen (2007) investigated consumers’ attitudes and purchase intention in relation to organic food in Taiwan. She discovered that it was perceived as healthier, natural, nutritious and environmentally friendly. The respondents’ attitudes towards organic foods purchase was positively related to the attitude towards such types of food. Chen’s (2009) second study indicates that health consciousness and environmental attitudes influence consumers’ attitudes towards organic foods. Positive attitudes can be determined by health 44 consciousness and environmental attitudes if they are keen to undertake healthy activities, such as organic food consumption (Chen 2009). This has profound implications for understanding mainland Chinese consumers’ purchase behaviour since consumers in Taiwan and in China share a similar cultural heritage and are heavily influenced by Confucianism and traditional values (Tai & Tam 1997). Another important study (Shepherd, Magnusson & Sjödén 2005) demonstrates a discrepancy between attitudes and behaviour of consumers having positive attitudes towards organic foods who, despite this, may not purchase or value the superiority of organic food. Magnusson at al. (2001) found that although between 46% and 67% of respondents held positive attitudes, only 4% to 10% of them indicated an intention to purchase different types of organic food. On the other hand, Roddy, Cowan & Hutchinson (1994) argue that organic food attributes such as higher price, unclear promotion and packaging can generate negative attitudes towards their purchase. However, another study (Tsakiridou, Mattas & Tzimitra-Kalogianni 2006) disputes these findings about attitudes and values, denying their effect on the consumption of organic food. Consumer attitudes towards organic products were not found to be significant determinants in the purchase of organic food. 2.6.2 Values A study in the Netherlands (Grunert and Juhl 1995) defined values as cognitive patterns by which individuals orientate themselves in society. They drive much consumer behaviour, and while people may possess the same values, they may have different value systems. Values can be viewed as representing motivations, since they are used by individuals to select and justify actions (Grunert & Juhl 1995). A Greek study (Chryssochoidis & Krystallis 2005) contends that a better understanding of consumers’ value structures is important in determining their motivation to purchase organic food. Based on their values, Greek consumers of organic foods can be categorised into three clusters, in order of importance: health-consciousness, pursuit of hedonism and environmental consciousness. Values have been determined to have a direct relation to organic food consumers’ 45 behaviour (Baker, Thompson & Engelken 2004; Chryssochoidis & Krystallis 2005; Essoussi & Zahaf 2008; Makatouni 2002; Zanoli & Naspetti 2002). Millock, Wier, and Andersen (2004) supported by a further study (Lea & Worsley 2005), indicate that organic food consumers have both public and private sets of values. Consumers who assign personal values such as personal health, taste and freshness are more likely to purchase organic food, while those who perceive public values such as environmental and animal welfare attributes are less likely to do so. 2.6.3 Beliefs An early Northern Ireland study (Davies, Titterington & Cochrane 1995) reveals that organic food is perceived as food without ‘chemicals’ and ‘hormones’. Consumers purchase organic food as they perceive it to contain lower pesticide and fertiliser residues. A UK study (Harper and Makatouni 2002) suggests that both buyers and non-buyers have similar perceptions of what organic food means. Perceptions are affected by beliefs about the safety and quality of conventional food production, and subsequent attitudes to conventional versus organic products. Similarly, a Belgian survey (Verdurme, Gellynck & Viaene 2002) focused on consumer beliefs (perceptions) about food production in general, organic food and GM food. 2 Respondents believed that food had become less safe, less tasty, less healthy and a bit more expensive than 20 years ago, while organic food was perceived to be healthy, safe and environmentally friendly compared to the negative attitudes harboured by respondents towards GM food. One of the first studies which examined the relationship between personal values and beliefs associated with organic food reveals that self-transcendence values are related to the purchase of organic food. Personal values related to nature, environment and equality were discovered to be the dominant predictor of positive beliefs associated with organic food 2 GM (genetically modified) food is derived from genetically modified organisms which have had specific changes introduced into their DNA by genetic engineering techniques. 46 (Lea and Worsley 2005). These studies suggest that consumers’ beliefs and perceptions towards organic food could be interesting to investigate. 2.6.4 Purchase intention Purchase intention is influenced by a number of elements. Attitudes, for instance, were found to be significant predictors of intention to purchase organic fruit and vegetables (Saba & Messina 2003). Consumers perceive organic food to possess better quality attributes than ordinary food and obviously these positive attitudes are likely to transform into positive intentions for its purchase (Honkanen, Verplanken & Olsen 2006). An Indian study argued that consumers’ intentions to buy organic food are influenced by positive attitudes towards the importance of personal health (Chakrabarti & Baisya 2007). Several studies have reaffirmed that factors such as health consciousness, subjective norms, familiarity and quality of organic food all have some bearing on purchasing intentions (Magnusson et al. 2001; Magnusson et al. 2003; Smith & Paladino 2010). According to an exhaustive Taiwanese study (Chen 2007), purchase intention are strongly influenced by six key food choice motives: mood, natural content, animal welfare, environmental protection, political values and religion. Shaharudin et al. (2010a; 2010c) claim that Malaysian consumers’ purchase intention of organic food is influenced by their perceived value and health consciousness. Food safety is considered to be of less importance in this respect. These findings concur with some western studies in that the perceived value has significant impact on consumers’ desire to purchase organic food and their willingness to pay the extra cost. Many positive attitudes towards organic food emerged on the basis of perceptions of it as being of better quality. Consumers with positive attitudes towards organic food are invariably more likely to form positive intentions to purchase it (Honkanen, Verplanken & Olsen 2006; Saba & Messina 2003; Sparks & Shepherd 1992). A positive attitude is one of the more important motivations, and very likely to influence the purchase intention. 47 However, although consumers may have positive attitudes towards organic food, this may not be translated into greater levels of purchase on a regular basis (Fotopoulos & Krystallis 2002a; Roddy, Cowan & Hutchinson 1996). Verdurme, Gellynck and Viaene (2002) also question the relationship between organic food buying intention and attitudes/beliefs. Organic food has been perceived as healthy, safe and nutritious, and consumers purchase organic food products mainly because of the belief in their beneficial effects. 2.7 Theories associated with previous studies Previous research has largely used the Consumer Decision Process (CDP) model (Blackwell, Miniard & Engel 2006) and the Theory of Planned Behaviour (TPB) to understand issues relating to organic food consumption. These have provided theoretical support and explored the motivations and barriers associated with the consumption of organic food. They also analysed consumers’ attitudes, purchase intention and purchase behaviour towards organic products, and investigated various determinants which influenced organic food consumers in making their purchasing decision. Table 2.1 summarises previous studies in alphabetical order and the respective use of these theories. The TPB was the model most encountered in the literature which primarily examined consumer attitudes in the psychological context. On the other hand, the CDP model examined the interaction between environmental and individual factors with respect to consumer purchase behaviour. These will be discussed in greater detail later in this study. Table 2.1 Summary of key attributes measured with associated theory in studies relating to organic products Author and year Key attributes measured Associated theory Barnes,Vergunst & Topp 2009 Knowledge of organic product Consumer Decision Process Chakrabarti and Baisya 2007 Attitudes Theory of Planned Behaviour Chen 2007 Attitudes, purchase intention Theory of Panned Behaviour Chinnici, D’Amico & Peorino 2002 Price, lifestyle Consumer Decision Process Chryssochoidis 2000 Information search Consumer Decision Process Guido et al. 2010 Purchase intention Theory of Planned Behaviour 48 Essoussi & Zahaf 2008 Knowledge, labelling, certification, trust Consumer Decision Process Gotschi et al. 2010 Knowledge of organic product Theory of Reasoned Action Hill & Lynchehaun 2002 Internal and external factors Consumer Decision Process Honkanen, Verplanken, & Olsen 2006 Motives, attitudes, intention Theory of Planned Behaviour Krystallis, Arvanitoyannis & Chryssohoidis 2006, Pre-purchase evaluation, product Consumer Decision Process attributes Lodorfos & Dennis 2008 Purchase intention Theory of Planned Behaviour de Magistris & Gracia 2008 Lifestyle, attitudes, purchase intention Consumer Decision Process Theory of Planned Behaviour Magnusson et al. 2001; Magnusson et al. 2003 Sensory characteristics, purchasing intention Theory of Reasoned Action McEachern & Willock 2004 Attitudes, behaviour Theory of Reasoned Action Theory of Planned Behaviour McEachern, & McClean 2002 Perception, attitudes, knowledge, purchasing behaviour Consumer Decision Process Padel & Foster 2005 Motivation/barrier to purchasing Consumer Decision Process Rimal, Moon & Balasubramanian 2005 Food safety Consumer Decision Process Saba & Messina 2003 Attitudes, intention Theory of reasoned action Theory of Planned Behaviour Schifferstein & Oude Ophuis 1998 Lifestyle, behaviour Consumer Decision Process Salleh et al. 2010 Purchase intention Theory of Planned Behaviour Shaharudin et al. 2010a; 2010b Purchase intention Theory of reasoned action Theory of Planned Behaviour Shepherd, Magnusson & Sjödén 2005 Attitudes Theory of Planned Behaviour Smith & Paladino 2010 Attitudes, intention, behaviour Theory of Reasoned Action Theory of Planned Behaviour Thøgersen 2009 Attitudes, intention Theory of Planned Behaviour Vindigni, Janssen & Jager 2002 Behaviour modelling Consumer Decision Process Source: Author, 2011 49 2.7.1 The Consumer Decision Process (CDP) model applied to food consumption There are many models related to consumer behaviour in the marketing discipline. One of the classical models utilised in the measurement of consumer buyer behaviour is the Consumer Decision Process (CDP) model (Blackwell, Miniard & Engel 2006). In recent years, there has been growing interest in the study of consumer behaviour for organic food along with the way in which consumers make their purchasing decisions (Vindigni, Janssen & Jager 2002). The CDP model was first proposed in 1968 by Engel, Kollatt and Blackwell. It has been developed and modified to a more advanced and sophisticated model comprising seven stages. Figure 2.2 portrays the model as ‘a road map of consumers’ mind that marketers and managers can use to help guide product mix, communication and sales strategies’ (Blackwell, Miniard & Engel 2006, p. 70). Consumer decision-making is influenced and shaped by many factors and determinants can be categorised into three sectors: individual differences, environmental influences and psychological processes. A full description of each level of the process is given after Figure 2.2. 50 Figure 2.2 The Consumer Decision Process 51 Stage One: Need recognition This stage is the start of the consumer decision-making process – the recognition of a need – in which consumers locate a product ‘to satisfy a need or want’ (Kotler et al. 2001, p. 8). Need recognition depends on how much discrepancy exists between a consumer’s current situation and the situation they want to be in. It can be triggered by discontent with existing products. Wier and Calverley (2002) suggest that product specific characteristics that benefit consumers provide the major motivation for them to purchase organic foods. However, this type of motivation varies between different consumer segments. Consumers could be motivated by organic food that appears fresh in appearance, which signifies quality, or by reliable nutritional content and absence of chemical residuals. This stage demonstrates that consumers have desires, and believe in products that have the ability to resolve their problems, hence they are worth purchasing sooner or later. Marketers examine consumers’ desires, including ability and authority to purchase. Hence this stage provides the platform in identifying the motivations of consumers intending to purchase or not to purchase organic food products. Blackwell, Miniard, and Engel (2006) also identified environmental influences and individual differences that impact consumers’ level of need. The questions one might ask in the context of organic foods are: What categories of Chinese consumers have the need to purchase organic food? Are they motivated by health consciousness, environmental issues or perceptions of organic products’ attributes? Stage Two: Search for information The second stage involves search for information about available purchase options. Once consumers have experienced the need for products or services, they have to search for the ‘motivated activation of knowledge stored in memory or acquisition of information from the environment about potential need satisfiers’ (Blackwell, Miniard & Engel 2006, p. 109). Information search may be determined by internal factors, scrutinising knowledge stored in memory, and by external information from the marketplace (Kotler et al. 2001). 52 Steenkamp (1997) claims that previous experience with food products is the most important information source. For example, the search for information about fresh meat might depend on consumers’ consumption experiences, information provided by retailers such as supermarkets or butchers, and information provided on the packaging. The questions one might ask in the context of organic food purchase are: Do consumers in their search for information rely on their existing knowledge about organic food? What is the level of consumers’ confidence in their existing knowledge? What is the level of satisfaction with their previous experience of consumption of organic food? When internal search proves inadequate, additional information might be collected. Stage Three: Pre-purchase evaluation of alternatives During this stage, consumers reflect on new or pre-existing evaluations stored in their memory and seek to get answers when evaluating products’ attributes, brands, personal values and lifestyle. They utilise different evaluative criteria, and the choice is influenced by both individual and environmental differences (Blackwell, Miniard & Engel 2006). Purchase behaviour is affected by many elements including demographics, psychographics, values, personality, consumer resources, motivation, knowledge and attitudes. It is also influenced by culture, social class, family, personal influence and situational behaviour (Steenkamp 1997). Consumers either like or dislike particular products based on attitudes towards available alternatives (Fishbein & Ajzen 1975). Stage Four: Purchase After the pre-purchase evaluation of alternatives stage, consumers move to the next stage which is that of purchase. Consumers first choose the place of purchase, which might be a general store, retail or speciality store, supermarket or online purchase, and then evaluate the alternatives available in-store. They may decide not to proceed further because of their experiences during the purchasing stage. Purchase behaviour is also influenced by shopping location preferences, frequency of purchase and advertising (Blackwell, Miniard & Engel 2006). In the context of consumers purchasing organic food in China, they might purchase particular types like imported or domestically produced. Other factors influencing their purchase might be branding, type of sales outlets and assistance provided during sales. They might also examine features of 53 organic food products such as appearance, quality, freshness, availability, presentation and pricing level. Stage Five: Consumption After the purchase is made, the consumption of the products occurs. It is important to identify the products’ end user. Sloof et al. (in Sun & Collins 2006) contend that the consumption of perishable products such as fruit might depend on the intended uses of the product, whether it is meant for personal consumption or as a gift to others. Consumers place more importance on the quality of the product if it is meant for the family or for children; whereas they would also look for attractive packaging if it is meant to be a gift. Stage Six: Post-consumption evaluation Following the purchase and consumption of the product, satisfaction or dissatisfaction occurs. If the products’ attributes do not match the buyer’s purchase intention, or if the consumer does not use the product properly, dissatisfaction may occur (Blackwell, Miniard & Engel 2006). When the consumer’s expectation matches the product’s perceived performance, satisfaction occurs. The consumer may re-purchase the product and also recommend it to others. Emotions play an important role in the postconsumption evaluation stage. Stage Seven: Divestment This is the final stage of the CDP model. Consumers might react differently during this stage. For example, they might dispose, recycle or re-sell the product (Blackwell, Miniard & Engel 2006). Post-purchase behaviour is based on satisfaction or dissatisfaction with the consumption (Kotler et al. 2001). Later in this chapter, other relevant variables which influence consumers’ organic food consumption are examined. The first three stages of this model – need recognition, information search and pre-purchase evaluation of alternatives – form the first phase of the CDP model. These three stages, and not all the seven stages of the model, are used in this study to evaluate consumers’ purchase intention towards organic food in China. 54 2.7.1.1 Applicability of the CDP model and Hierarchy of Effects model in the context of food A significant Belgian study was conducted by Verbeke (1999, 2000), built on a fourstage conceptual framework (Engel, Miniard & Blackwell 1993; Engel, Kollat & Blackwell 1968) which was designed for consumer decision-making for fresh meat, as seen in Figure 2.3. The stages of this process include problem recognition, information search, alternative evaluation and choice/behaviour. The model is linked and integrated, firstly, with a Hierarchy of Effects model as proposed by Lavidge and Steiner (1961) and later adapted by Barry and Howard (1990). This model indicates the different mental stages that consumers go through when making purchasing decisions and responding to messages in sequence. It is generally agreed that the structure includes cognitive (learning, knowing), affective (thinking, feeling) and conative (intending, doing) components, although there is no clear-cut evidence about the sequence and inter-distance of these hierarchical stages (Verbeke 2000). Dubé, Cervellon and Han (2003) demontrate that a hierarchical model performed better than any alternative representation in capturing consumers’ attitudes and also proved to have superior ability to predict purchase behaviour of food. Secondly, concepts related to information-processing as presented by McGuire (1978) and re-elaborated by Scholten (1996) are supplemented in Verbeke’s (1999, 2000) model. There is a linkage to potential influences on consumer decision-making from the impact of communication and marketing. The information-process concept is underpinned in the study of persuasion in social psychology, and focuses on the impact from persuasive communication (Verbeke 2000). Finally, a classification of factors or variables that potentially influence consumer decision-making is adopted (Pilgrim 1957; Steenkamp 1997). The Pilgrim model suggested that the acceptance of food behaviour can be grouped into attitudes, sensory tests and consumption. Steenkamp echoed this in that generally there are three groups of factors influencing consumption of food: properties of the food, personal related factors and environmental factors. 55 In the main, Verbeke’s model is entwinned in the CDP model. These include need recognition (or problem recognition), information search and alternative evaluation as essential stages before a choice of product purchase can be affected (Steenkamp 1997; Verbeke 1999, 2000). Verbeke’s research is valuable for analysing consumer decision-making, in this case, towards fresh meat. The results conclude that consumers go through relatively extended problem solving and intensive evaluation of food attributes: ‘The rather intensive decision-making process, with extended evaluation of product alternatives on a large number of attributes, points towards relatively high involvement with respect to fresh meat consumption decision and includes periods of continuous attention to food and meat safety issues’ (Verbeke 2000, p. 530). Figure 2.3 depicts Verbeke’s conceptual framework for analysing consumer decision-making towards the purchase of meat. 56 Source: Verbeke 1999, 2000 Figure 2.3 Conceptual framework for analysing consumer decision-making towards the purchase of meat 57 The CDP model is also used in a study by Bareham (1995) which examines various factors that influenced European consumers’ purchase of food and beverages. Bareham uses both Engel, Blackwell and Miniard’s (1990) consumer behaviour approach and Fishbein & Ajzen’s (1975) attitudes and behaviour measurement theories. His study covers political, economic and technical influences, cultural and social influences, psychological and marketing influences. Non-economic factors include health and diet concerns, the growth of vegetarianism, the convenience of buying food, the life cycle of the households and advertising impact on consumers’ food choice. Bareham (1995, p. 98) also argues that ‘taboos, religious doctrines, cults, scares and the status of different foods can all have a considerable impact on food consumption’. (Another study undertaken by Wheelock (in Mitsostergios & Skiadas 1994) suggests that food consumption can be influenced by economic and non-economic factors. Food price and personal income have been two dominant factors post-1945 which have determined consumer food purchase. However, the influence of prices on the choice of food is diminishing. In addition, Cronin, Brady and Hult (2000) suggest that the consumer decision-making process is the best model that incorporates both direct and indirect influences on consumer behavioural intentions. The study by Furst et al. (1996) identifies value factors in the food choice process, including sensory perceptions, monetary considerations, health and nutrition, managing relationships and quality. Similarly, Lennernas et al. (1997) suggest that taste, health, long shelf-life, quality or freshness, price, family preference and habits are the most important factors which impact the choice of food. These studies concur with an organic food study by Wier and Calverley (2002) in eliciting personal user values, such as products’ health benefits, taste and freshness, which are more important than public values, such as environmental and animal welfare. Hansen (2001) found many similarities in the motives of Chinese and Western buyers regarding the purchase decision behaviour of seafood, with the most important factors being high product quality, suppliers’ reliability and service level, product availability, product certification by an authorised organisation and competitive prices. 58 2.7.1.2 Applicability of the CDP model to organic food An earlier study in the Netherlands (Schifferstein & Oude Ophuis 1998) investigated various determinants of organic food consumption and their interrelationship using the CDP model. They suggested that in the problem recognition stage, the primary reason for buying organic food was some manifested or potential health problem. Furthermore, findings from the evaluation of alternatives stage suggest that health food buyers are attaching more importance to absence of food additives, health value, naturalness and biodynamic influences, and giving less importance to freshness and appearance of food. Vindigni, Janssen and Jager (2002, p. 634) remark that they have witnessed in recent years a large amount of interest in the study of consumers’ behaviour towards organic food: ‘Various approaches have been developed aiming at describing, modelling and predicting the various ways that consumers make purchase decisions’. Sanjuán et al. (2003) suggest that it could be valuable to develop a model to analyse the different influences of independent variables on the consumer decision-making process in future organic studies. An exploratory Canadian study (Essoussi & Zahaf 2008) provides another theoretical contribution to understand decision-making processes with regard to organic food. This study has an integrative and complex structure, and explains not only how, what and why consumers buy or don’t buy in terms of organic food, but also examines macro-environmental factors (in Canada). Five themes relating to the consumer decision process of community organic food consumers were identified: organic food definition and recognition, consumer’ motivation, trust with regards to organic food, labelling and certification process, and organic food distribution channels. Similarly, de Magistris & Gracia (2008) investigated the consumer decision process of organically produced food products. It found consumers’ attitudes towards health attributes, such as healthy diet and balanced lifestyle, and towards the environment are the most important factors which explain consumers’ decision-making processes for organic food products for Italian consumers. On the contrary, Padel and Foster (2005), in their UK study, underline that the decisionmaking process about organic food is complex, and the importance of motivations and barriers may differ between categories. They demonstrate a need to differentiate diverse 59 reasons to buy organic food according to product categories, and provide insights into trade-offs that consumers make between competing and conflicting values and needs. Bonti-Ankomah and Yiridoe (2006) developed a conceptual framework which affected organic attitudes and purchase decisions, based on reviewing a large amount of literature. As depicted in Figure 2.4, this suggests that knowledge and awareness have a direct and indirect effect on attitudes towards organic products and the willingness to pay a premium, and are critical in the consumer purchasing decision process since organic food is a certified product. Consumers cannot identify whether a product is produced using organic or conventional methods unless they have been told or can see organic logos or labels attached to the products. Several studies (Chang & Zepeda 2005; de Magistris & Gracia 2008; Radman 2005; Tanner & Kast 2003; Zanoli & Naspetti 2002) have found that available information in the market creates higher organic food knowledge and positively influences attitudes towards organic food products. Consumers who are more knowledgeable about organic food are more tolerant of higher prices and inaccessibility and are more likely to purchase organic foods. On the other hand, knowledge and awareness may not directly translate into purchase, as barriers might limit the ability of consumers to transform such knowledge and perceived demand into actual demand (Bonti-Ankomah and Yiridoe 2006). 60 Figure 2.4 Framework of factors which affect organic consumers’ attitudes and purchase decisions Source: Bonti-Ankomah and Yiridoe 2006, p. 18 61 2.7.2 Attitude theory and measurement Grønhøj (2006) suggests that the tendency of using individualist approaches is widespread in the growing environmentally oriented study of organic products. Therefore, predicting the individual consumer’s behaviour towards buying organic food is often used in the attitude or value based model evidenced in the Theory of Planned Behaviour model. Attitude information and decision-making is more complex and closely related to personal values. 2.7.2.1 Introduction to the Theory of Planned Behaviour (TPB) The Theory of Planned Behaviour (TPB) has been widely used as a model of social cognition in psychology (or social behaviour). It is an extension of the Theory of Reasoned Action (TRA) but differs from this by its addition of perceived behavioural control (Ajzen 1991). Psychologist and marketing researchers have used both TRA and TPB to explain consumers’ food choice behaviour (Chen 2007; Eves & Cheng 2007; Petrovici, Ritson & Ness 2004; Verdurme, Gellynck & Viaene 2002; Verdurme & Viaene 2003). These models have also been used extensively in applied food decisionmaking, food consumption attitudes and predicting food choices, as they are good predictors of food consumption behaviour as well as they explain the cognitive processes that determine consumer behaviour. According to Vindigni, Janssen and Jager (2002), they explain the art of the process that determines consumer behaviour. The TRA was first developed by Fishbein and Ajzen (Ajzen & Fishbein 1980; Fishbein & Ajzen 1975) and was introduced in 1967 as Fishbein’s multi-attribute attitudes model (Blackwell, Miniard & Engel 2006; Fishbein 1967). It sets out a method of explaining and exploring the underlying determinants of intentions and behaviour (Fishbein & Ajzen 1975). The TRA is one of the most developed ‘knowledge-attitudes-behaviour’ (KAB) models and is extensively used in social psychology and consumer decisionmaking. It also has been applied to a number of health and environmental behaviour studies (Donovan & Henley 2003). It suggests that a person’s behaviour is determined by his or her intention to perform the behaviour, and states that ‘a person’s intention is a function of two basic determinants, one personal in nature and other reflecting social influences’ (Ajzen & Fishbein 1980, p. 6). These are termed normative beliefs. The 62 subjective norm is a measure of the perceptions held by the person and may not reflect the importance of what others actually think they should do. It is defined as an individual’s perception of behaviour influenced by peers and family (Ajzen & Fishbein 1980; Donovan & Henley 2003). Fishbein and Ajzen (1975, p. 14) state that ‘The totality of a person’s beliefs serves as the informational base that ultimately determines one’s attitudes, intentions and behaviours’. According to these authors it is necessary to distinguish between beliefs, attitudes, intentions and behaviour. Over the years, the TPB has been developed, tested and refined. Ajzen (1991, p. 206) states it ‘traces attitudes, subjective norms, and perceived behaviour control to an underlying foundation of beliefs about the behaviour’. According to the Planned Behaviour model (see Figure 2.5), intention is the best predictor of behaviour and thus behaviour can be deliberately planned. There are three major determinants embraced in the TPB: attitudes towards the behaviour, subjective norms and perceptions of behavioural control (Ajzen 1991; Ajzen 2005; Ajzen & Albarracin 2007). Demographic variables become external in social psychological studies because they are external to the cognitive structure associated with making a specific decision (Ajzen 2005). Figure 2.5 illustrates the role of background factors in the TPB. It describes a multitude of variables which may be related to or influence the beliefs consumers embrace. These background factors are divided into three categories: dispositions, demographics and information. The effects of background factors are assumed to indirectly influence intention and behaviour by affecting behavioural, normative and control beliefs (Ajzen & Albarracin 2007). 63 Source: Ajzen & Albarracin 2007, p. 6 Figure 2.5 The Theory of Planned Behaviour 64 The TPB has been successfully applied to explain aspects of consumer behaviour (Bredahl, Grunert & Frewer 1998). It offers an excellent starting point for the examination of organic food purchase and consumption (Shepherd, Magnusson & Sjödén 2005). Ajzen and Albarracin (2007) as well as Blackwell, Miniard and Engel (2006) suggest that the TRA model has some limitations. These include a significant risk of confounding attitudes and norms, since attitudes can be reframed as norms and vice versa. A further study (Schütte & Ciarlante 1998) suggests that the TRA is not valid in the Asian context, particularly in those cultures influenced by Confucianism, where the purchase intention is more influenced by the face-saving pressures and social influences of group conformity. Attitudes towards consuming a product have been found to be one of the most important antecedents of predicting and explaining consumers’ choice of food (Bredahl 2001; Conner et al. 2003). This was evidenced by an earlier UK study (Bredahl, Grunert & Frewer 1998) which used two models to explain consumer attitudes and decision-making towards genetically modified food: (1) Fishbein’s multi-attribute attitude model explained consumer attitudes towards GM food, which are determined by both consumers’ beliefs regarding the product process and their beliefs regarding products, and (2) consumer purchase behaviour with regard to GM food is explained by Ajzen’s TPB. The purchase (or avoidance) intention is summarised in the attitude to behaviour, subjective norm, perceived control, perceived difficulty and perceived moral obligation similar to that described in the original TPB model as shown in Figure 2.5. 2.7.2.2 Applicability of attitude theories in the context of organic food Previous studies (see Table 2.1) suggest that the Theory of Planned Behaviour can be applied to food choices, and this view is supported by the efficacy of its application towards organic food choice and intention. French et al. (in Hansen 2005) claim that food consumption is not just an issue of obtaining nutrients, but is also psychologically related. Maslow’s Theory of Motivation explains why people are driven by particular needs at a particular time (in Kotler et al. 2001). Food consumption can satisfy physiological needs, such as hunger and thirst, and also meets other higher level of needs. 65 A study in the UK (Lodorfos & Dennis 2008) uses the TPB as the conceptual framework to examine consumers’ intention to buy organic food. The findings offer considerable support for the model in explaining organic food consumption intentions. This study also suggests that organic food product attributes and the subjective opinions of others are important determinants of consumers’ intention. The result echoes previous studies (Kalafatis et al. 1999; McEachern & Willock 2004), which suggests that the TPB represents a reliable predictive model of intention to purchase environmental products. A study in Taiwan (Chen 2007) examined consumer attitudes and purchase intention in relation to organic food, and adopted the TPB. This study showed that the model explained quite effectively the results of the main effects of Taiwan consumers’ organic food intentions. Consumers’ intention in purchasing organic food is supported by their attitude to organic food purchase, subjective norms, perceived behavioural control and perceived difficulty. Furthermore, it proposed six food motive determinants which contribute towards consumers’ positive attitudes towards organic food: consumer mood, natural content, animal welfare, environmental protection, political values and religion. Inconvenience in purchasing organic foods negatively affected attitudes towards purchase. This study was one of the first attempts to examine the moderating effects of intention and attitudes on the purchase of organic food in the TPB model. Smith and Paladino’s (2010) Australian study utilised the TPB to form a causal model which established the relationship between organic knowledge, subjective norms and familiarity, revealing that intentions were mediated by attitudes towards organic knowledge, subjective norms and environmental concerns. This study has contributed to the growing organic food research and provides theoretical implications in an Australian context. In addition, an Indian study applied the TPB to examine both regular and occasional organic food buyers’ motivations and attitudes, showing that Indian consumers’ intention to buy organic food is relatively strong as they have positive attitudes towards organic food (Chakrabarti & Baisya 2007). Consumers across the board with positive attitudes towards organic food are more likely to form intentions to purchase it (Honkanen, Verplanken & Olsen 2006; Saba & Messina 2003; Sparks & Shepherd 1992). 66 2.7.3 Integration of three models: CDP, TPB and Hierarchy of Effects 2.7.3.1 Limitation of organic food studies The TPB argues that behaviour, such as choice of organic food, is predicted by intention, which is in turn is predicted by attitudes, perceived social pressure and degree of personal control (Ajzen 1991). Criticism related to the TRA and the TPB suggests that these theories can only partly explain food consumption (Petrovici, Ritson & Ness 2004). They fail to effectively incorporate some key features that are important in the choice of organic food. There are affective responses to food and moral concerns about the way in which it is produced (Shepherd, Magnusson & Sjödén 2005). They also only explain parts of the processes that determine consumer behaviour (Vindigni, Janssen & Jager 2002). Armitage, Conner and Norman (1999) address some of the criticisms, and argue that the models are too reliant on rational decision-making and the influence of individualistic behaviour. The TRA and TPB are multi-attribute attitude measurements which emphasise consumers’ beliefs, feelings, attitudes and intentions. However, they are not the only important beliefs in terms of how consumers behave. Others range from what consumers believe about future economic conditions to what they believe about the trustworthiness of alternative information sources (Blackwell, Miniard & Engel 2006): ‘For a complete insight into attitude formation, the multi-attribute model should incorporate both the importance of the evaluative criteria as well as their evaluation’ (Steenkamp 1997, p. 151). Criticism has also been raised about the CDP model in that it may be a generalised model applied to the purchase of any product or service. Various authors have focused on the sets of variables which have impacted on food choices (Bareham 1995). An Italian study (de Magistris & Gracia 2008, p. 931) suggests that ‘consumers’ attitudes towards different organic food attributes and towards the environment are the most important factors that explain consumers’ decision-making process for organic food products’. The CDP model fails to explore how beliefs and attitudes influence purchase decision made in advance of product evaluation (Shaw & Clarke 1999). 67 Ajzen (2001) suggests that attitudes contain both cognitive (thinking) and affective (feeling) components. The multi-components view of attitudes assumes that evaluations are influenced by cognition as well as affect. This is supported by Dean et al. (cited in Aertsens 2009) who claim that both cognition and affect are relevant to predict purchase intention of organic food in a compensatory way. Consumers evaluate behaviour not only in terms of cost and benefits, but also by the positive and negative feelings generated. These findings highlight the importance of including both cognitive and affective components of attitudes in food purchase intention models. The current study has identified the importance of integrating the three models: CDP, TPB and Hierarchy of Effects. 2.7.3.2 Integrating the theories Need recognition/information search The CDP model (Blackwell, Miniard & Engel 2006) suggests that no-one buys a product unless they have a need or a want. Barriers that significantly influence organic food purchase are lack of trust in the control mechanisms, lack of interest in organic foods and lack of trust that the food is healthy. Hence teasing out the most important factor in influencing consumers in their decision to purchase or consume organic foods can be a critical issue in the CDP model. From a marketing perspective, it is important to understand why consumers consume a certain quantity of organic food, their changing consumption patterns and their motives (Vindigni, Janssen & Jager 2002). The phase of need recognition arises when discrepancy exists between the consumer’s current circumstance and the situation they want to be in. On the other hand, need recognition does not happen if the difference is below the threshold level (Blackwell, Miniard & Engel 2006). Previous studies identify concerns for personal health, food safety and environmental issues which seem to be the most commonly stated triggers to motivate purchase of organic food products. Consumers may desire to get quality food, since they have heard of or purchased organic food for a while. According to 68 Ajzen and Fishbein (1980), intentions are assumed to capture the motivational factors that influence behaviour and are indications of how people would try and how much effort they would make to purchase organic foods. Search for information is the most important factor related to the previous purchase experience. How knowledgeable are consumers? A study based on food with therapeutic claims suggests that most consumers read nutritional facts and then, based on this information, make their decision (Bhaskaran & Hardley 2002). Although they do not trust claims made by the producers, and may not believe that the product would help with an existing health problem, their purchase decisions are based on the hope that the product has a therapeutic feature. Pre-purchase evaluation Steenkamp (1997) points out in relation to the CDP model that studies from seven European countries investigating the importance of a large set of evaluation criteria for the choice of generic food products found that there were five important criteria: product quality, price, brand name, freshness and guarantee. Guarantee was implied based on brand name, quality labelling and geographic origin which, in turn, allowed for consumers to easily process this information. However, the criteria used by consumers in the evaluation of food choice depended on the type of food products involved. For instance, taste intensity, taste evaluation and fatness were key criteria for meat products. An Australian study of consumer perception and demand for organic food (Chang & Zepeda 2005, p. 164) suggests that ‘consumers preferences and food choices appear to be influenced by socio-demographic factors such as income, family structure, lifestyle, dietary restrictions and some social values and beliefs’. Larger studies (Chen 2007; Wong 2004) show that the correlation between organic food purchasing decisions associated with people’s attitudes is positive and significant. In the TPB, consumers with positive attitudes are more likely to display purchase intention behaviours. On the contrary, those with negative attitudes towards organic food will be inclined not to purchase. Barriers towards purchase of organic foods can be investigated in this phase as indicated in the TPB and the 69 model of Hierarchy of Effects. For example, price may elicit both positive and negative attitudes. It is positive when a higher price of organic food denotes a symbol of quality, and negative when it signifies burden of purchase. At this stage, consumers decide whether they will purchase or not. They evaluate product attributes, brands, personal values and lifestyle. It is important to recognise that ‘attitudes influence consumers’ purchase and consumption intentions’ (Blackwell, Miniard & Engel 2006, p. 392). Consumers are reluctant to pay higher prices for organic food even if they show positive attitudes towards them. Such attitudes do not necessarily translate into a final decision to purchase. 2.8 Summary of previous studies Consumers perceive that organic food has a healthier image attached to it. They believe it is better for the environment, promotes animal welfare and has a better taste. It has a more trustworthy quality and consumers are willing to pay more for it. On the other hand, some of the reasons consumers have indicated for not purchasing organic food include perceptions of it being expensive and limited in availability and choices (Bhaskaran et al. 2006; Chang, Zepeda & Griffith 2005; Connor & Douglas 2001; Lea & Worsley 2005; Radman 2005; Sanjuán et al. 2003). Non-purchase is also influenced by unsatisfactory appearance, low profile distribution channels, misleading labelling and certification, and lack of perceived value (Brennan, Gallagher & McEachern 2003; Chang, Zepeda & Griffith 2005; Essoussi & Zahaf 2008; Krystallis, Fotopoulos & Zotos 2006; Stobbelaar et al. 2007). In recent years, studies of consumers’ purchase behaviour related to organic food have been carried out in many countries, including the US, UK, Netherlands, Ireland, Germany, Denmark, Sweden and Italy. They have explored consumer characteristics, awareness, knowledge, attitudes, values and even risk perceptions. Various approaches with the aim of describing, modelling and predicting consumers decision-making process in purchasing organic foods have been investigated (Vindigni, Janssen & Jager 2002). 70 Studies in Asian countries such as Japan, Taiwan, India and Thailand confirm that the most important reasons and motivation for consumers to purchase organic food products are health, food safety and concern for lifestyle. Furthermore, these studies examined whether consumers are willing or not to buy organic food. What is evident from this chapter is the inconclusive nature of the literature on this topic. The review has uncovered related extant literature on a global and regional basis and highlights important yet partial findings. There is an extensive and well developed European literature on the topic. Curiously, there was significant literature in countries such as Greece, yet much less in countries such as Germany. Some of this imbalance might seem in part due to language accessibility. The literature in the Americas was also extensive as it was in other localities such as in Australia. The key authors from selected studies conducted in different continents and countries are summarised in Appendix 2. 2.9 Limitation of previous studies Various studies relating to consumers purchase’ behaviour towards organic products have been carried out in Europe and internationally (Krystallis, Fotopoulos & Zotos 2006). Only a handful of studies relating to organic food consumption have been conducted in Asian countries such as Japan, Taiwan, India and Thailand (Chakrabarti & Baisya 2007; Chen 2007; Roitner-Schobesberger et al. 2008). These reveal that purchase behaviour in Asian countries is different from that in the west. Issues such as environment, for example, are not a major motivation for Indian organic buyers (Chakrabarti & Baisya 2007). Furthermore, very few studies have been conducted in China (Dai, Zhu & Ying 2006; Yin et al. 2010). Some studies in the Chinese language (including Dai, Zhu & Ying 2006) examined consumers in Nanjing on their propensity to purchase organic vegetable products, and their purchase behaviour based on age, education, knowledge of organic vegetables and concern about food safety and environmental protection issues. A similar study (Wang, liu & Tian 2008) in Beijing revealed that the older the consumer, the greater the intention to purchase organic food. This was due to their own health concerns about food. On the other hand, the greater the educational levels, the greater the levels of scepticism about organic 71 food production and certification. These two studies focused mostly on demographic factors influencing Chinese consumers and the data was collected in a single city. Yin, Wu and Chen (2008) investigated consumers in Shandong province and highlighted the greater level of confidence which consumers had in organic food and in the certification process. The study also found that age, education level and environmental issues are not so important to Chinese consumers’ organic food purchase behaviour. In a more recent study, Yin et al. (2010) claim that none of the previous studies had analysed factors that affect consumers’ choice of organic food in the mainland Chinese market. This survey of 432 consumers was conducted in three southern cities: Shenzhen, Guangzhou and Zhuhai. The results concur with Yin, Wu and Chen (2008) that Chinese consumers’ intention to purchase organic food is strongly influenced by income level, health, degree of trust in organic food and degree of acceptance of organic food price (Yin et al. 2010). A weakness with this study is that the survey data was conducted in the same region, the Zhujiang Triangle Zone of Guangdong province in southern China. China constitutes a heterogeneous consumer market, with distinct differences in buyer behaviour in inland and coastal cities, and many disparities across north, south, east and west (Cui & Liu 2000; Dou, Wang & Zhou 2006; Sun & Collins 2006; Uncles & Kwok 2008; Zhou et al. 2010). Therefore this study fails to provide a consistent overview of the entire Chinese market. The most recent study by Nees (2011) faced a similar problem. Her study only collected data from Guangzhou over a period of three days and only focuses on the purchase of organic vegetables. However, the range of organic food products in the Chinese market now not only includes fruit, vegetables and meat, but also extends to rice, green tea, dairy produce and other processed food (International Trade Centre 2011). There is scant empirical research on consumer behaviour towards organic food across China, and the market suffers from significant misunderstanding of the terms ‘organic’ and ‘green’ types of food which are used interchangeably by consumers. Due to some significant omissions in previous studies, it is the intention of this study to empirically investigate the determinants related to Chinese consumers’ pre-purchase evaluation of organic food in a way which fills the void left by numerous but incomplete studies. 72 2.10 Chapter summary This chapter has examined previous organic food studies across numerous years and countries. The research undertaken is both significant and in part inconclusive. This literature review has sought to investigate and define the most significant findings of previous studies on organic food and consumer behaviour towards it. The chapter began by investigating the stereotype and myths around organic food. It examined the importance in the literature of demographic and other profile factors which may have played a role in understanding consumer behaviour towards organic food across the globe. Consumer attitudes towards organic food and decision-making in choosing organic food have been well documented across much of the developed world, but only a handful of studies have been conducted in Asian countries. These studies, especially in western markets, are important so as to better understand the issues and whether there are any patterns or trends in consumer behaviour which may have been pertinent for the Chinese context within this study. Consumer purchase of organic food might be dependent on the perception that it is safer, healthier and more environmental friendly than conventional food. There are several important factors and motivations why consumers are motivated to purchase organic food products, and barriers to non-purchase. This chapter has identified that the Consumer Decision Process model, the Hierarchy of Effects and the Theory of Planned Behaviour are widely utilised within food and organic food studies. They have provided significant theoretical underpinning for research in the areas of consumer buying behaviour. These theories analyse consumers’ attitudes, purchasing intentions and behaviour towards organic products, and examine various determinants that influence organic food consumers to make purchase decisions. Chapter Three, which follows, develops an applicable conceptual framework that incorporates the Consumer Decision Process in choice of consumption of organic food in China. The conceptual framework developed is based on integration of these theories and also based on the assessment of consumer perception, belief, attitude, behaviour and other related issues towards organic food. 73 CHAPTER THREE: DEVELOPMENT OF THE CONCEPTUAL FRAMEWORK AND RELATED HYPOTHESES 3.1 Introduction Chapter Two evaluated related extant literature including the three key models of consumer purchase behaviour: Consumer Decision (making) Process (CDP), Hierarchy of Effects and the Theory of Planned Behaviour (TPB). These are central to most studies in the food and organic products area. This chapter aims to formalise the conceptual framework (Section 3.2). It discusses the influencing stage (Section 3.3), cognitive/affective stage (Section 3.4), evaluation of alternatives stage (Section 3.5) and behavioural intentions stage (Section 3.6). In addition, demographic influences are analysed as control variables (Section 3.7) and a chapter summary is provided (Section 3.8). The hypotheses are developed as result of the interrelationship between the different constructs of the conceptual framework. Figure 3.1 diagrammatically illustrates the structure of Chapter Three. 74 Figure 3.1 Structure of Chapter Three Source: developed for this research 3.2 Development of the proposed conceptual framework As identified by Cavana, Delahaye and Sekaran (2001), literature reviews play an important role in identifying and scoping models which can then become the basis for a new theoretical framework. The proposed conceptual framework is developed on an extensive review of the relevant body of literature and is designed to address the research questions. Chapter Two explored the key characteristics of three models: the CDP, Hierarchy of Effects and TPB. These have been widely utilised within studies of food consumption behaviour (Drichoutis, Lazaridis & Nayga Jr 2007; Tanner & Kast 2003; Verbeke 2000) as 75 well as more specific organic food studies (Chen 2007; Essoussi & Zahaf 2008; de Magistris & Gracia 2008; Verdurme, Gellynck & Viaene 2002). Despite their importance, the proposed conceptual framework for the current research is unique and incorporates the idea developed by Verbeke (1999, 2000). For the purposes of explaining consumer purchase of organic food in urban China, this study has established that the CDP model is far too general and lacks specificity for examining specific indicators influencing consumer purchase behaviour. A five stage conceptual framework for analysing consumer decision-making towards organic food in urban China is proposed (Figure 3.2). This is embedded in the three underpinning models mentioned above. Essentially it is a staged model which assumes that consumers move through a rational problem solving process in ultimately purchasing organic food. The five stages are influencing, cognitive/affective, evaluation of alternatives, behavioural intentions and purchase. The problem solving process in the minds of consumers (cognitive/affective stage) involves the search and processing of information as suggested by the CDP model and also by the TPB (beliefs, attitudes and norms). The Hierarchy of Effects model (Barry & Howard 1990) is in line with this thinking as it advocates the different mental stages of cognitive, affective and conative stages that consumers go through when making buying decisions. Pilgrim (1957) and Steenkemp (1997) suggest a classification of factors or variables that potentially influence consumer decision-making. This forms the first stage of the proposed conceptual framework. As a result of the extensive literature review of organic food studies in Chapter Two, it was decided to include other relevant constructs which influence consumer decision-making towards the purchase of organic food in urban China, that is, product, regulatory, lifestyle and ethnocentrism. The product construct includes physical appearance, smell and taste, which are part of sensory appeal. This construct also covers product nutritional value and price perception. The regulatory construct includes items related to certification and labelling, government regulation and policy, and also environmental influences as suggested by Steenkamp (1997). The lifestyle construct includes three types of lifestyles: self-indulgence, variety seeking and opinion leadership. 76 Finally, the ethnocentrism construct includes aspects of country of origin and the level of ethnocentrism. The attitudes and beliefs of consumers which form the second stage of the proposed conceptual framework are akin to the need recognition and search for information stages of the CDP model. It involves consumers’ awareness and understanding of the product, and is embedded in the cognitive and affective domains of the Hierarchy of Effects model. This model suggests that three mental stages – cognitive, affective and conative – are involved when consumers make purchase decisions. The cognitive stage is the mental activity as reflected in knowledge and beliefs, the affective stage is related to feelings and emotions, and the conative stage refers to purchasing intention or behaviour (Barry & Howard 1990). Essentially, attitudes are consumers’ feelings or affective responses (Verbeke 1999, 2000). The third stage of the conceptual framework addresses consumers’ pre-purchase evaluation. This stage is exactly the same as that proposed by the CDP model. Consumers seek the best options when they evaluate and select various products (Blackwell, Miniard & Engel 2006). The prospective purchaser weighs up the pros and cons of each product, which naturally depends on personal judgement. Evaluative criteria are derived from beliefs, attitudes and intentions (Bareham 1995). At this stage, urban Chinese consumers would compare what they know about different safe and healthy foods in the Chinese market with what they consider to be the most important attributes. They could use new or pre-existing knowledge stored in memory to select organic food products that would most likely result in their overall satisfaction. Blackwell, Miniard and Engel (2006) suggest that pre-purchase evaluation is influenced by both individual and environmental influences. The fourth stage of the conceptual framework is related to behavioural and purchase intention. This stage is suggested by the TPB, which then leads to actual purchase. As a rule of thumb, the more favourable the attitude, subjective norm and perceived control, the stronger should be the person’s intention to perform the behaviour (Ajzen 1991). Finally, the fifth stage is the actual purchase, which is what marketers are really interested in and is articulated in all the three associated theories of CDP, TPB and Hierarchy of 77 Effects. Most studies relating to consumer buyer behaviour only investigate their behavioural or purchase intention as the actual purchase involves collection of data from the sample at some future time (longitudinal studies). Hence, this stage is depicted by dotted lines in the conceptual framework and was not investigated in this research. Generally, positive attitudes are likely to lead to positive purchasing intentions, which may eventually translate to actual purchase (Tarkiainen & Sundqvist 2005). The proposed conceptual framework as shown in Figure 3.2 is developed to illustrate and explain Chinese consumers’ purchase behaviour and includes key determinants which influence their decision-making process for the purchase of organic food. Details of the individual stages of this conceptual framework and the development of hypotheses are explained from Section 3.3 onwards. 78 79 3.3 Influencing stage The term ‘influencing’ was adapted by Verbeke (1999, 2000) based on one of the earliest models of consumer buyer behaviour of food (Pilgrim 1957) and also on the influencing factors suggested by Steenkamp (1997) who asserts that there are three types of influencing factors on consumers: environmental factors, personal related factors and properties of the food. Based on this idea and on factors identified in the literature related to organic food consumption as discussed in previous chapters, four significant constructs are elicited: product, regulatory, lifestyle and ethnocentrism, as discussed in the sections that follow. 3.3.1 The product construct More than 40 years ago, Lancaster published his now famous product characteristic theory which reflects the importance of product characteristic or attributes (Drichoutis, Lazaridis, & Nayga Jr 2007). This construct is categorised as ‘properties of food’ (Steenkamp 1997; Verbeke 1999). It includes two key factors: sensory appeal and price perception. Product attributes of food vary in intensity, favourability and the degree of conviction which individuals have (Ophuis et al., cited in Chryssochoidis 2000). Pilgrim (1957, p. 171) suggests that the criterion of food acceptance may be described as ‘consumption with pleasure, the nutrition of body and soul’. Although current studies suggest little consistency across countries, they are nonetheless similar in terms of consumers’ perceptions towards organic product attributes (Bonti-Ankomah & Yiridoe 2006). Consumers consider sensory characteristics to be one of the key factors in the choice of organic food (Magnusson et al. 2001; Magnusson et al. 2003; Torjusen et al. 2001). For most, organic food purchase motivations arise largely from product-specific characteristics directly benefiting the consumers (Wier & Calverley 2002). Obviously, taste is one of the important elements which influence consumers’ choice of organic food, followed by appearance and price (Stobbelaar et al. 2007). However, non-sensory 80 attributes such as absence of food additives and residuals and nutritional value (Torjusen et al. 2001) are increasingly important. Steptoe, Pollard, and Wardle (1995) found that concern for health, convenience of food preparation and sensory appeal, price and convenience are mechanisms in the selection of food at the individual level. The statements related to appearance, smell and taste contribute towards sensory appeal. Food attributes such as price, taste and nutrition are referred to as measures of product involved factors when making food purchasing decision. Accordingly, different consumers can be associated with different food attributes (Drichoutis, Lazaridis & Nayga Jr 2007; Steptoe, Pollard & Wardle 1995). A Greek study reveals that organic food purchase is motivated by consumers’ sensory appeal and concern for health (Arvanitoyannis, Krystallis & Kapirti 2003), and a New Zealand study (Wong 2004) discloses that participants considered the taste of organic food to be better and to possess superior flavour as compared to conventional food. However, organic food is not always aesthetically appealing due to its shorter shelf life. Its physical appearance is sometimes considered as a hindering element to its consumption (Roddy, Cowan & Hutchinson 1996). A Belgian study (Verdurme, Gellynck & Viaene 2002) found that consumers of organic food regarded organic food as being, in order of importance, ‘healthy’, ‘tasty’ and ‘environmentally friendly’. Another interesting Danish study by Millock, Wier and Andersen (2004) suggests that attitudes towards taste, freshness and aspects of health associated with organic food are more important drivers than the environment and animal welfare attributes in its actual purchase. In contrast, a Greek study by Chryssochoidis (2000) questions whether organic food attributes such as physical appearance do actually influence the purchase intention decision. It indicates that physical appearance can in fact be non-appealing to some consumers, which in turn impacts their purchase intention. As the standardised coefficient between physical appearance of organic food and purchase intention in this study was weak (0.08), the finding needs further investigation. 81 The price of organic food is controversial and may be associated with other variables such as gender and socio-economic positioning. Steptoe, Pollard and Wardle (1995) refer to cost as being an important issue for low income earners. It is also seen as more crucial amongst female rather than male consumers. However, this study observes that those female consumers concerned about dietary issues are less influenced by price. This is consistent with a Belgian study (Verdurme, Gellynck & Viaene 2002) which noted that existing consumers consider organic food as healthy, environmentally friendly and less expensive. On a different note, one reason for the non-purchase of organic food is perceptions of it as being expensive. Organic production means more work and lower yields; organic food is also overpriced by unscrupulous middlemen, according to Chang & Zepeda’s (2005) Australian study. However, an Italian study (Cicia, Giudice & Scarpa 2002) demonstrates that, in terms of consumers’ perception of quality in organic food, price is often perceived as a proxy for quality. In China, its price is perceived to be relatively higher in line with its perception of excellent quality and reliability (Lu 2008). Development of hypotheses associated with the product construct Based on the above discussion, the following relationships between the product construct and other stages of the proposed conceptual framework are hypothesised. The first hypothesis deals with the product construct and the second stage of the conceptual framework, which is the cognitive/affective stage, and is postulated as follows: H1a: The product-related construct has a positive influence on the beliefs and attitudes of urban Chinese consumers towards the purchase of organic food products The next hypothesis deals with the product construct and the third stage of the conceptual framework, which is pre-purchase evaluation. The importance of product attributes is supported by a Greek organic meat study (Krystallis, Arvanitoyannis & Chryssohoidis 2006) which suggests that pre-purchase evaluation involves visual judgement like colour, leanness and similar elements. Accordingly the following has been hypothesised: 82 H1b: Organic food product-related construct is positively correlated to urban Chinese consumers’ pre-purchase evaluation Finally, a hypothesis between the product construct and the fourth stage of the conceptual framework which is behavioural and purchase intention is elicited. Product-related attributes are vital to Chinese consumers. Historically, they place high expectations on the original flavour, taste and freshness of food products. Many Chinese, particularly the elderly, prefer to purchase fresh foods on a daily basis. The freshness, flavour and texture add value to the food products (Zhao et al. 2000). It is evident from the literature that the motivation of European consumers for buying organic food is product specific characteristics which directly benefit them. However, this motivation appears to vary between different consumer segments (Wier & Calverley 2002). A study in Taiwan (Chen 2007) acknowledges that the characteristics of specific food and the relevance that consumers assign to them have both a direct and an indirect effect on the intention to buy organic products. Based on the above discussion, the following hypothesis is formulated: H1c: Organic food product-related construct is positively correlated to urban Chinese consumers’ behavioural/purchase intention 3.3.2 The regulatory construct The regulatory construct relates to two areas: government regulation and the complex area of labelling. There is a relatively poor understanding of the legal use of the term ‘organic’ on the labelling of food products including the inspection and certification systems (Padel & Foster 2005). Some studies have highlighted the seemingly stringent regulatory process (Courville 2006; Wier & Calverley 2002). Others suggest the confusing nature of labelling and regulatory processes (Aarset et al. 2004), and still other studies strongly question the validity of the labelling and certification and highlight the distrust felt by consumers in relation to what they are told about organic food (Williams & Hammitt 2000). From a purely material and technical point of view, organic food certification ensures for consumers the reliability of the claims that the labels represent (Getz & Shreck 2006). 83 Consumers generally do not know whether a product is organically produced unless they have been told or have seen the evidence of certification or the label of the product concerned (Lin, Smith & Huang 2008). As Ennis (2007, p. 104) noted: ‘A credible good is one whose quality cannot be identified with consumption, but whose history affects some consumers’ attitudes towards the good’. Consumers generally avoid purchasing food products which do not have signs of the products’ quality. As a credible product, organic food quality can influence positive attitudes on consumers who have greater knowledge and experience with the product. Food safety concerns appear as the most important influence on purchase decision-making. A European study highlighted the lack of organic product knowledge and confusion among European Union consumers. Many were unsure, even dubious, about the concept of organically farmed salmon and were distrustful of the regulatory process (Aarset et al. 2004). Labelling seems to be one of the important issues in relation to the purchase of organic food. Consumers perceive organic food labels as providing a guarantee of the product (Magnusson et al. 2001; Torjusen et al. 2001). Despite the general awareness of organic food labels in Greece, almost half of the buyers were confused by the differentiation between organic and conventional food products (Krystallis, Fotopoulos & Zotos 2006). Not surprisingly, organic food consumers are more likely to rate food labels as more important than conventional food consumers, but ‘organic’ or ‘pesticide-free’ labels can be misleading (Williams & Hammitt 2000). There is also a difficulty for consumers awareness due to an absence of a well-known organic brand, unlike conventional products where brand recognition is well understood. As such there is a lack of variety of branded organic products. Organic labelling maximises the effectiveness of organic products which thus reveals a direct relationship between price and brand awareness. There are also many organic certification bodies in the market, and it is debatable as to which certification stamps or logos are genuine and can be trusted by consumers. For example, in Taiwan, there are more than 14 kinds of certificates and labels (Chen 2007), and there are 27 domestic and six overseas (including US, Europe, Australia and Japan) organic certification bodies for the Chinese market (Yuan & Liu 2009). Euromonitor 84 International (2010) reports that many Chinese consumers do not believe the descriptions on the packaging and have no confidence in the manufacturers’ claims. This report confirms Taylor’s (2008) findings regarding Chinese consumers’ distrust of labelling of organic food. The regulatory regime associated with organic food can be a compelling factor across numerous markets. For example, Denmark has the highest consumption of organic food per capita in the world, and consumers are not unduly concerned about related health and food safety issues. This is because Denmark has a relatively well functioning processing and distribution system in the organic food industry and also boasts a reliable certification and labelling system. Organic food is offered in a variety of forms and is available through mass retailing and supermarkets (Wier & Calverley 2002). On the other hand, perceptions about government regulations can attract ongoing feelings of mistrust towards the government in relation to organic food and the environment. A study undertaken in Croatia shows that only 11% of respondents thought that organic agriculture is properly regulated by law, 61% did not agree, and more than 25% did not know anything about organic agriculture legislation. This is further evidence of the low awareness of organic systems and mistrust of organic products regulation (Radman 2005). The recent food scandals in China which involved melamine-tainted dairy products have increased consumers’ concerns about domestically produced food products and, according to US sources, have led them to doubt the food regulatory system (US Department of Agriculture 2008). A survey (Euromonitor International 2011) conducted by the Beijing Times during 2011 found that 70% of consumers would not purchase milk powder produced in China. The main concern for many is that toxic substances are illegally added to milk products to boost their apparent protein content. Chinese consumers prefer to buy imported products with foreign organic certification as this is viewed as being more trustworthy. The government has sought to restore consumer confidence by introducing stricter organic rules and regulations (International Trade Centre 2011). 85 A recent study (Xie, Li & Yi 2011) indicates that China has had voluntary national organic standards since 2005, and attempts have been made to provide protection for consumers against deceptive and misleading claims on organic products and to strengthen the organic industry’s capacity to respond to the domestic market. Lack of equivalency and non-harmonisation of standards and regulation are some of the impediments for trading in the Chinese organic food market. Development of hypotheses associated with the regulatory construct Based on the above discussion, the following relationship between the regulatory construct and attitudes of urban Chinese consumers is hypothesised: H2a:The regulatory construct associated with organic foods has a positive influence on beliefs and attitudes of urban Chinese consumers A US study (Bellows et al. 2008) suggests that the use of organic labelling information normally increases the organic label familiarity. A certification stamp or logo provides a guarantee to the consumer that the product has been produced organically (McDonald 2001). Some food industry stakeholders argue that labels are often confusing and open to misinterpretation by the public. However, if a product is certified as organic by an accredited organisation and its logo is displayed on the label, it is a guarantee to consumers that it is the closest thing to what nature intended (Abbey 2005). Lea and Worsley’s (2005) Australian study concedes that about half of the participants did not find organic food labelling as being trustworthy. Other studies (Aarset et al. 2004; Hughner et al. 2007) further contend that there is lack of trust in the organic product certification process. With the increasing affluence derived from rapid economic growth, certain sections of Chinese society and also the Chinese government have begun to realise the hazardous impact of environmental deterioration on national wellbeing (Chan 2001). It is the government’s desire to implement anti-pollution laws with stronger punishment for offences against the environment (Chan 2001). The concern for the environment by consumers is also evidenced in developed markets and green behaviour, and perceived 86 consumer effectiveness can positively affect the choice of organic food (Verhoef 2005). In light of the above discussion, the following relationship between regulations and prepurchase evaluation is hypothesised: H2b: The regulatory construct associated with organic foods is positively correlated to urban Chinese consumers’ pre-purchase evaluation The regulatory construct has an impact on organic food safety. An Irish study (O’Donovan & McCarthy 2002) reports that respondents who purchased or had intention to purchase organic meat placed higher levels of importance on food safety, compared to those with no such intention. A US study (Williams & Hammitt 2000) found that organic food buyers value safe food. They do not trust conventional food supply and believe that the federal food safety agencies are not trustworthy. Organic meat consumers believe that it is superior in terms of quality, safety, labelling, production methods and value (O’Donovan & McCarthy 2002). With these points in mind, the following hypothesis is developed: H2c: The regulatory construct associated with organic food is positively correlated to urban Chinese consumers’ behavioural/purchase intention 3.3.3 The lifestyle construct Lifestyle segmentation is often used to predict consumer behaviour. The concept of lifestyle was introduced in marketing research from the psychology area in 1964 by William Lazer who described it as ‘a distinctive or characteristic mode of living, in its aggregative and broadest sense, of a whole society or segment thereof’’(cited in Zhu et al. 2009, p. 300). To further elaborate its importance, Kotler et al. (2001, p. 208) provide another definition. They say that a person’s lifestyle profiles their whole pattern of acting and interacting in the world and can help the marketer to understand changing consumer values and how they affect purchase behaviour’. Brunsø and Grunert (1995, p. 475) define food-related lifestyle as ‘the system of cognitive categories, scripts and their associations, which relate a set of products to a set of values’. Lifestyle has a multidimensional basis for explaining consumer behaviour (Reid et al. 2001) and has been 87 widely utilised in marketing management for designing advertising strategy, segmentation of populations, new product development and market surveillance (Brunsø & Grunert 1995). A study conducted in the Netherlands suggests that the consumption of organic food is part of lifestyle of people who may have related interests regarding nature, society and the environment. According to this study: ‘Lifestyle results from an ideology connected to a particular value system that affects personality measures, attitudes and consumption behaviour’ (Schifferstein & Oude Ophuis 1998, p. 119). Health and environmental consciousness which are de facto lifestyle expressions have become some of the motivations for Dutch consumers to buy organic food. A US study based in Boston reinforces the findings that organic food consumers (compared to purchasers of conventional foods) saw this as a lifestyle choice (Williams & Hammitt 2000). For some consumers, hedonic consumption is linked to a search for sensorial pleasure, and organic food could well be one such example (Piron 2006). Chen (2009) supports previous findings that health consciousness and environmental attitudes towards lifestyle are the most common motives for purchasing organic food in Taiwan. It also shows that a healthy lifestyle mediates the relationship between health consciousness and environmental attitudes and the consumer’s attitude towards organic food. In Malaysia, increasing consumption of organic food resulted from changes in lifestyles owing to an awareness of food safety issues (Shaharudin et al. 2010b). On the contrary, Wier and Calverley (2002) suggest that changes in lifestyle can have adverse effects on the consumption of organic food. This is because life is becoming busier, and consumers may either frequently eat outside the home or prepare meals from convenience products. One Australian study depicts current stereotypical images of organic consumers as being ‘greenies’, ‘health nuts’ or ‘yuppies’, with consumption of organic food reflecting their ‘green’ lifestyle and interest in these products as an element of trend (Lockie et al. 2002). Organic foods are perceived in some cases as being trendy symbols, as suggested by a Thai study (Roitner-Schobesberger et al. 2008). For regular Italian organic consumers, health and wellbeing are the most transcendental principles combined with expressions of hedonism, pleasure and achievement (Zanoli & Naspetti 2002). A Greek study 88 (Chryssochoidis & Krystallis 2005) indicates that there is almost no difference between consumers who wish to enjoy life and those who are ‘health conscious’ when sourcing organic food. Yang (2004) categorises consumers into three distinct lifestyle groups on the basis of their lifestyle characteristics: experiencers, traditionalists and self-indulgents. Experiencers seek variety and novelty, and prefer trendy and fashionable products. Typical traditionalist shoppers conform to norms and are less experimental. Selfindulgents are risk takers and self-centred individuals. The goal of a ‘well-off’ society in China is to translate economic growth into a better quality of life. Hence there is optimism surrounding hyper-consumption from the perspective of good health and a long life (Zhang et al. 2008). The LOHAS (Lifestyle of Health and Sustainability) lifestyle which evolved in the US is seen as a trend for some consumers to follow (Essoussi & Zahaf 2008; Kristiansen et al. 2010). A lifestyle related study (Tai & Tam 1997) suggests that women in China consider imported ‘green’ products to be fashionable. Gil, Gracia & Sanchez (2000) introduced lifestyle factors to investigate Spanish consumers’ willingness to pay for organic products. They found that those whose lifestyle was based on healthy diets and who were concerned about environmental degradation were more likely to buy organic food. Spanish consumers who are aware about health, balanced lifestyle, natural diet and environmental issues showed positive attitudes towards organic food and were willing to pay higher prices (Sanjuán et al. 2003). However, some consumers who believed that organic products are merely fashion statements had negative attitudes towards their purchase (Gil, Gracia & Sanchez 2000). Development of hypotheses associated with the lifestyle construct Based on the above discussion, the following relationships between the lifestyle construct and the following three stages of the proposed conceptual framework are hypothesised: H3a: The lifestyle construct has a positive influence on the beliefs/attitudes of urban Chinese consumers in relation to the purchase of organic food products 89 H3b: The lifestyle construct is positively correlated to consumers’ pre-purchase evaluation of organic food products in urban China H3c: The lifestyle construct is positively correlated to consumers’ behavioural/purchase intention of organic food products in urban China 3.3.4 The ethnocentrism construct Wild et al. (2007, p. 59) propose that ‘ethnocentricity is the belief that one’s own ethnic group or culture is superior to that of others’. Shimp & Sharma (1987, p. 280) define consumer ethnocentrism as: Beliefs held by consumers about the appropriateness, indeed morality, of purchasing foreign made products. Consumer ethnocentrism gives the individual a sense of identity, feelings of belongingness, and an understanding of what purchase behaviour is acceptable or unacceptable to the in-group. In some societies this is translated as being nationalistic or patriotic. In the context of consumer behaviour it impacts in two ways: it defines one’s own country as being superior, and it views the products of others as being potentially inferior. Preferences for foreign products can ruin home produced goods and hurt the domestic economy. On the other hand, discrimination against imports from certain countries also creates invisible barriers to globalisation (Agbonifoh & Elimimian 1999; Wild et al. 2007). In a Greek study (Fotopoulos & Krystallis 2002b), consumers generally refrained from purchasing foreign organic products because of their ‘patriotic’ stance. The country of origin seems to have had a powerful influence on Greek consumers. They also seem to be indifferent towards advertising, and seek information about organic products from their family and friends. Generally Greek consumers displayed strong ethnocentric tendencies. Some consumers may be willing to purchase organic food to support local producers, hence they tend to prefer shipments from locally grown organic produce compared to products shipped from other areas (Bonti-Ankomah & Yiridoe 2006). Locally produced 90 goods involve short food miles, implying lower fuel costs. This concept motivates consumers who are concerned for the environment (Wirth, Stanton & Wiley 2011). A US study (Williams & Hammitt 2000) found that organic food buyers were twice as likely as conventional buyers to rate locally grown produce as an important factor influencing their purchase decision. Howard and Allen (2006) suggest that locally grown products are preferred by older consumers and by households with children. From ancient times, the Chinese people have valued patriotism. This is widely reflected among people of all ages and areas, and in turn influences attitudes towards foreign-made products. A recent study () indicates that Chinese consumers’ ethnocentrism tendencies play an important role in shaping their purchase attitudes towards domestic fresh fruit. In the last 20 years this patriotism has become somewhat diluted although there are signs of a return to higher level of ethnocentrism, especially among older and less educated people with less foreign contact and influence (Hsu & Nien 2008). Chinese consumers’ attitudes towards western luxury goods are generally positive owing to the fact that there are currently no viable Chinese alternatives (Lu 2008). The McKinsey survey (StMaurice, Süssmuth-Dyckerhoff & Tsai 2008) finds that as Chinese consumers become more sophisticated, nationalistic tendencies play a smaller role when they shop. However, Chinese nationalism surfaces in various way, particularly as the country increases its economic development and international visibility, such as hosting the 2008 Olympic Games. Kwok, Uncles and Huang (2006) suggest that Chinese consumers generally prefer local branded products with regards to fast moving consumer goods. There is a growing tension between consumers’ pride and their interest in imitating and integrating into the western way of life (Huliyeti, Marchesini & Canavari 2008). As organic food advocates environmental and local produce, ethnocentrism comprises items related to country of origin. Previous studies demonstrate that country of origin is an important issue for organic food consumers. Pilgrim (1957) indicates that it is related to preferences for many foods. Studies on country of origin conducted in developed countries have revealed that ethnocentric consumers tend to perceive domestic products as being of better quality than imported products. However, there is a variance in the level of ethnocentric tendencies measured across different type of products (Herche 1992). 91 Country of origin of organic food is important to most Danish consumers: 72% of respondents preferred to buy domestically produced conventional fruit and vegetables rather than foreign produced organic food; 80% had confidence in the quality of food produced in Denmark when they read the package description of where and how it was produced (Millock, Wier & Andersen 2004). Irish consumers prefer local, fresh, seasonal and ‘chemical-free’ fruit and food over imported certified organic products (Moore 2006). In developing countries, purchasing of foreign products is generally seen as a social symbol, and one associated with status searching. This is particularly the case with those products which originate from prestigious countries (Batra et al. 2000). However, Anic’s (2010) study in Croatia (a developing country) found that consumers displayed a relatively high level of ethnocentrism towards domestically made retail food products. Another study by Ozretic-Dosen, Skare and Krupka (2007) concurs that almost half of young Croatian respondents would purchase more expensive domestically produced chocolate rather than western European brands if the national brand was of similar quality. There are signs that after more than a decade of embracing all things Western, the Chinese are turning to things Chinese (Prasso, Ting & Dan 2007). A contrasting report portrays Chinese consumers as being no longer interested in foreign brands or products, while domestic products are seen to provide better value with good and matching quality at lower prices (Croll 2006). Watson and Wright (2000) suggest that cultural similarity is an important consideration for highly ethnocentric consumers in evaluation of foreign products. Another Chinese study (Liu et al. 2006) shows that the differing attitudes of consumers towards goods imported from western countries and from Japan. This has been attributed to the historical animosity between China and Japan (Klein, Ettenson & Morris 1998). Country of origin impact on Chinese consumers’ ethnocentric beliefs and attitudes relating to foreign brands can be country specific, for example, it might impact more on US than Australian brands. Development of hypotheses associated with the ethnocentrism construct 92 Based on the above discussion, the following relationships between the ethnocentrism construct and the following three stages of the proposed conceptual framework are hypothesised: H4a: The ethnocentrism construct has a significant influence on the beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products H4b: The ethnocentrism construct is significantly correlated to the pre-purchase evaluation of Chinese consumers towards the purchase of organic food products H4c: The ethnocentrism construct is significantly correlated to consumers’ behavioural/purchase intention of organic food products in urban China 3.4 Cognitive/affective stage Consumers’ attitudes and beliefs The second stage of the conceptual framework is termed cognitive/affective. The most significant components are consumers’ beliefs and attitudes. Fishbein and Ajzen (1975) propose that attitudes are composed of several aspects of beliefs. These are called normative beliefs and are developed from other factors such as peers and surrounding environment. The present study does not differentiate between attitudes and beliefs in relation to organic food consumption intentions. According to the Theory of Planned Behaviour (Ajzen 1991), human action is guided by three kinds of considerations: attitude towards the behaviour, subjective norm and perceived behaviour control. Perceived control refers to the consumer’s perception about their ability to perform a given behaviour. It is determined by beliefs about the presence of factors that may facilitate or impede performance of the behaviour. In agreement with this view, Chen (2007) claims that consumers may have positive attitudes towards organic food purchase, but may not buy it when faced with a perceived impediment. Aertsens et al. (2009) suggest that perceived behaviour control may cover ‘perceived barriers’ that may impede and ‘perceived abilities’ that may facilitate behavioural intention and behaviour towards consumption of organic food, for example, relatively higher price might be perceived as negative attributes. 93 Studies on personal beliefs and motivations indicate that food safety concern appears to be the most important factor in influencing the decision-making and purchase of organic food (Rimal, Moon & Balasubramanian 2005; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006). Also consumers of organic food are motivated by their beliefs and concern for the ecology and are more attuned to matters of personal health (Shepherd, Magnusson & Sjödén 2005) and environmental consciousness (Honkanen, Verplanken & Olsen 2006). Interestingly, other factors such as lifestyle and cultural influences create particular personal beliefs for consumers of organic food and can also be a driver for its purchase (de Magistris & Gracia 2008). The European Action Plan states that ‘increasing consumers’ organic knowledge is of vital importance for the development of organic food demand because organic knowledge influences attitudes towards organic food products that directly determine the decision or intention to buy the product’ (de Magistris & Gracia 2008, p. 943). A Danish study examined the relationship between attitudes and the purchase of organic foods. It describes eleven motivational values: universalism, benevolence, spirituality, conformity, tradition, security, power, achievement, hedonism, stimulation and selfdirection. These values are both self-centred and socially based (Grunert & Juhl 1995). The results suggest that Danish school teachers with ‘green’ attitudes are more likely to buy organic foods. In a sense, they possessed more environmental protection attitudes that unified them with nature. It is also evident in a Spanish study that consumers’ attitudes towards the environment and the agricultural system are highly significant in the decision to pay a premium for organic olive oil (Soler et al. 2002). On the other hand, a Swedish study (Shepherd, Magnusson & Sjödén 2005) has demonstrated a discrepancy between attitudes and behaviour in that positive attitudes towards organic food do not always guarantee its purchase. Generally consumers with positive attitudes towards organic food are more likely to display positive behavioural and purchase intention (Honkanen, Verplanken, and Olsen 2006). After all, behaviour is determined by the intention of individuals’ motivation (Fishbein and Ajzen 1975). Consumers who are more knowledgeable about organic food believe that it is healthier and of a higher quality. They therefore present more positive attitudes towards organic 94 food (de Magistris & Gracia 2008). An Italian study (Saba & Messina 2003) suggests that those consumers who hold positive attitudes towards eating organic fruit and vegetables on average agree that organic food products are healthier, environmentally friendly, taste better and are more nutritious than conventional food. Tanner and Kast (2003) suggest that personal influential factors, such as attitudes towards food, and personal norms play a significant role in the purchase of green food in Sweden. Consumers who attempt to pursue a healthy diet and balanced lifestyle are more likely to have positive attitudes towards organic food products and the environment, and are thus more likely to purchase organic food (de Magistris & Gracia 2008). 3.5 Evaluation of alternatives stage Pre-purchase evaluation A Greek study (Drichoutis, Lazaridis & Nayga Jr 2007) related to purchase behaviour of food suggests that nutritional knowledge is often regarded as a proxy for prior food knowledge. The action of searching for nutritional information is as important as that of searching for price information prior to the purchase of food. In line with this thinking, a study conducted in the Netherlands (Schifferstein & Oude Ophuis 1998) suggests that the conceptual model of consumers’ organic food choice decision process should embrace problem recognition, search information and alternative evaluation stages. Problem recognition might occur due to some obvious or latent health problem, which then motivates consumers to seek information about different types of food, and finally consider alternative evaluations prior to purchase. An Australian study (Chang & Zepeda 2005) emphasises that knowledge of organic food products appears to be an important attribute in influencing consumers’ attitudes and consumption. Regular consumers have more knowledge, and are more tolerant of higher prices and inaccessibility. Verbeke (2000) suggests that the consumer decision-making process regarding purchase of meat involves stages such as problem recognition, search and processing of information and evaluation of product alternatives. Development of hypotheses associated with the pre-purchase evaluation 95 Based on the above discussion, the following relationship between consumers’ attitudes/ beliefs and pre-purchase evaluation of the proposed conceptual framework is hypothesised: H5: Chinese consumers’ beliefs/attitudes towards organic foods are directly and positively correlated to their pre-purchase evaluation 3.6 Behavioural/purchase intention stage Intentions are generally good predictors of behaviour (Honkanen, Verplanken & Olsen 2006). Consumers who attempt to pursue a healthy diet and balanced lifestyle have stronger intentions of purchasing organic food products (de Magistris & Gracia 2008). Those who have positive attitudes towards organic foods naturally have greater intention to purchase them (Tarkiainen and Sundqvist 2005). Hansen (2005) developed a better understanding of consumers’ perception of food quality. This quality perception process framework was constructed using a number of variables which included stimuli, expected quality attributes, feeling of pleasure and purchase intention. Previous studies (Chen 2007; Honkanen, Verplanken & Olsen 2006) demonstrate that consumers with positive attitudes towards organic food have stronger intentions to buy it: ‘Intentions are assumed to capture the motivational factors that influence behaviour. They are indications of how hard people are willing to try, of how much of an effort they are planning to exert, in order to perform the behaviour’ (Ajzen 1991, p. 181). This means that the higher the intention to engage in purchase behaviour (of organic food), the more likely they are to undertake the purchase. Relevant studies (de Magistris & Gracia 2008; Saba & Messina 2003) show that consumers who believe that organic food is healthier and of higher quality have higher intentions to purchase it. Consumers who have a more positive attitude towards organic food have greater intention to purchase it (Tarkiainen & Sundqvist 2005). Development of hypotheses associated with the pre-purchase evaluation 96 Based on the above discussion, the following relationship between pre-purchase evaluation and behavioural/purchase intention of the proposed conceptual framework is hypothesised: H6: The pre-purchase evaluation of Chinese consumers to organic food is directly and positively related to their behavioural/purchase intention 3.7 Demographic variables As discussed in the literature review, several studies exhibit the impact of demographic variables on organic food consumption. The socio-demographic profile of consumers which includes level of income, education, occupation, gender, marital status, age group and social status significantly influences consumer attitudes towards the purchase of organic food. A higher level of disposable household income and education significantly increases the probability of being a heavy organic food consumer (Millock et al. 2004). The influence of demographic variables is not explicitly mentioned in the Theory of Planned Behaviour or Hierarchy of Effects model, both of which underpin the proposed conceptual model for the current study. They are nonetheless present implicitly, as consumer groups are divided along socio-demographic lines which may produce differences in relation to attitude, intentions and behaviour (Aertsens et al. 2009): ‘The influence of demographic factors is pervasive, affecting various stages of the consumer decision process, but the level of theorising is not commensurate with their importance’ (Steenkamp 1997, p. 161). Since the current study only focuses on the impact of the influencing stage (product/regulatory/lifestyle/ethnocentrism) on consumers’ beliefs/ attitudes, pre-purchase evaluation and purchase intention, the demographic variables are explained as perceived abilities in the stages of consumption of organic food. Hence, they are positioned as control variables during the purchase intention stage. As organic food is more expensive than conventional food, numerous studies (Rimal, Moon & Balasubramanian 2005; Tregear, Dent & McGregor 1994) suggest that higher income households are more likely to purchase it. Such households have positive attitudes towards purchase of health labelled products, as compared to those with lower 97 income levels. Other studies (Lockie et al. 2002; Squires, Juric & Cornwell 2001) question the strong link between higher income and consumption of organic food. A pan-European survey of consumer attitudes towards healthy eating highlights that educational levels appeared to be the strongest influence on perceptions of healthy diet, food choice, food quality and freshness (Margetts et al. 1997). Educational levels seem to be higher in the organic consumption group than the conventional food group (Storstad and Bjørkhaug 2003). Gender can also be an influential demographic variable in the decision-making and choice of organic food. Women are more concerned about health, nourishment and the environment and more likely to demand more health information. As such, they are more likely than men to purchase organic food (Davies, Titterington & Cochrane 1995; McEachern & McClean 2002; Rimal, Moon & Balasubramanian 2005). In Lockie et al.’s (2002) Australian study, 44% of women respondents claimed to have consumed certified organic food in comparison with 34% of men. These figures concur with a previous study undertaken in Northern Ireland by Davies, Titterington and Cochrane (1995) in which 46% of women consumed organic food as opposed to 34% of men. These studies identified women as being more aware of organic food and having more positive attitudes towards it. Women are more likely than men to hold positive attitudes towards the purchase of organic foods (Lea & Worsley 2005; Lockie et al. 2002; McEachern & McClean 2002). On the other hand, because women are more price conscious than men, men are more willing to pay a premium for organic food (Ureña, Bernabéu & Olmeda 2008). A Canadian study claims that males in a single household holding a bachelor degree and individuals older than 51 years are more likely to pay a premium price (Haghiri & McNamara 2007). Do demographic variables influence Chinese consumer attitudes towards organic food? Does gender affect Chinese consumers’ attitude? Do income, professional and education levels positively influence organic food consumption in urban China? The answers to these questions are explored later in this study. 98 Sun and Collins (2006) suggest that there are five intended uses of imported fruit among Chinese consumers: for gift giving, for aged parents, for their only child, for visiting patients and for self-consumption. Generally in urban China, both parents and grandparents may spend large amounts of money for children who are products of a onechild policy3 (Schütte & Ciarlante 1998). An Indian study (Chakrabarti and Baisya 2007) explains that one intended use of organic food is for consumers seeking remedies in poor health conditions. Sun and Collins (2006) point out that Chinese consumers purchase imported fruit not only to receive high quality products, but they also expect to gain some pleasure and symbolic benefits, such as ‘superior appearance’, ‘expensive’ and ‘high prestige’. The gift reflects both the status of the donor and respect towards the receiver. The Hofstede cultural dimension suggests that status is important to show power, success and achievement. National or global brands are regarded as revealing one’s status or being in the ‘right place’ in society (Mooij & Hofstede 2002). McEwen et al. (2006) argue that there is a misconception that Chinese consumers are all well-off and have a lot of money to spend. Yet at the same time they spend more on basic household goods and products like mobile phones, televisions and air conditioners, which can be seen as status as well as concern about products’ reliability. Bao, Zhou and Su (2003) demonstrate that purchasing certain kinds of products is a ‘face saving’ issue and positively influences consumers’ brand-consciousness and price-equals-quality and negatively affects the price-conscious and value-for-money perception. Development of hypotheses associated with the behavioural/purchase intention Based on the above discussion, the following relationships between demographic variables and behavioural/purchase intention of the proposed conceptual framework are hypothesised: 3 The Chinese government’s one-child policy refers to family bearing a single child applied to the Chinese mainland. The policy was introduced in 1978 and officially restricts married and especially urban couples to a one-child reproduction. It does allow some exemptions, primarily in rural areas and to ethnic minorities. 99 H7a:Women are more likely to purchase organic food than men in urban China H7b:Younger consumers are more likely to purchase organic food than older consumers H7c:Highly educated consumers are more likely to purchase organic food as compared to those who are less educated H7d:Consumers earning higher income are more likely to purchase organic food as compared to those earning less H7e:The important attributes of organic food products have positive influences on their intended usage pattern in urban China Summary of research hypotheses The development of hypotheses is based on the proposed conceptual framework in Figure 3.2 which illustrates and explains the purchase of organic food in urban China, and the key determinants which influence consumers’ decision-making process. Seven series of hypotheses have been developed because of the complex range of determinants in the process of consumer decision-making. The H1, H2, H3 and H4 battery of hypotheses address the relationship between product/regulatory/lifestyle/ethnocentrism and various stages of the proposed conceptual framework. H5 addresses the correlation of consumers’ beliefs/attitudes to pre-purchase evaluation. H6 addresses the relationship between prepurchase evaluation and behavioural/purchase intention. H7 examines the impact of control variables: gender, age, education, income and usage pattern, towards purchase intention. A summary of all research hypotheses is provided in Table 3.1. Table 3.1 List of all research hypotheses No. Hypotheses H1a The product-related construct has a positive influence on the beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products. Organic food product-related construct is positively correlated to the Chinese consumers’ pre-purchase evaluation. Organic food product-related construct is positively correlated to Chinese consumers’ behavioural/purchase intention. The regulatory construct associated with organic food has a positive influence on beliefs/attitudes of urban Chinese consumers. The regulatory construct associated with organic food is positively correlated to the Chinese consumers’ pre-purchase evaluation. The regulatory construct associated with organic food is positively correlated to Chinese H1b H1c H2a H2b H2c 100 H3a H3b H3c H4a H4b H4c H5 H6 H7a H7b H7c H7d H7e 3.8 consumers’ behavioural/purchase intention. The lifestyle construct has a positive influence on the beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products The lifestyle construct is positively correlated to consumers’ pre-purchase evaluation of organic food products in urban China The lifestyle construct is positively correlated to consumers’ behavioural/purchase intention of organic food products in urban China The ethnocentrism construct has a significant influence on the beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products The ethnocentrism construct is significantly correlated to the pre-purchase evaluation of organic food products in urban China The ethnocentrism construct is significantly correlated to Chinese consumers’ behavioural/purchase intention of organic food products Chinese consumers’ beliefs/attitudes are directly and positively correlated to their prepurchase evaluation. The pre-purchase evaluation of Chinese consumers to organic food is directly and positively related to their behavioural/purchase intention. Women are more likely to purchase organic food than men in urban China Younger consumers are more likely to purchase organic food than older consumers Highly educated consumers are more likely to purchase organic food as compared to those who are less educated Consumers earning higher income are more likely to purchase organic food as compared to those earning less The important attributes of organic food products have positive influences on their intended usage pattern in China Chapter summary This chapter describes the development of the proposed conceptual framework for the purchase of organic food in urban China. The proposed conceptual framework was built on and developed from the Consumer Decision Process (CDP) model along with the Theory of Planned Behaviour model (TPB) and Hierarchy of Effects model. This framework provides the theoretical foundations to examine the five stages of consumers’ decision-making process towards organic food, namely, influencing, cognitive/affective, evaluation of alternatives, behavioural/purchase intention and finally purchase. The current study only investigates consumers’ behavioural/purchase intention as the actual purchase involves collection of data of the same sample at some future period of time (longitudinal study). This chapter also puts greater emphasis on investigating the extent to which consumers’ beliefs and attitudes pertaining to organic foods are influenced by constructs such as product, regulatory, lifestyle and ethnocentrism. Along with the proposed conceptual framework, a series of elaborate and comprehensive hypotheses are developed to examine the relationship between the various constructs of the proposed 101 conceptual framework. Table 3.1 provides a full listing of all these hypotheses. The methodology used is explained in the following chapter. 102 CHAPTER FOUR: METHODOLOGY 4.1 Introduction The purpose of this chapter is to introduce and justify the research design employed in this study, which is a deductive approach, predominantly involving testing of the hypotheses developed in Chapter Three. Figure 4.1 diagrammatically illustrates the chapter structure Figure 4.1 Structure of Chapter Four Source: developed for this research Section 4.2 follows up on the conceptual framework developed in Chapter Three, and explains the research design. Section 4.3 establishes the development of the survey instrument, which includes measurement of the constructs of the proposed conceptual framework. It also deals with issues associated with reliability and validity of the data. Section 4.4 explains the sampling and the data collection procedures. Two stages of data collection were employed in this study. They included online and paper-based surveys. 103 4.2 Research design 4.2.1 Research paradigm A research paradigm is a philosophical framework that guides how scientific research should be conducted (Creswell 2009). The two main traditional paradigms are categories called positivism and interpretivism, which are associated with quantitative and qualitative methods of analyses. Collis and Hussey (2009, p. 56) affirm: Under positivism, theories provide the basis of explanation, permit the anticipation of phenomena, predict their occurrence and therefore allow them to be controlled. Explanation consists of establishing causal relationships between the variables by establishing causal laws and linking them to a deductive or integrated theory. A set of interrelated variables and propositions that specify relationships among the variables is a theory. Variables are attributes of the phenomenon which can be changed and measured, indicating cause and effect relationships. On the other hand, interpretivism attempts to understand subjective human perception by focusing on exploring, seeking and describing the meaning rather than measurement of statistics (Collis & Hussey 2009). Creswell (2009) summarises previous studies which demonstrate that there are five philosophical assumptions – ontological, epistemological, axiological, rhetorical and methodological – referred to as quantitative and qualitative paradigms. The nature of this research paradigm is influenced by the literature and research issues examined in the previous chapters. The main research problem is to understand the most important factors that influence consumers’ purchase intention relating to organic food in urban China. In the positivist paradigm, this research is predominantly deductive. Individual researchers have the freedom to choose the best method to suit the research purpose. A particular research paradigm may be more acceptable by personal choices and follow a tradition in the research discipline although it is determined by the research problem (Collis & Hussey 2009; Creswell 2009). Current literature studying consumers’ 104 awareness of organic food has been well developed in North America and Western Europe (Bonti-Ankomah & Yiridoe 2006). Previous studies (Chen 2007; Fotopoulos & Krystallis 2002b; Storstad & Bjørkhaug 2003) have provided highly specific and precise measurements to examine consumers’ purchase behaviour in relation to organic food. They emphasise that a survey instrument is most suitable to collect data in the organic food studies, particularly when the population of an entire country needs to be presented. A survey instrument provides more control over the research process, as described by Babbie (in Creswell 2009, p. 12): Survey provides a quantitative or numeric description of trends, attitudes, or opinions of a population by studying a sample of that population. It includes cross-sectional and longitudinal studies using questionnaires or structural interviews for data collection, with the intent of generalising from a sample to a population. A survey instrument is a proper sampling technique which allows the collection of a large amount of data from a sizeable population in a highly economical way, hence its popularity (Saunders, Lewis & Thornhill 2007). Although the term ‘survey’ is often used to describe the collecting of data using questionaries, it also includes structured observation and structured interviews (Saunders, Lewis & Thornhill 2007). In this study, survey instruments will be referred to as questionnaires. 4.2.2 Research context This section discusses the setting of the research environment to achieve stated research objectives. This study is conducted in the context of a developing and emerging market, China. Like in many industrialised countries, there is a realisation amongst Chinese consumers of the need for healthy and nutritious products such as organic food (Moon et al. 1998). Although a small and almost insignificant percentage of the 1.3 billion Chinese consume organic food products, China is developing an affluent urban middle class market for domestic organic food. Willer and Yussefi (2006) estimate that the Chinese domestic organic market increased by 30% compared to the previous year and this growth predominantly came from eastern China. Demand is also increasing for imported 105 organic products. Large international retailers like Wal-Mart and Carrefour have substantially increased sales of organic food products, even though they are more expensive (Baer 2007). As more consumers become health-conscious, China will potentially develop the world’s largest organic food market (CMP Information Ltd 2006). 4.2.3 Unit of analysis ‘The unit of analysis refers to the level of aggregation of the data collected during the subsequent data analysis stage. The research objective determines the unit of analysis’ (Cavana, Delahaye & Sekaran 2001, p. 119). In China, consumers in major cities are more affluent than in other cities, hence they are ideal targets for consumer good studies (Eves & Cheng 2007). Urban consumers also differ from those in rural areas (Dickson et al. 2004). In this study, the urban Chinese consumer is defined as the unit of analysis. Consumers were sourced across major supermarkets in four selected Chinese cities where the demand for organic food products is growing. Justification for choosing the four cities is provided in Section 4.4.5. The unit of analysis may compose the primary empirical research aim, individual or group under investigation (Davis 2005). The unit of analysis has been selected in line with the research objectives stated in Chapter One. 4.2.4 Refining the unit of analysis In a Chinese organic food study by Li, Cheng & Ren (2005), only 37.2% of respondents were aware of ‘organic food’. The term ‘organic’ is little understood by Chinese consumers, as organic food is only available in supermarkets in major cities (Paull 2008a). Studies suggest that using a screening question about the awareness of organic food at the commencement of a survey discriminates between buyers or potential buyers of organic food and non-buyers (Fotopoulos & Krystallis 2002a, b; Krystallis, Fotopoulos & Zotos 2006; Roitner-Schobesberger et al. 2008). In this study, a screening question – ‘Have you ever heard of the term “organic food”?’ – was asked at the beginning of the survey to ensure that data was collected only from valid participants. In other words, it was a requirement that respondents should have heard of the term and should have a basic understanding about organic food. It was determined for this study the respondents 106 should be regular, occasional or even potential organic food consumers. Respondents who answered ‘no’ were not included in this study. Therefore, this study only focuses on urban Chinese consumers who are aware of organic food products. 4.3 Design of the survey instrument The survey instrument comprised statements which were considered important in relation to the purchase intention of organic food in urban China. These statements were obtained from previous literature and three focus group discussions in China, and were refined by the researcher on the premises that: x They were appropriate for all respondents of the survey; x They were able to generate sufficient data to test the research hypotheses 4.3.1 Focus groups Focus groups are used for many purposes, including exploring issues, opinions and ideas of the research, developing survey questionnaires, increasing the understanding of survey findings and generating hypotheses that can be tested in the research (Boyce 2002). They often provide insights that direct questioning cannot present (Malhotra 2008). Focus groups are used to generate information to guide further quantitative research (Kotabe et al. 2005) and enlighten the content of further surveys (Saunders, Lewis & Thornhill 2007). They have been widely used in organic food research (Aarset et al. 2004; Chang & Zepeda 2005; Essoussi & Zahaf 2008; Lockie et al. 2004) and have been recognised as a tool for aiding survey design, providing better knowledge of consumer perceptions and demand for organic food and suggesting directions for further research. The objective of these focus groups was to investigate Chinese consumers’ attitudes to and motivations in purchasing organic food. Three focus groups each consisting of eight participants were organised in China with participants from different socio-demographic background. Structured questionnaires and open-ended questions were employed to understand Chinese consumers’ attitudes and beliefs towards the purchase of organic food. The focus groups’ processes were intended to gain a better understanding of attitudes towards the purchase of organic food, hence 107 develop meaningful survey questions. All participants had heard of the term ‘organic’ and purchased organic food products. The major questions discussed in the focus groups were: x Are Chinese consumers confused about differences in the terms ‘green food’, ‘non-polluted food’ and ‘organic food’? x What do you know about the following factors which influence the consumption of organic food: a. Product-related factors including sensory attributes and price perception b. Government policies and regulations, certification and labelling c. Sociocultural factors and lifestyle d. Country of origin and ethnocentric factors? x Why do Chinese consumers purchase organic food? x What are the salient demographic features of organic food consumers in China? The researcher who originally hails from China and has a thorough understanding of the nature of this study was the moderator for the three focus groups. After discussion with each focus group, the researcher modified and re-phrased the initial questions which were developed from previous literature. Additional items were identified, and comments and suggestions were utilised to arrive at the pre-test version of the survey instrument (see Section 4.3.5). 4.3.2 Operationalising the constructs ‘A strategy to develop scale measurement is to find a suitable existing instrument whose results generalise to a broader class of measures that legitimately employ the same name’ (Nunnally & Bernstein 1994, p. 53). Most statements were adapted from published research articles and validated measures for the seven key constructs of the proposed conceptual framework: product, regulatory, lifestyle, ethnocentrism, personal beliefs/attitudes, pre-purchase evaluation and behavioural/purchase intention. Table 4.1 depicts the components of the product-related construct and associated measures. Four statements relating to food product characteristics – texture, appearance, 108 smell and taste – which are termed ‘sensory appeal’ were selected. Chen (2007) adopted these measures to explain Taiwanese consumers’ attitudes and purchase intention in relation to organic food, and the internal consistency results were positive (α=0.87). The reasons for the non-purchase of organic food include consumers’ perception of it being expensive and lacking in value. Price perception items were also chosen by previous studies (Burton et al. 1998; Gil, Gracia & Sanchez 2000; Lichtenstein, Ridgway & Netemeyer 1993; Yoo, Donthu & Lee 2000). Table 4.1: Components of product-related construct and associated measures Component Author and year Adapted measure Sensory appeal Steptoe et al. 1995; Chen 2007 As above As above As above Price perception Gil et al. 2000; Chen 2007 Krystallis & Chryssohoidis 2005 Lichtenstein et al. 1993; Burton et al. 1998 Verhoef 2005; Yoo et al. 2000 Steptoe et al. 1995; Chen 2007 Factor loading Cronbach’s alpha Instrument item Has a pleasant texture 0.70 3.1 Looks nice Smells nice Tastes good Consuming organic food is trendy Organic food has high nutritional value Generally speaking, the higher the price of a product, the higher the quality The price of organic food is too high Organic food is good value for money 0.72 0.80 0.53 0.71 Chen: 0.87; Steptoe et al.: 0.70 As above As above As above N/A 0.70 N/A 3.6 N/A Lichtenstein 0.78; Burton et al. 0.85 3.7 Yoo et al. 0.94 Yoo et al. 0.88 3.8 Steptoe et al. 0.76 Chen: 0.72 3.9 3.2 3.3 3.4 3.5 Food safety is one of the most important issues influencing the purchase of organic food (Ho, Vermeer & Zhao 2006). There is a relatively poor understanding of the legal use of the term ‘organic food products’, especially in relation to inspection and certification systems (Krystallis & Chryssohoidis 2005). An organic food logo is the consumer’s guarantee that the product has been produced organically (Botonaki et al. 2006). However, owing to the huge number of labels and organic certification bodies, there is a need for government regulations to address this (Özcelik & Uçar 2008; Walsh, Laczniak 109 & Carlson 1998). There is also doubt about which of those can be trusted by consumers. Eight attributes associated with the regulatory construct were selected as shown in Table 4.2. Table 4.2: Components of regulatory construct and associated measures Component Certification and labelling Government regulation/ policy Author and year Botonaki et al. 2006 Krystallis & Chryssohoidis 2005 As above Focus group Walsh 1998 Focus group Özcelik & Uçar 2008 Ho et al. 2006 Adapted measure When I buy a food product, I always read the label I don’t trust organic food certification bodies I trust the outlets which sell certified organic food Chinese government regulate food marketing, and have developed significant policies Lack of adequate government control of media allows advertisers to take advantage of consumers Logos depicting types of organic foods should be controlled More land should be allocated for organic farming The market for ‘organic’ and ‘green’ is chaotic Factor loading N/A Instrument item 3.10 0.76 3.11 0.81 3.12 N/A 3.13 N/A 3.14 N/A 3.15 0.43 3.16 N/A 3.17 Consumption of organic food is related to lifestyle. Previous studies (Chen 2009; Lockie et al. 2002) demonstrate that lifestyle has an impact on consumers’ attitudes and consumption of organic food. The lifestyle variables comprising 12 items shown in Table 4.3 were adapted from Yang (2004). These scales were used to identity which lifestyle segments influence consumers’ attitudes towards purchase of organic food products in China. 110 Table 4.3: Components of lifestyle construct and associated measures Component Adapted measure Factor 1: selfWhen I like something, I will buy it without too indulgence (a = 0.70) much deliberation I always do whatever I feel like and whenever I feel like it The sole purpose of making money is to spend it Sometimes I feel like spending money on anything I lay my eyes on I often make impulse purchases Factor 2: variety I always try something new and unique seeking (a = 0.70) I love fashionable and trendy products It does not hurt to be trendy if I feel like it I am often influenced by advertisements of new products Factor 3: opinion I can easily influence people around me during leadership (a = 0.52) conversation My friends often consult me when they cannot make up their own mind I have a strong desire to be successful Factor loading 0.592 Instrument item 3.28 0.512 3.29 0.609 0.727 3.30 3.31 0.760 0.587 3.32 3.33 0.639 0.723 0.706 3.34 3.35 3.36 0.707 3.37 0.628 3.38 0.554 3.39 Source: Yang 2004 A product’s country of origin is an important attribute to influence organic food consumers’ willingness to pay. Consumers generally perceive locally produced products to be of better quality than imported ones (Krystallis & Chryssohoidis 2005). In addition, those with ‘local’ concern are more likely to purchase organic food (Torjusen et al. 2001). Locally produced goods involve short food miles implying lower fuel costs. This concept motivates consumers who are concerned for the environment (Wirth, Stanton & Wiley 2011). Organic food consumers show stronger ethnocentrism attitudes as compared to non-organic food consumers (Fotopoulos & Krystallis 2002a, b). Components for the ethnocentrism construct and associated measures were adapted from previous related studies (Shimp & Sharma 1987; Klein et al. 1998; Lindeman & Vaananen 2000; Chryssochoidis, Krystallis & Perreas 2007) and are shown in Table 4.4: 111 Table 4.4: Components of ethnocentrism construct and associated measures Component Ethnocentrism Country of origin Author and year Shimp & Sharma 1987; Chryssochoidis et al. 2007 As above Klein, Ettenson & Morris 1998 Lindeman & Vaananen 2000 Ethnocentrism Shimp & Sharma 1987; Chryssochoidis et al. 2007 Adapted measure Chinese people should buy domestic products rather than imported products China should levy heavy tariff on foreign products to reduce their quantity into China If two organic food products were the same in quality, but one was imported and the other was Chinese, I would pay more for the imported product I would purchase organic food coming from a country I approve of politically Chinese people should not buy foreign products, because it would hurt domestic business, and cause more unemployment Factor loading 0.65 Instrument item 3.40 0.58 3.41 N/A 3.42 0.85 3.43 0.67; 3.44 0.80 Consumers’ purchase of organic food is based on subjective experiences and perceptions (Hughner et al. 2007). The strongest motivation is related to health (Kraft & Goodell 1993; Squires, Juric & Cornwell 2001; Verhoef 2005). Another is driven by growing interest in ecological values, which is related to environmental sustainability and the impact of conventional agriculture on the environment (Honkanen, Verplanken & Olsen 2006). Consumers’ positive motivations towards organic food products have a bearing on their subsequent positive attitudes towards their purchase (Chen 2007). Ten attributes depicted in Table 4.5 were elicited to determine Chinese consumers’ beliefs and attitudes towards purchase of organic food. 112 Table 4.5: Attributes of beliefs/attitudes construct and associated measures Component Author and year Beliefs/ attitudes RoitnerSchobesberger et al. 2008; Magnusson et al. 2003 Gil et al. 2000 Focus group Kraft & Goodell 1993 RoitnerSchobesberger et al. 2008 As above Gil et al. 2000 Focus group Botonaki et al. 2006 McEachern & Willock 2004 Adapted measure Organic food is good for myself and my family’s health Organic food has no harmful effects I like the brands associated with organic food I worry about there being harmful chemicals in my food Organic food does not contain pesticides Organic food is good for the environment I believe that organic food has superior quality I personally think I should always buy organic food To me, it is important that the food I usually eat can be easily found in the food outlets near my house or workplace Organic food labels mean high quality food products Factor loading 0.822 Instrument item 3.18 0.74 3.19 N/A 3.20 0.759 3.21 0.696 3.22 0.612 3.23 0.74 3.24 N/A 3.25 N/A 3.26 0.646 3.27 There is a prevalent belief that organic food is healthier, safer, more nutritious and of higher quality. Consumers with positive attitudes towards organic food are more likely to form a stronger intention to purchase it. Hence Chinese consumers are willing to pay a premium price (Botonaki et al. 2006). Four items used to identify their pre-evaluation of organic food consumption were previously used by Bower (2001), and an additional item was used by Bennett (1999). Although some of the wording has been slightly altered, they are all related to pre-evaluation. Three behavioural intentional items were utilised by Cronin Jr et al. (2000), and the wording has also been slightly modified to suit Chinese consumers’ organic food purchase intention. Table 4.6 shows the measures for the prepurchase evaluation and behavioural intention constructs. 113 Table 4.6: Attributes of pre-purchase evaluation and behavioural/purchase intention constructs and associated measures Component Pre-purchase evaluation Author and year Bower 2001 As above As above As above Bennett 1999 Behavioural/ purchase intention Cronin Jr et al. 2000 As above As above Adapted measure I am eager to check out the product because of advertisements and promotion I intend to try out the product I am interested in experiencing the benefits of using the product It is likely that I will buy the product when it becomes available I can recall the brand names and labelling of some of the organic food products I will probably use the product in the future I will recommend usage of the product to my friends and relatives If I had to purchase the product again, I would make the same choice Factor loading N/A Cronbach’s alpha Reliability of scale = alpha of 0.80 Instrument item 4.1 N/A As above 4.2 N/A As above 4.3 N/A As above 4.4 N/A N/A 4.5 N/A 4.6 N/A Reliability of scale = alpha of 0.87 As above N/A As above 4.8 4.7 Twenty-three statements (Section Two of the Survey Instrument) were used to measure the important attributes of the purchase intention of organic food. These statements were utilised by previous studies, with some of the wording being slightly modified to suit the Chinese context. Obviously, product-related measures such as taste, smell, appearance and overall quality have been considered as important (Steptoe et al. 1995). In addition, limited availability and choice of organic food has been regarded as one of the obstacles to purchasing organic food (Steptoe, Pollard & Wardle 1995; Verhoef 2005). Food not only satisfies hunger, but in the Chinese context it is also a form of social status and prestige. Face saving (mianzi) is important for Chinese consumers. This means that people are more concerned with the perception of others towards themselves and with maintaining their own status. As such, ‘saving face’ is likely to influence Chinese consumers’ purchasing decisions (Qian, Razzaque & Keng 2007). An organic study 114 (Fotopoulos, Krystallis & Ness 2003) reveals that organic wine satisfies the expectations of high social status consumers who search for novel forms and styles. Similarly, Chinese consumers perceive certain types of imported fruit to be superior in appearance, and their consumption reflects their wealth and social status (Sun & Collins 2006). Chinese consumers are also becoming increasingly interested in grape wines (they traditionally only drink rice wine) and this is contributing to lifestyle changes. Initially, grape wines were considered to be symbols of social status, but they are now appreciated because of their taste and pleasure (Euromonitor International 2011). Table 4.7 depicts the important attributes when purchasing organic food. Table 4.7 Important attributes when purchasing organic food Author and year Krystallis & Chryssohoidis 2005 As above As above de Magistris & Gracia 2008 Krystallis & Chryssohoidis 2005 Verhoef 2005; Steptoe et al. 1995 Focus group Fotopoulos & Krystallis 2002 a, b Krystallis & Chryssohoidis 2005 Qian, Razzaque & Keng 2007 Sun & Collins 2006; Fotopoulos, Krystallis & Ness 2003 de Magistris & Gracia 2008; Schifferstein & Oude Ophuis 1998 Fotopoulos & Krystallis 2002 a, b Krystallis & Chryssohoidis 2005 As above As above Focus group Focus group Adapted measure The taste of organic food Instrument Items 2.1 The smell of organic food The appearance of organic food The overall quality of organic food 2.2 2.3 2.4 The price of organic food 2.5 The availability of organic food in convenience stores and supermarkets The promotion and advertising of organic food The value of organic foods relative to its price 2.6 The environmental benefits of organic food 2.9 The idea of face saving (mianzi) when purchasing organic food The social status of people purchasing organic food 2.10 The knowledge of organic food products 2.12 The awareness of organic food products 2.13 The country of origin of the organic food 2.14 Organic food that is produced in China The brand name of the organic food Government regulations and policies relating to the sale of organic food Food safety in relation to organic food 2.15 2.16 2.17 2.7 2.8 2.11 2.18 115 Krystallis & Chryssohoidis 2005 Focus group Krystallis & Chryssohoidis 2005 Krystallis & Chryssohoidis 2005 Krystallis & Chryssohoidis 2005 Certification relating to the quality of organic food 2.19 Enforcement relating to the quality of organic food Packaging of organic food 2.20 2.21 Correct labelling of organic food 2.22 Information about the nutritional value of organic food 2.23 4.3.3 Description of the survey instrument The survey instrument shown in Appendix 1 comprises five sections, as below: Section one: Investigated Chinese consumers’ general knowledge about organic food. It required respondents to indicate whether they recognised any of the organic food labels (Roitner-Schobesberger et al. 2008). Three logos used in the domestic market – ‘nonpolluted products’, ‘green food’ and ‘organic products’ – were selected and three logos belonging to imported organic foods. This was followed by questions related to their knowledge about organic food (Essoussi & Zahaf 2008). Section two: Respondents were required to indicate the level of importance of the attributes for the purchase of organic food. Twenty-three attributes were selected using a five point Likert scale. Section three: Examined perceptions, attitudes and beliefs towards organic food. The 44 statements operationalised constructs related to product, regulatory, lifestyle, ethnocentrism and beliefs/attitudes. Section four: Contained attributes concerning pre-purchase evaluation and purchase intention. The 8 statements operationalised constructs related to pre-purchase evaluation and behavioural/purchase intention. All statements used a five point Likert scale. Section five: Investigated socio-demographic variables of respondents, which included age, gender, occupation, family income and education level. Additionally, it sought to obtain their usage pattern. Five types of organic food were selected because these are the most popular and widely available in Chinese organic food stores. They represent staple 116 foods in the Chinese diet, and are significantly different from one another (US Department of Agriculture 2008). The last part of the survey was open ended, inviting respondents to comment on any aspect of organic food not covered in the survey instrument. 4.3.4 Scaling and measurement Attitudes and opinions are effectively evaluated and quantified by using Likert scales (Leedy & Ormrod 2005). A Likert scale is often used to collect data relating to opinion, and it is designed to examine how strongly the respondent agrees or disagrees with a statement (Cavana, Delahaye & Sekaran 2001; Saunders, Lewis & Thornhill 2007). A range of attitudes, guided by previous literature, were developed in relation to themes surrounding organic food and were measured using rating scales. Previous studies (Eves & Cheng 2007; Kim, Ma & Kim 2006) confirmed that five point Likert scales are comprehensible by Chinese consumers, and widely applied in Chinese food consumption studies (Eves & Cheng 2007; Sun & Collins 2002; Zhang 2002). Hence it was decided to use such scales to evaluate the strength of respondents’ agreement to the statements in the survey instrument. The efficacy of the five point Likert scales was also supported by the three focus groups. Chen (cited in Lee et al. 2002) suggests that collectivist and individualist 4 cultures produce different response patterns. An individualist would present an extreme response, whereas a collectivist would be happy to provide a moderate response to be in agreement with the group (Wild et al. 2007). In order to minimise the chances of Chinese respondents selecting neutral answers, both positive and negative statements were employed to ensure that all statements were read carefully before being answered (Saunders, Lewis & Thornhill 2007). 4 According to the Hofstede framework, people in collectivist cultures feel a strong association to groups, including family and work units, and the goal of maintaining group harmony is probably most evident in the family structure. Individualistic culture is opposite (Wild et al. 2007, p. 86). 117 4.3.5 Validity and reliability of survey instrument 4.3.5.1 Validity ‘Validity is concerned with whether the findings are really about what they appear to be about’ (Saunders et al. 2007 p. 150). Three validity tests were employed: face validity, content validity and construct validity. Face validity and pre-test Face validity indicates that the statements being presented in the questionnaire are clear and understandable (Cavana, Delahaye & Sekaran 2001). Prior to large-scale data collection, a pre-test was employed for both hard copy and online questionnaires to assess face validity. The purpose was to refine the questionnaires, so that respondents would not have any difficulty in answering all questions (Malhotra 2008; Saunders, Lewis & Thornhill 2007). The initial questionnaires were pre-tested using a set of 20 selected Chinese organic food consumers. Follow up emails or phone calls were made to the pre-test participants. Extra questions were asked, including: How many minutes did it take to complete the questionnaire? Are there any sentences difficult to understand? Is the layout of the survey clear and attractive (Saunders, Lewis & Thornhill 2007)? How about grammar, structure and content of the questions? Consequently, a number of changes were made to make the statements more understood by Chinese respondents. In order to maintain the validity and reliability of the final data set, respondents who were involved in the pre-test were not invited to participate in the actual data collection process. Content validity and translation into Chinese The primary data originates for the specific purpose of dealing with the research problem (Malhotra et al. 2006). In order to successfully interpret the meaning of the statements and collect primary data, Chinese cultural and linguistic characteristics need to be considered in this methodology (Ho, Vermeer & Zhao 2006). 118 People who speak more than one language have an advantage of being able to understand many different international points of view (McMurray, Pace & Scott 2004; Carroll 1974). The researcher possesses a bilingual capability in the Chinese and English languages and this was obviously an advantage. The original survey instrument was written in English, and then translated into Chinese by the bilingual researcher. Significant misunderstanding or confusion caused by a cross-cultural transformation can be detected through the backtranslation, and ensure the validity for a cross-cultural setting (La Ferle & Lee 2003; Phau & Min 2009; Wang et al. 2010). The translated Chinese version was back-translated into English to ensure the same meaning (Chan 2001). Discrepancies in the translation need to be carefully inspected and corrected to ensure the measurement instruments mean the same thing (Mullen 1995), and also contain no jargon and complex social judgements. To eliminate any errors, the translated version should be confirmed by different individuals (Kotabe et al. 2005; Zhuang et al. 2006). ‘Care must be taken to ensure that apparent cross-cultural differences are real and actually do not stem from poor translation’ (Lee et al. 2002, p. 298). Therefore, the translated versions were cross-checked by three other bilingual researchers who were competent in both Chinese and English and were selected from Australian and Chinese universities. Construct validity ‘Construct validity testifies to how well the results obtained form the use of the measure fit the theories around which the test is designed’ (Cavana, Delahaye & Sekaran 2001, p. 213). Two specific forms of construct validity, convergent validity and discriminant validity were used in this study, as discussed in Chapter Six: Sections 6.5.1 and 6.5.2. 4.3.5.2 Reliability ‘Reliability refers to the consistency and stability of a score from a measurement scale’ (Davis 2005, p. 188). In other words, the valid scale must be a reliable scale. EasterbySmith, Thorpe and Lowe (2002, p. 53) suggest that three questions must be considered before and whilst undertaking research: x Will the measures yield the same results on other occasions? x Will similar observation be reached by other observers? 119 x Is there transparency in how sense was made from the raw data? Robson (in Saunders, Lewis & Thornhill 2007) argues that there are four circumstances which invalidate the reliability: subject or participant error, subject or participant bias, observer error and observer bias. Subject or participant error may occur in a survey completed at different times of the week. The web link to the online survey was sent to selected food outlets for onward submission to customers. This enabled respondents to complete the survey at their convenience during the survey period. The paper-based data was collected on weekdays and weekends during supermarket trading hours (for details of data collection procedures, see Sections 4.4.3 and 4.4.4). In order to reduce subject or participant bias, both the online and paper-based responses were strictly anonymous and confidential, and respondents were free to decline or withdraw at any time. This was made known to them by a cover letter which accompanied the survey. Observer error and observer bias might arise due to different personnel involved in conducting data collection, with potentially different ways of asking questions and different ways of interpreting the replies. To eliminate such bias, the paper-based survey was administered by trained personnel who assisted the respondents in completing the survey. ‘Reliability means that a measure should consistently reflect the construct that it is measuring’ (Field 2009, p. 673). To assess the reliability of the statements, test-retest, alternative form and internal consistency approaches can be employed (Malhotra et al. 2006). Test-retest: ‘the reliability coefficient obtained with a repetition of the same measure on a second occasion is called “test-retest reliability”’ (Cavana, Delahaye & Sekaran 2001, p. 211). To find a cohort of the same respondents to answer repeats of the survey is quite challenging, therefore 20 Chinese consumers were selected using the snowball technique during the pre-test. The survey was sent to them by email in word document format and, after two weeks, it was sent again using a web link requesting them to complete the online version. The test-retest technique was also applied to the data 120 analysis in this study explained in Section 6.2.2, which split the data into different data set and then tested the data reliability. The alternative form approach ‘forms techniques of estimating reliability and assesses the equivalency of the content of sets of items. This technique usually involves the administration of equivalent scales to the same individuals, with or without a time interval’ (Davis 2005, p. 191). Saunders, Lewis and Thornhill (2007) suggest that it is often difficult to ensure the alternative form as it might involve increasing the length of the questionnaire. ‘Considerable debate centers around which of several alternative reliability estimates is best’ (Hair et al. 2006, p. 777). A commonly used measure of reliability is internal consistency which involves correlating the responses to each question in the survey with those of other questions (Saunders, Lewis & Thornhill 2007). Cronbach’s alpha is widely used to measure internal consistency of reliability (Hair et al. 2006). In order to main the internal consistency of data in this study, Cronbach’s alpha was used in the analysis phase. 4.4 Data Collection 4.4.1 Ethics approval Research ethical issues generally occur in the design and conduct of research on human subjects (Veal 2005). Prior to undertaking this research, ethical approval was sought from the Swinburne University of Technology, Human Research Ethics committee. A copy of the approved ethics proposal is attached (Appendix 3). This research was supported by the Chinese National Professional Committee of Organic Health Industry, and permission was acquired from the organisation to assist in data collection. An ‘informed consent’ was introduced at the beginning of the survey, which read as follows: x All questions and discussion about the research have been answered to respondents’ satisfaction; 121 x Respondents’ participation in the research is voluntary and they may withdraw at any time; x The survey is strictly anonymous. Data obtained is kept confidential, and only aggregated results (not individual responses) are mentioned in the research output. 4.4.2 Sampling The population of this study can be defined as ‘urban Chinese persons aged 18 and above who are familiar with the term organic food’. Survey participants were adults who were either regular, occasional or potential organic food consumers. The constructs and components of the conceptual framework were operationalised using different scales from past studies. Hence it was decided to perform a comprehensive pilot study. This would best be achieved by doing an online survey of respondents in China who were aware of organic food products. The objectives of the online data survey were to (i) perform an EFA to identify dimensions and latent variables of the proposed conceptual model, hence eliminating unnecessary indicators, (ii) inform the refinement of the survey instrument for further administration of the paper-based survey, (iii) initially scrutinise the validity and reliability of the factors and (iv) examine interrelationships between constructs and components of the conceptual framework in order to test research hypotheses at a later stage. Following the pilot study, it was decided to obtain a representative sample size using paper-based surveys. The purpose of this survey was to (i) confirm interrelationships between the constructs and components in the measurement model, (ii) assess the convergent and discriminate validities of various items in the survey, (iii) modify and finally confirm the measurement models, hence addressing the research hypotheses, (iv) estimate the best fit model, including the goodness-of-fit and significance of loadings and (v) use invariance tests to evaluate the impact of demographic and psychographic variables on the behavioural and purchase intention of respondents. 122 An explanation of the two stages of data collection, that is, stage one (pilot study using an online survey) and stage two (main data collection using a paper-based survey) follows: 4.4.3 Stage one: pilot study (online survey) The pilot study was conducted as an online experiment (Kim & Lennon 2010). Online surveys are an economical method of collecting data in the organic food industry from a national base (Rimal, Moon & Balasubramanian 2005). Completing the survey online has the advantage that data is delivered in an electronic format which can be instantly analysed (Saunders, Lewis & Thornhill 2007; Veal 2005) and respondents are geographically dispersed (Saunders, Lewis & Thornhill 2007). As part of the Chinese economic reforms, 1999 was declared ‘the year of getting on the internet’, with all ministries ensuring that they had their organisational websites (Kahal 2001, p. 70). According to the China Internet Network Information Centre (2009), there were over 338 million online users in China in 2009. A Chinese study (Cai et al. 2008) provided initial evidence that internet surveys are as reliable as paper surveys. However, although online data collection has its obvious advantages, respondents may not be representative of the general population (Malhotra 2008), hence this methodology of data collection was used only for the pilot study. The online survey was made available via Swinburne University of Technology’s Opinio platform. Opinio software enables production and reporting of a survey and ensures anonymity, confidentiality and privacy to the participant. The web link to the online survey was sent to the Chinese National Professional Committee of Organic Health Industry, which in turn forwarded it to selected Chinese food outlets. These outlets sent the web link to their customers. The sampling frames were drawn up using the participating food outlets’ customer mailing lists and participants were randomly selected from sampling frames. In order to guarantee the survey’s accuracy, and to prevent multiple completion by the same respondent, the default in Opinio was set as follows: ‘not allow multiple submissions’, and ‘prevent with cookies and IP-address check’. This means that the 123 Opinio software recognised every respondent’s IP address (computer ID), hence only one completed survey was received from a particular computer. Online surveys can be simply completed by respondents when screening questions are involved (Veal 2005). ‘Organic’ is a relatively new term and many Chinese consumers are still unaware of it. Hence the screening question used at the start of the survey was ‘Have you ever heard of organic food?’. A total of 405 responses (computer) IDs were detected by the Opinio software and a final total of 204 valid online surveys with no missing data were received. Hence the response rate was approximately 50%. The online survey was active for the whole of December 2009. 4.4.4 Stage two: main data collection (paper-based survey) After the online data was analysed, the survey instrument was slightly modified for use in paper-based surveys. Organic foods are relatively new products, primarily available in large supermarkets in major Chinese cities (Li, Cheng & Ren 2005). Previous organic studies suggest that it is efficient to collect data through food outlets of retail chains (Drichoutis, Lazaridis & Nayga Jr 2007; Krystallis, Arvanitoyannis & Chryssohoidis 2006; Krystallis & Chryssohoidis 2005). The paper-based surveys were administered at major supermarkets in four selected first and second tier5 cities: Beijing, Shanghai, Shenzhen and Chengdu, over a period of one month. Details of selected survey locations are justified in the next section. These four cities are economically and politically prominent and are the main engines of China’s phenomenal economic development. Hence consumers are more affluent, and are more likely to be aware of organic foods. All participating supermarkets had organic food sections. 5 First tier cities are those granted approval by the central government to establish joint venture retail enterprises. They include Beijing, Guangzhou, Shanghai and the Special Economic Zones (SEZs). Second tier cities are those without such approval but with good potential due to their advantageous location of being in close proximity to first tier cities (Wong and Yu 2003). 124 Shoppers were approached randomly by trained personnel and asked to participate in self-administered surveys at the main entrance of the supermarkets. The trained personnel who had Bachelors Degrees in Commerce were staff members of the Chinese National Professional Committee of the Organic Health Industry. They were instructed to seek out demographic variation where possible and the surveys were presented to consumers entering these supermarkets throughout the day during the stores’ trading hours. Consumers were asked ‘Have you ever heard of the term organic food?’. If they answered ‘Yes’, they were then asked ‘Would you like to participate in the survey?’. If they answered ‘No’, the next customer entering the store would be approached and so on. Then, every tenth individual consumer who walked into the supermarket was asked the same questions. A small gift of organic food (e.g. nuts, vegetables, milk) was offered in appreciation of their time to answer the questions. The paper-based data was collected over a period of four weeks in April 2010. The researcher was stationed in China throughout the duration of the data collection. Previous studies (Fotopoulos & Krystallis 2002b; Kuhar & Juvancic 2010; Magnusson et al. 2001; McEachern & Willock 2004; Roddy, Cowan & Hutchinson 1996) suggest that a sample size of more than 1000 respondents is sufficient to conduct a national survey. A total of 2000 surveys (500 in each city) were distributed, and 964 valid questionnaires were obtained. Both the online and paper-based survey response rates were as high as 50%. This is consistent with extant literature which suggests that the response rates of consumers of specialised goods like organic food products can be as high as 40% to 50% (Haghiri & McNamara 2007; Honkanen, Verplanken & Olsen 2006; McEachern & Willock 2004; Tregear, Dent & McGregor 1994). 4.4.5 Selection of cities in China for stage two: paper-based data collection China’s economic reforms have helped alleviate poverty and have been successful in exponentially increasing per capita income. However, they have also led to increasing disparity and inequality in income and wealth (Yeon 2008). It is important to recognise that China is a country of a conglomeration of regional diversity and economic disparity (Cui & Liu 2000; Ho & Tang 2006; Wong & Yu 2003), so market segments cannot be 125 oversimplified (Dickson et al. 2004). According to Euromonitor International (2008a), there are 22 provinces, four municipalities and five autonomous regions. The two most populous cities, Beijing and Shanghai, have the highest household expenditure. First tier cities, such as Beijing, Guangzhou and Shanghai have propelled the Chinese economy forward. Second-tier ones such as Chengdu, Dalian, Hangzhou and Xian also contribute to economic growth, with consumption patterns comparable to those in the first tier cities. Previous related studies chose four cities, i.e. Beijing, Shanghai, Chengdu and Guangzhou, to draw sample of consumers, representing regions with different geographic location, political and commercial backgrounds (Uncles & Kwok 2008; Wang et al. 2000; Wang & Chen 2004). However, Shenzhen and Guangzhou have a lot in common, partly because they are both located in the same province and just a few hours distant from each other by car. Shenzhen is a city representing modern Chinese lifestyle where organic food products are available in plenty (Yuan & Liu 2009). Hence this study has intentionally selected Shenzhen in lieu of Guangzhou. Figure 4.2 depicts the four cities selected in this study which are geographically dispersed throughout China. Figure 4.2 Map showing the geographic dispersion of the four selected cities Source: Google Image 2010 126 Shanghai Shanghai is the most populous city in China with 20 million people. It is the largest commercial centre located in the eastern seaboard. It has been the main engine of China’s phenomenal economic development. The per capita GDP in Shanghai in 2007 was US$9900, notably higher than the national average of US$2490. Its growing middle class has taken a lead role in purchasing organic food products in urban China. The first organic food shop in China opened in Shanghai (US Department of Agriculture 2008). Consumers in Shanghai are more adventurous than those in other cities and are willing to accept new products (Zhao et al. 2000). They are also likely to be trendsetters as compared to those in Beijing who are regarded as being more serious and sober in their judgement (Cui & Liu 2000; Schütte & Ciarlante 1998). Beijing Beijing is the political and cultural capital of China. It is home to over 8600 diplomatic missions, international organisations and Chinese provincial and city representative offices, with a population of 15 million. It is the one of the largest and wealthiest cities in China (US Department of Agriculture 2012a). With a history dating back more than 3000 years, Beijing is an amazing mix of old and new China (Beijing Municipal Government 2010). Chengdu Chengdu is the largest south-western city in China, and the capital of Sichuan province with a substantial population of 11 million (National Bureau of Statistics of China 2009). It is the wealthiest and most advanced city in the region, and symbolises the more traditional part of China (Wang & Chen 2004). It opened up to foreign business much later than Shanghai (Fram, Le & Reid 2005). Chinese economist Zhang Wuchang regards Chengdu as the economic headquarters of western China (Yin 2010). Although per capita incomes are lower than in some coastal cities, residents have a strong interest in food and are prepared to pay a premium price for high quality products (Stewart & Bean 2006). 127 Shenzhen The rise of its coastal cities has played a critical role in China’s economic development. The Special Economic Zones (SEZ) are looking to replicate the economies of Asian Tiger countries. Shenzhen is a young city, built in the early 1980s. It was the first Chinese SEZ resulting from the economic reforms and it attracts a migration population from the rest of China (Ho & Tang 2006). According to a government report (Shenzhen government online 2009), Shenzhen’s economic power is ranked fourth in China with its per capita GDP the highest in the country at US$13,153 in 2008, and the per capita annual disposable income reaching RMB26,726 (equivalent to US$3,818). There are 10.36 million permanent residents and its domestic migration makes up 86% of its total population (Atsmon et al. 2009). It is a city more open and exposed to contemporary lifestyles, and constitutes the most important future organic food market in mainland China (Yin et al. 2010). 4.5 Chapter summary This chapter introduced the research design, which is predominantly quantitative in nature, and explained why surveys were the best method to use to address the research objectives of this study. It detailed several stages in the development and design of the survey instruments. Focus groups were conducted prior to establishment of survey instruments. To ensure their validity and reliability, pre-testing was undertaken prior to the final survey distribution. The data was collected in China in two stages: 204 valid online questionnaires were collected for the purpose of the pilot study, and later 964 completed paper-based questionnaires were collected in four major cities. The next two chapters discuss the analysis of data collected during the pilot and main stages. 128 CHAPTER FIVE: ANALYSES AND RESULTS OF THE PILOT STUDY (STUDY ONE) 5.1 Introduction As explained in the previous chapter, online data was collected for the purpose of the pilot study. This was analysed using the exploratory factor analysis technique to generate relevant factors and associated latent variables in each of the constructs of the proposed conceptual model. Statistical Software Package for Social Sciences (SPSS) version 17 was employed to compute and analyse the data. Section 5.2 introduces the objectives of the data analysis of the pilot study. Section 5.3 describes the statistical methods employed. Section 5.4 identifies the factors and latent variables in each construct. Section 5.5 concludes with the chapter summary. Figure 5.1 diagrammatically illustrates the structure of Chapter Five. Figure 5.1 Structure of Chapter Five Source: developed for this research 5.2 The objectives of the data analysis in the pilot study It is always desirable to conduct a pilot study before administering a self-completion questionnaire to the sample (Bryman & Bell 2007). ‘A pilot study is a dress rehearsal of 129 the main study where you replicate the conditions as closely as possible to the main study’ (McMurray, Pace & Scott 2004, p. 268). The survey should be piloted with a reasonable sample size of respondents who come from or closely resemble the target population (Cavana, Delahaye & Sekaran 2001). The sample of the pilot study was drawn from adult Chinese consumers who had heard of the term ‘organic food’. The objectives of the pilot study were: x To perform an EFA to identify factors and associated latent variables for the constructs of the proposed conceptual model. During this process, variables which did not load on the factors would be eliminated. x To develop initial scales which would be subsequently translated into measurement models by confirmatory factor analysis (CFA) using data from the main study; x 5.3 To initially examine the validity and reliability of the online data. Choosing the right statistical method Data screening The online data for the pilot study was generated using Swinburne’s Opinio platform. A total of 405 respondents logged into the online web link, of whom 204 (50.4%) completed the survey. There were no missing data as the online survey had a built-in filter system. Each respondent could attempt to complete the survey only once on any one computer. The final response rate is consistent with literature which suggests that it can be high as 40% to 50% in organic food studies (Honkanen, Verplanken & Olsen 2006; McEachern & Willock 2004). 48% of valid respondents were females and 52% were males. Generally statistical techniques are sensitive to outliers. ‘An outlier is a case with such an extreme value on one variable (a univariate outlier) or such a strange combination of scores on two or more variables (multivariate outliers) that it distorts statistics’ (Tabachnick & Fidell 2007, p. 72). The descriptive function in SPSS calculates means and 5% trimmed means, which give an indication of the problem associated with outliers. If the two mean values are closely similar, it is indicative that the outliers are not 130 problematic (Pallant 2007). The results (Appendix 4 Table A) of this type of analysis showed that there were no significant differences between the means and the 5% trimmed means (when 5% of highest and lowest values are deleted). Therefore, no outliers were deleted from the data of the pilot study. Exploratory factor analysis Data was analysed using exploratory factor analysis (EFA) which is often carried out in the early stages of research to consolidate variables and to produce hypotheses about underlying processes (Tabachnick & Fidell 2007, p. 935). The main purpose is to identify the smallest number of meaningful latent variables or factors that closely reproduce the original correlations and covariances amongst a larger set of measured variables (Gorsuch 1997). In order to ascertain the constructs or factors and latent variables and also explore the correlation of each factor in the proposed conceptual model, a series of EFAs were performed on the data of the pilot study. Rotation Principal components factor extraction has been widely applied in food consumption studies (Botonaki et al. 2006; Chryssochoidis, Krystallis & Perreas 2007; Sanjuán et al. 2003). This method is commonly used in the analysis of psychological data (Pallant 2007; Todman & Dugard 2007). There are two general categories of rotation in factor analysis, orthogonal and oblique. Factors are always rotated at right angle to each other in orthogonal rotation, which means they are uncorrelated to each other. Oblique rotation assumes that the factors are correlated, which allows more freedom in selecting the position of factors in factor space than does orthogonal rotation (Gorsuch 1983; Kline 1994). In practice, orthogonal and oblique often result in very similar solutions (Tabachnick & Fidell 2007). Many researchers conduct both approaches and then report the result which is clearer and easier to interpret (Pallant 2007). Hence in this study principal components factor extraction with direct oblimin (oblique) rotation was selected to derive the factor loadings. 131 Factor loadings are ‘the correlations of a variable with a factor’ (Kline 1994, p. 5), which explain weighted combination of the variables’ loadings on each factor, within the range of 0 to 1 (Gorsuch 1983; Pallant 2007). They score the weight of impact of each variable on the endogenous variables. There is no consensus about what constitutes a ‘high’ or ‘low’ factor loading; however, in the social sciences the threshold is 0.30, the higher the factor loadings, the better the results (Hair et al. 2006; Peterson 2000). If more than one factor is generated during the EFA, a structure matrix of correlations between factors and variables and a pattern matrix of unique relationships between each factor and each observed variable are presented (Tabachnick & Fidell 2007). 5.4 Identifying factors and latent variables The analyses of the pilot study comprised a series of EFAs which were conducted to eliminate variables with factor loading < 0.3 (Hair et al. 2006; Peterson 2000). In order to be considered suitable for factor analysis, Bartlett’s Test of Sphericity value should be statistically significant at p <0.056 and the Kaiser-Meyer-Olkin (KMO) value should be 0.6 or above (Pallant 2007). Bartlett’s Test of Sphericity indicates that the data is suitable for factor analysis. A non-significant (p>0.05) Bartlett’s Test of Sphericity indicates that the data is not suitable for factor analysis. KMO reports the amount of variance in the data that can be explained by the factors and is a measure of sampling adequacy (Allen & Bennett 2010). Both Eigenvalues and Scree Plot estimate the number of factors that should be selected. ‘Eigenvalues represent variance. Because the variance that each standardised variable contributes to a principal components extraction is 1, a component with an eigenvalue less than 1 is not as important, from a variance perspective, as an observed variable’ (Tabachnick & Fidell 2007, p. 644). It is important to study the Scree Plot provided by SPSS. Catell (in Pallant 2007) recommends retaining all factors above the elbow or break in the plot as these contribute the most to the explanation of the variance in the data set. 6 P-value is the probability or chance that the null hypothesis is true. The lower the p-value, the less likely the result is if the null hypothesis is true, and consequently the more ‘significant’ the result is, in the sense of statistical significance (Cavana, Delahaye & Sekaran 2001). 132 To be considered suitable for factor analysis, the correlation matrix should also show at least some correlation of r =0.3 or above (Pallant 2007). However, larger sample sizes tend to produce smaller correlations (Tabachnick & Fidell 2007). The internal consistency was examined by estimating the Cronbach’s alpha coefficients in this study. ‘Internal consistency reliability is concerned with the homogeneity of the statements within a scale. Scales based on classical measurement models are intended to measure a single phenomenon which is typically equated with Cronbach’s coefficient alpha, α’ (DeVellis 2003, p. 27). The reliability of the measurement scale was checked in order to meet the reliability of the survey instrument. Cronbach’s alpha measures the inter-statement consistency reliability (Cavana, Delahaye & Sekaran 2001). Cronbach’s alpha values above 0.7 (Cronbach 1951) are considered acceptable, and values above 0.8 are preferable. This is a measurement of internal consistency (Pallant 2007). Following the conceptual framework developed in Chapter Four, 52 statements were used to examine consumers’ purchase intention relating to organic food in urban China. These statements were subjected to EFA to identify underlying factor patterns, and in the process eliminating unnecessary indicators. It was decided to conduct EFA for each construct separately rather than for a combination of the 52 statements. This is because most constructs predominately used items from previously validated scales as has been explained in Chapter Four. 5.4.1 Results of the EFA for the product-related construct Table 5.1 shows the results of the EFA for the product-related construct and associated measures. The KMO value was 0.696 which exceeded the recommended value of 0.6. Bartlett’s Test of Sphericity value was significant (p<0.05), therefore factor analysis was appropriate for this construct. Only the first two components recorded eigenvalues above 1 (2.628, 1.226), and they explained a total of 55.10% of the variance. The Scree Plot also suggested the existence of two components which were labelled ‘Sensory appeal’ and ‘price perception’ (details are shown in Appendix 4, Table B). 133 Table 5.1 Pattern and structure matrix for EFA of the product-related construct and associated measures Factor Factor 1: Sensory appeal Factor 2: Survey Content of statement statement 3.2 Organic food looks nice 3.1 Organic food has a pleasant texture 3.3 Organic food smells nice 3.9 Organic food is good value for money Price perception Pattern matrix .882 .865 Structure matrix .848 .878 .694 .779 .730 .743 .648 .693 .643 .697 .485 .470 Cronbach’s alpha .767 .547 3.5 Consuming organic food is trendy 3.6 Organic food has high nutritional value 3.7 Generally speaking, the higher the price of a product, the higher the quality Total Variance Explained (%) 55.053 KMO: .696 Bartlett’s Test of Sphericity: .000 From the initial nine statements in this construct, two were eliminated: statement 3.4 was deleted due to cross-loadings, and statement 3.8 was deleted as the loading was less than 0.4. The results (Appendix 4, Table B in the Component Correlation Matrix) showed that the correlation matrix of the two factors was 0.298, which is close to 0.30. This is considered to be moderately correlated as r =0.30 to 0.49 signifies a medium correlation (Pallant 2007). As this was a pilot study with a reasonable final response of 204, this value is acceptable. Overall, the factor loadings were acceptable, with the highest loading of 0.882 (pattern) and the lowest loading of 0.485 (pattern). The Cronbach’s alpha value for ‘sensory appeal’ was 0.767, which suggests good internal consistency of the scale. Nunnally (1967) suggests that the reliability of Cronbach’s alpha ranging from 0.50 to 0.60 is sufficient for early stages of basic research. Therefore, although the Cronbach’s alpha of ‘price perception’ was 0.547, which is less than the desirable value of 0.70, it was retained for the next stage of analysis. 134 5.4.2 EFA results for the regulatory construct EFA was selected for analysing of the regulatory construct and associated measures, as the KMO was 0.771 (greater than the recommended 0.6), and Bartlett’s Test of Sphericity was significant. The initial EFA results suggested a two factor solution for the eight statements. However, one of the factors comprised only two statements (statement 3.12 ‘I trust the outlets which sell certified organic food’ and statement 3.13 ‘Chinese government regulate food marketing, and have developed significant policies’ shown in Appendix 1). Two statements are generally insufficient for factorial stability, unless highly correlated (r>0.7) with the factor and relatively uncorrelated with other variables (Worthington & Whittaker 2006). The two statements had a correlation value of r=0.389 which is lower than the bench mark. Hence these two statements were deleted from the results of the EFA. The remaining six statements suggested one factor which had an eigenvalue above 1 (2.558), and which explained 42.63% of the variance. The Scree Plot also recommended a single factor for the regulatory construct (see Appendix 4, Table C). Table 5.2 illustrates the EFA results for the regulatory construct and its associated measures. Overall, the factor loadings are acceptable, with the highest loading of 0.783 and the lowest loading of 0.401. The Cronbach’s alpha value for this factor was 0.709. This suggests good internal consistency of the sample data. Table 5.2 Factor loadings for EFA of the regulatory construct and associated measures Survey statement 3.15 Contents of statements Logos depicting types of organic foods should be controlled 3.17 The market for organic and green is chaotic 3.16 More land should be allocated for organic farming 3.14 Lack of adequate government control of media allows advertisers to take advantage of consumers 3.10 When I buy a food product, I always read the label 3.11 I don’t trust organic food certification bodies Total Variance Explained (%) 42.627 KMO: .771 Bartlett’s Test of Sphericity: .000 Loadings Cronbach’s alpha .783 .750 .702 .709 .637 .567 .401 135 5.4.3 EFA results for the lifestyle construct The KMO value was 0.748, and Bartlett’s Test of Sphericity value was significant, hence EFA was employed to analyse the data. Only the first three components recorded eigenvalues above 1 (3.624, 1.553, 1.330), which explained a total of 59.15% of the variance. The Scree Plot also recommended a three factor solution for this data. Statement 3.31 ‘Sometimes I feel like spending money on anything I lay my eyes on’ was deleted, due to this statement’s significant cross-loadings on both variety seeking and self-indulgence. The correlations between factors ranged between r=0.268, and r=0.340, indicating that they were moderately correlated (details are shown in Appendix 4, Table D). Table 5.3 shows the results of EFA for the lifestyle construct and associated measures. Table 5.3 Pattern and structure matrix for EFA of the lifestyle construct and associated measures Survey Contents of statements statement Factor 1: 3.34 I love fashionable and trendy Variety products seeking 3.32 I often make impulse purchases 3.36 I am often influenced by advertisements of new products 3.33 I always try something new and unique 3.35 It does not hurt to be trendy if I feel like it Factor 2: 3.28 When I like something, I will Selfbuy it without too much indulgence deliberation 3.29 I always do whatever I feel like and whenever I feel like it 3.30 The sole purpose of making money is to spend it Factor 3: 3.38 My friends often consult me when they cannot make up their own mind Opinion 3.37 I can easily influence people leadership around me during conversation 3.39 I have a strong desire to be successful Total Variance Explained (%) 59.149 KMO: .748 Bartlett’s Test of Sphericity: .000 Factor Pattern matrix .812 Structure matrix .809 .736 .682 .745 .685 .679 .684 .673 .693 .916 .906 .898 .890 .672 .732 .829 .821 .734 .759 .590 .594 Cronbach’s alpha .773 .801 .570 136 The resultant factor loadings of between 0.590 and 0.916 are all acceptable. The twelve statements which measure lifestyle were adapted from Young’s (2004) study, which similarly reported a three factor solution. However, the internal consistencies for the factors in this study were stronger than those reported in Young’s study, that is, the Cronbach’s alpha values of the three factors in this study were 0.773, 0.801 and 0.570 as compared to Young’s 0.700, 0.700 and 0.520. Similar to Young, the three factors generated in this study have been named variety seeking, self-indulgence and opinion leadership. 5.4.4 EFA results for the ethnocentrism construct Table 5.4 shows the EFA results for the ethnocentrism construct and associated measures. The KMO value was 0.689, and Bartlett’s Test of Sphericity value was significant, hence it was appropriate to use EFA. Only one factor recorded an eigenvalue above 1 (2.07), which explained a total of 69.12% of the variance. The Scree Plot also recommended a one factor solution. Two of the five statements in this construct were deleted due to the low correlation of the two statements (r =0.251) and low Cronbach’s alpha (α= 0.400). These were 3.42 ‘If two organic food products were the same in quality, but one was imported and the other was Chinese, I would pay more for the imported product’ and 3.43 ‘I would purchase organic food come from a country I approve of politically’. The overall results were good with all loadings above 0.8 (details are shown in Appendix 4, Table E). The Cronbach’s alpha value was 0.776, which suggests good internal consistency Table 5.4 Result for EFA of the ethnocentrism construct and associated measures Survey Contents of statements statement 3.40 Chinese people should buy domestic products rather than imported products 3.41 China should levy heavy tariff on foreign products to reduce their quantity into China 3.44 Chinese people should not buy foreign products, because it would hurt domestic Loadings Cronbach’s alpha .864 .821 .776 .808 Total Variance Explained (%) 69.115 KMO: . 689 Bartlett’s Test of Sphericity: .000 137 5.4.5 EFA results for the beliefs and attitudes construct Table 5.5 depicts the EFA results for the beliefs and attitudes construct and its associated measures. The KMO value was 0.900, and Bartlett’s Test of Sphericity value was significant, hence it was appropriate to use EFA. Only one factor recorded an eigenvalue above 1 (4.898), and explained a total of 48.99% of the variance. The Scree Plot also recommended the retention of only one factor in this construct. Overall, the loadings were acceptable, with the highest being 0.801 and the lowest 0.476 (details are shown in Appendix 4, Table F). The Cronbach’s alpha value was 0.881, which suggests good internal consistency of the items in this factor. Table 5.5 Result for EFA of the beliefs and attitudes construct and associated measures Survey statement 3.24 3.23 3.18 Contents of statements I believe organic food has superior quality Organic food is good for the environment Organic food is good for myself and my family’s health 3.27 Organic food labels mean high quality food products 3.25 I personally think I should always buy organic food 3.20 I like the brands associated with organic food 3.19 Organic food has no harmful effects 3.22 Organic food does not contain pesticides 3.21 I worry about harmful chemicals in my food 3.26 To me, it is important that the food I usually eat can be easily found in the food outlets near my house or workplace Total Variance Explained (%) 48.985 KMO: .900 Bartlett’s Test of Sphericity: .000 Loadings Cronbach’s alpha .801 .778 .736 .729 .881 .711 .703 .699 .686 .627 .476 5.4.6 EFA results for the pre-purchase evaluation construct Table 5.6 shows the EFA results for the pre-purchase evaluation construct and associated measures. The KMO value was 0.711 and Bartlett’s Test of Sphericity value was significant, hence it was appropriate to use EFA. Only one factor recorded an eigenvalue above 1 (2.512), which explained a total of 50.24% of the variance. The Scree Plot also recommended the retention of only one factor. Overall, the loadings were acceptable, with the highest being 0.851 and the lowest 0.452 (details are shown in Appendix 4, 138 Table G). The Cronbach’s alpha value was 0.699, which suggests adequate internal consistency of the items in this factor. Table 5.6 Result for EFA of the pre-purchase evaluation construct and associated measures Survey statement 4.3 Contents of statements I am interested in experiencing the benefits of using organic foods 4.2 I intend to try out organic food products 4.4 It is likely that I will buy organic food products when they become available 4.5 I can recall the brand names and labelling of some of the organic food products 4.1 I am eager to check out organic food products because of advertisements and promotion Total Variance Explained (%) 50.235 KMO: .711 Bartlett’s Test of Sphericity: .000 Loadings Cronbach’s alpha .851 .839 .803 .699 .485 .452 5.4.7 EFA results for the behavioural/purchase intention construct Table 5.7 shows the EFA results for the behavioural/purchase intention construct and associated measures. The KMO value was 0.704, and Bartlett’s Test of Sphericity value was significant, hence it was appropriate to use EFA. Only one factor recorded an eigenvalue above 1 (2.233), which explained a total of 74.43% of the variance. The Scree Plot suggested the retention of only one factor for items of this construct (details are shown in Appendix 4, Table H). All loadings were above 0.80, indicating their strength. The Cronbach’s alpha value was 0.827, which suggests good internal consistency of items in this factor. Table 5.7 Result for EFA of the behavioural/purchase intention construct and associated measures Survey statement 4.7 Contents of statements I will recommend usage of organic foods to my friends and relatives 4.8 If I had to purchase organic foods again, I would make the same choice 4.6 I will probably use organic food products in the future Total Variance Explained (%) 74.431 KMO: .704 Bartlett’s Test of Sphericity: .000 Loadings Cronbach’s alpha .890 .876 .827 .821 139 5.4.8 Consolidated details of the ten factors Ten factors were retained from the process of a series of EFAs based on the proposed framework for the purchase intention of organic foods in urban China (Figure 3.2). The overall reliability showed good internal consistency of the items within the ten factors. Hence 45 out of the 52 statements from the original survey were retained and were used in the next stage of data collection, the paper-based survey. Table 5.8 summarises the ten factors obtained from this analysis. Table 5.8 Summary of the ten factors Construct Product (two factors) Regulatory Lifestyle (three factors) Ethnocentrism Beliefs/attitudes Pre-purchase evaluation Behavioural/ purchase intention 5.5 9 Number of statements deleted during EFA 2 Number of statements retained in EFA 7 8 12 2 1 6 11 5 10 5 2 Nil Nil 3 10 5 α 1=.767 α 2=.547 .709 α 1=.773 α 2=.801 α 3=.570 .776 .881 .699 3 Nil 3 .827 Original number of statements Reliability (Cronbach’s alpha) Chapter summary To summarise, a series of EFAs were performed on the data of the pilot study to identify the factors and associated latent variables. The analysis was based on data collected from a final sample of 204. Of the 52 statements, a total of seven were deleted during the process of EFA. These results suggest that there are ten factors underpinning seven constructs in the proposed conceptual framework: sensory appeal, price perception, regulatory, self-indulgence, variety seeking, opinion-leadership, ethnocentrism, beliefs/attitudes, pre-purchase evaluation and behavioural/purchase intention. This pilot study has provided strong evidence to support further analysis of the proposed conceptual 140 model. The next chapter discusses analysis of the main study’s data obtained using paperbased surveys. 141 CHAPTER SIX: ANALYSES AND RESULTS OF THE MAIN STUDY (STUDY TWO) 6.1 Chapter overview This chapter gives details of Study Two which aims to validate the measurement models, test the hypotheses and address the research questions. Data screening and cleaning were performed based on the paper-based data collected in four Chinese cities (Section 6.2). Respondents’ profiles were analysed using demographic variables (Section 6.3). The measurement models were assessed using CFA (Section 6.4), along with a comprehensive measurement model evaluation and hypotheses testing using structural equation modelling (SEM) (Section 6.5). Additionally, control variables and important attributes of organic food purchase were examined, then a comparative study of the four cities was conducted using invariance tests (Section 6.6). Computer software applications PASW Statistics version 18 and AMOS (Analysis of Moment Structures) version18 were used to perform the analyses. Figure 6.1 diagrammatically illustrates the structure of Chapter Six. 142 Figure 6.1 Structure of Chapter Six Source: developed for this research 6.2 Data screening Prior to the analysis, the data was examined using various functions of SPSS version 18. It was checked for accuracy, missing values, fit between their distributions and the assumption of multivariate analysis. 6.2.1 Screening and cleaning missing data A total of 1117 paper-based survey responses were collected. The data was screened and cleaned using descriptive analysis for out-of-range values and missing values, and 143 frequency analyses were conducted for each variable to screen for out-of-range values. The maximum and minimum statistics were examined and extreme values were revisited and corrected using descriptive analysis. Questionnaires which had more than 25% missing data were discarded (Byrne 2010). This brought the total number of valid responses to 964. Out of the 52 statements included in the initial online survey of the pilot study, seven were eliminated during the process of EFA. Hence, only 45 statements were used in the paper-based survey. Cohen and Cohan (1983) suggest that as a rule of thumb, if more than 10% of the variables are missing, interpretation of the results could be problematic. Hence any completed survey which was found to have more than four statements missing was eliminated. There were 17 such cases, which reduced the number of usable surveys to 947. Since structural equation modelling (SEM) requires a complete data set (Kaplan 2009), there are various methods to handle missing data. One of the most popular methods is imputing missing data (Tabachnick & Fidell 2007). The statements associated with consumers’ purchase intention relating to organic food (Sections three and four of the survey) were imputed through the missing value function which uses the expectationmaximisation algorithm (EM). This process eliminated another ten cases, hence the final number of usable responses was 937. 6.2.2 Assessing reliability and normality As explained in Chapter Four, the reliability of measures is established by testing both consistency and stability. One such stability of measures can be assessed through testretest reliability (Cavana, Delahaye & Sekaran 2001). Field (2009) suggests that the simplest way to practise test-retest reliability is to use split-half reliability. This method randomly splits the data into two data sets. Cronbach’s alpha was used to measure internal consistencies of reliability for all constructs in this study. To ensure that all subjects came from the same population before treating them differently, random sampling techniques were used in SPSS (Tabachnick & Fidell 2007). Random sampling reliability analysis was repeated every time when an additional batch of 200 random 144 responses was added within the 937 cases of the paper-based data file. This technique was repeated to validate these measures based on factors obtained from the EFA in Study One. Table 6.1 illustrates the progressive reliability tests done for each batch of responses. The results reveal that the overall internal consistencies of the various batches of responses are acceptable, as the lowest Cronbach’s alpha obtained was 0.527 and the highest 0.835. This is within the acceptable limits in the early stage of research (Nunnally 1967). The results of these tests provide evidence that generalisation can be made to the population. Table 6.1 Progressive reliability scores for the ten factors Sample size Sensory appeal Value for money Regulatory Variety seeking Self-indulgence Opinion-leadership Ethnocentrism Personal beliefs/attitudes Pre-purchase evaluation Purchase intention 50 .557 .615 .529 .771 .746 .774 .814 .821 .571 .562 200 .649 .581 .624 .821 .788 .644 .840 .819 .708 .678 400 .680 .543 .586 .830 .746 .559 .835 .815 .700 .643 600 .680 .533 .577 .813 .758 .618 .817 .814 .692 .674 800 .683 .527 .561 .804 .758 .637 .810 .811 .691 .698 6.2.3 Outliers Models may fail to fit the data due to the presence of influential outliers or extreme data points (Cunningham 2008). Univariate outliers can be detected by visual checking of histograms and plots of individual variables (Pallant 2007). Multivariate outliers can be detected by visual inspection based on the ‘Mahalanobis distance (D) statistic, which indicates the distance in standard deviation units between a set of scores for an individual case and the same means for all variables’ (Kline 2005, p. 51). Cases with outliers can generate very different results, therefore outliers found to influence the result were deleted at every step of the CFA (see Section 6.4.1). Six cases with outliers were deleted; this resulted in a final count of 931 cases which comprised the entire data set of Study Two. Comrey and Lee (in Tabachnick and Fidell 2007) suggest that as a guide a sample size of 100 is poor, 200 is fair, 500 is very good and 1000 is excellent. 145 6.2.4 Assessing normality Screening continuous variable data for normality is important in the early stages of every multivariate analysis, and can be assessed by either statistical or graphical methods (Tabachnick & Fidell 2007). Normality means that the data is sampled from a normally distributed population (Allen & Bennett 2010). To assess the degree of normality, values of skewness and kurtosis can be examined. West, Finch and Curran (1995) suggest that absolute values of skewness and kurtosis exceeding 2 and 7 respectively are indicative of moderately non-normal distributions. Kline (2005) recommends that absolute kurtosis values greater than 10.0 are indicative of problematic non-normality, and values greater than 20.0 are indicative of serious deviations from multivariate normality. In addition, a standardised residual exceeding an absolute of 2.58 for larger models suggests that the model is unable to explain much of the covariation that exists between particular statement pairs (Byrne 2010). In this study, distribution assumption was used to assess normality. 6.3 Profile of respondents This section examines the profile of the final 931 respondents. All respondents had heard of the term ‘organic food’ and were at least 18 years old. Figure 6.2 illustrates recognition of the six selected logos used in the survey. Figure 6.2 Recognition of organic food logos 146 The results show that 89.8% of respondents recognised the ‘Chinese Green Food’ logo, 62% recognised the ‘Chinese Non-polluted Food’ logo, and only 58.3% recognised the ‘Chinese Organic Product’ logo. This is in line with Li et al.’s (2005) study which revealed that only 37.2% of respondents recognised ‘organic food’. The understanding of organic food in China seems to have improved in the last few years. With regards to imported organic food products, 9.1% of respondents recognised the USDA organic logo, which again confirms that Chinese consumers are more familiar with American products in comparison with other countries’ imported products. 6.8% recognised the German Bio organic logo. The lowest awareness rating was with the Australian Certificated Organic logo (ACO) which represented 5.2% of respondents. With respect to knowledge about organic food, 44.6% of respondents selected ‘I know a little about what ‘organic’ means’, 39.7% selected ‘I have heard of organic food, but am not sure what it means’, only 15.7% answered ‘I know a lot about organically produced food’, which again confirms the low awareness and understanding of organic food products. Figure 6.3 illustrates responses to the question ‘What is the first thing which comes to your mind when you think about organic food?’. Figure 6.3 First thing in mind when thinking about organic food 147 Most respondents answered ‘no pesticide’ (30.3%), followed by ‘something natural’ (18.2%) and ‘green’ (16.1%), ‘no additives’ (13.5%), ‘environmentally produced’ (13.9%), ‘free of GM contents’ (7.2%); 0.8% selected ‘others’, which included ‘healthy’, ‘healthier than non-polluted and green food’, ‘hormone free’, ‘using organic fertiliser’ and ‘no chemical pesticides’. 77.5% of respondents had purchased organic food. With respect to the last time of their purchase, 6.9% answered ‘more than a year ago’, 17.2% half a year ago, 30.4% a month ago, and 43.2% of respondents a week ago. Figure 6.4 illustrates the types of organic foods that respondents had recently purchased. Figure 6.4 Types of organic food recently purchased 66.9% of respondents had purchased organic fruit and vegetables, followed by 24.1% who had purchased grains, 21.3% organic dairy products, 15.4% organic meat and seafood, and 14.2% had purchased ‘others’ which included tea, beverages, and eggs. 148 Figure 6.5 illustrates the usage pattern for their most recent purchase of organic food. Figure 6.5 Usage of the most recently purchased organic food 42% of respondents purchased organic food for family use, 38% for own consumption, 14.6% for children, 6.7% elderly people, and 2.4% purchased organic food as gifts. Finally10.6% selected ‘others’, which could possibly include things such as ‘employer bonus’. Figure 6.6 illustrates how much people were willing to pay extra for organic food. Figure 6.6 Willing to pay extra for organic food When asked ‘How much extra are you willing to pay for organic food as compared to conventional food?’, 60.8% were willing to pay an extra 20% to 50%, and 9.4% an extra 149 51% to 100%. Only 0.8% were willing to pay in excess of an extra 100%. 29% said they ‘don’t want to pay extra’. The price of organic food seems to be an issue for Chinese consumers. Figure 6.7 illustrates the respondents’ preferred country of origin when they purchase imported organic food. Figure 6.7 Country of origin of imported organic food Surprisingly, 17.8% preferred organic food from Australia/New Zealand. This could possibly be owing to Australia being perceived as clean and green (Chang & Kristiansen 2006). 13.9% preferred European organic products, 7.9% preferred Japanese/Korean organic food, 6.4% preferred US organic products and 6.1% preferred organic food from Hongkong and Taiwan. Not surprising, nearly half (47.9%) did not mind which country or region the organic food came from. Figure 6.8 depicts respondents’ choice of distribution channels: 71.2% prefer to buy organic food in a supermarket, 10.3% from speciality stores, 9.3% from organic farms, 4.3% from local market, 2.1% prefer to purchase online, and only 0.4% prefer to purchase organic food through mail order catalogues. These findings confirm that ‘organic’ is still a new concept, and organic food products in China are mainly available in supermarkets in the larger cities. 150 Figure 6.8 Choice of distribution channel Figure 6.9 depicts the frequency of purchase of organic food products. 44.4% of respondents purchased ‘occasionally’, 27% ‘at least once a week’, 16.4% ‘every fortnight’ and 12.2% ‘monthly’. Figure 6.9 Frequency of purchase of organic food products 151 Figure 6.10 shows the monthly family income level of respondents: 43% earned less than RMB 5000, 33.8% earned between RMB 5,001 and RMB10,000, 16% earned between RMB 10,001 and RMB 20,000, and only 7.2% earned more than RMB 20,000. Figure 6.10 Monthly family income levels Types of household 21% of respondents were single, 19.2% were couples without children, 23.5% lived with parents, 22.4% lived with young children and 13.9 lived with adult children. Gender Although the administrators of the survey were instructed to seek out demographic variation where possible, the gender breakdown of respondents was 59.7% females and 40.3% males. Perhaps this can be explained by women taking more responsibility for purchasing food in the family, and going shopping more frequently than men. Age group 48.7% of respondents were between 18 and 30 years old, 35.3% were between 31 and 45, and 12.4% were between 46 and 60. Only 3.6% were older than 60, which indicates that younger consumers prefer to shop in supermarkets where organic food products are readily available. 152 Education level 10.1% of respondents had a postgraduate degree or above, 40.1% had a bachelor degree, 22% had a two years college or associate’s degree, 17.5% went to high school, and only 10.2% were educated less than high school. That is, more than half had higher education degrees, implying that most affluent people in China are aware of organic food products. Occupation 36% of respondents were white collar employees, 35.1% chose ‘others’, which could possibly be ‘teacher’, ‘public servant’, ‘doctor’, ‘owner of small business’, 13.9% were blue collar employees, 6.3% were not working and 8.6% were students. The demographic mix of respondents appears to concur with previous studies in relation to organic food consumers being both highly educated and high income earners (Tsakiridou et al. 2008). 6.4 Confirmatory factor analysis and measurement models 6.4.1 Confirmatory factor analysis A series of confirmatory factor analyses (CFA) were conducted on the data to validate the EFA results from Study One. CFA is like creating a structure and testing an objective against the structure to see how well it fits. In comparison to EFA, CFA is a more complicated and sophisticated technique, used in the advanced states of the research process to test the theory about latent processes or to test specific hypotheses usually performed through SEM (Pallant 2007, Tabachnick & Fidell 2007). To evaluate the measurement models for each construct, covariance structure analysis was carried out using the software application AMOS version 18. 6.4.2 Maximum Likelihood Maximum likelihood (ML) is the most widely used fitting function for general SEM (Bollen 1989). Hu and colleagues (in Tabachnick & Fidell 2007) found that when the 153 normality assumption was reasonable, the ML performed well with sample sizes over 500. In this study, the paper-based survey generated about 1000 cases, therefore ML estimation was chosen to provide parameter estimates and standard errors that are asymptotically unbiased, consistent and efficient (Bollen 1989). (Tarkiainen & Sundqvist (2005) also adapted this technique to avoid the problem with missing values when examining organic food consumers’ purchase intention, by using SEM. ML estimation was employed to estimate all models. 6.4.3 Measurement model fit indices The primary fit index used in measurement models is the Chi-square (χ2) test, which is used to determine whether the data supports the model. A number of other fit indices are also used to test the overall model. This section describes absolute goodness-of-fit statistics used in this study. Some of the fit indices used for the CFA are described below: x Chi-square value The model evaluation was assessed by χ2 and its accompanying significance test. If the associated p value is not significant (p>0.05), which means there is no significant difference between the sample variance/covariance matrix and the model-implied variance/covariance matrix, it can be assumed the data fits the model well. However, Kline (2005) states that χ2 is sensitive to sample size. The larger the sample size, the more likely the p-value associated with the χ2 will result in a significant difference between the model and the data, therefore this study also takes into account the normed chi-square value. To reduce the sensitivity of χ2 to sample size, the values of χ2/ df was assessed. According to Bollen (in Kline 2005), the values of χ2/df of two, three, even as high as five have been recommended as indicative of reasonable fit. The column CMIN/DF is the χ2 value divided by the df, and is referred to as the normed χ2. Bollen-Stine bootstrap adjusts for the non-normality in the data, and can be used to assess the model fit. The bootstrapping technique produces a new chi-square distribution from which an adjusted p-value is computed (Cunningham 2008). 1000 bootstrapping samples 154 were generated in this study, producing the chi-square distribution that corrects for departure from multivariate normality (pa = Bollen-Stine bootstrap p value). x Goodness-of-fit index (GFI) and adjusted goodness-of-fit index (AGFI) According to Kline (1994, p. 97) ‘GFI is the ratio of the sum of the squared discrepancies to the observed varices’. GFI=1.0 indicates perfect model fit, and GFI>0.90 indicates good fit (Kline 2005). The AGFI is the GFI adjusted for the degree of freedom of the model relative to the number of variables. Both the GFI and the AGFI with values exceeding 0.95 are considered to be indications that the data fits the model well (Cunningham 2008). x Tucker Lewis index (TLI) There are two commonly used incremental fit indices widely used in SEM, known as TLI and comparative fit index (CFI). The TLI was originally proposed by Tucker Lewis as early as 1973. It indicates the improvement per degree of freedom of the target model over an independent model. It has a range from 0 to 1, where values exceeding 0.95 indicate good model fit (Hu & Bentler 1998). However, the TLI can frequently exceed a value of 1, especially for over-fitting models (Cunningham 2008). x Comparative fit index (CFI) CFI is derived from the χ2, and it measures an improvement on the normed fit index (NFI). The CFI is the most widely used indicator, and has a range from 0 to 1 (Hair et al. 2006; Kline 2005). Hu and Bentle (1998) suggest that as a rule of thumb, CFI values greater than roughly 0.90 may indicate a reasonably good fit of the model. A value of 0.95 is preferable (Kline 2005) and was applied in this study. x Root-mean-square error of approximation (RMSEA) RMSEA is strongly recommended by many statisticians since the availability of its confidence interval provides important information about the precision of the estimate of fit (Browne & Cudeck 1992; Hu & Bentler 1998; Kline 2005). It measures the 155 discrepancy per degree of freedom, with values of about 0.05 or less indicative of close fit of the model to the data, and with values between 0.05 and 0.08 indicative of reasonable fit (Browne & Cudeck 1992). Hair et al. (2006) claim that RMSEA values for 90% confidence interval (90% CI) which range between 0.03 and 0.08 are preferable. x Standardised root-mean-square residual (SRMR) The SRMR is a measure of the average difference between corresponding elements of the sample and model-implied correlation matrices. A value of less than 0.05 suggests good fit. Large value of the SRMR indicates outliers in the data (Cunningham 2008). Table 6.2 summarises all model fit indicators adopted in this study. The model fit evaluates various ranges of model fit indices in AMOS version 18. Table 6.2 Model fit indicators adopted in this study Indicator CMIN/df P-value or pa (BollenStine bootstrap p value) GFI/AGFI TLI CFI RMSEA SRMR Definition Value used in this study Value of normed χ2 No significant difference below 5.0 >0.05 Goodness-of-fit index, adjusted goodness-of-fit index Tucker Lewis index Comparative fit index Root-mean-square error of approximation Standardised root-meansquare residual >0.95, or even >0.90 >0.95, sometimes >1.00 >0.95, sometimes only >0.90 <0.05, sometimes <0.08 <0.05 The re-specified structural equation model should not only rely on statistical information, but also on the theoretical reasoning/judgements, as theory is at the centre of any respecification (Anderson & Gerbing 1988). The results of this study are based on both theoretical and statistical indications. AMOS produces diagrams such as those displayed in the figures that follow. Circles embody the error terms associated with each attribute or statement. Rectangles symbolise the indicator variables or statements. Ovals represent the latent or unobserved factors and arrows indicate the relationships between the variables. 156 6.4.4 Results of CFA 6.4.4.1 CFA for the product-related construct The two components of the product-related construct, that is, ‘sensory appeal’ and ‘price perception’, were initially linked to form a two factor CFA measurement model as shown in Figure 6.11a. Figure 6.11a A two factor CFA model of the product-related construct Chi-square = 73.084, df=13, pa = .001, CMIN/DF=5.622, GFI=.978, AGFI=.954, TLI=.902, CFI=.940, RMSEA=.070, 90%CI=(.055,.086), SRMR=.0403 As can be seen, the correlation between the two components was (0.83) and there was a lack of discriminant validity (see Appendix 5, Table A). As can also be seen from the results, the data did not fit the model well. This suggests that the two factors are strongly co-related and are measuring a similar construct. Hence, the statements of the two components were combined to form a one factor solution. Statements 3.7 and 3.9 (statements associated with price perception) were removed in order to improve the model fit, and the final best fit congeneric measurement model is shown in Figure 6.11b. 157 Figure 6.11b A re-specified one factor CFA model for the product-related construct Chi-square=16.690, df=5, pa=.160, CMIN/DF=3.338, GFI=.993, AGFI=.978, TLI=.971, CFI=.985, RMSEA=.050, 90%CI= (.025, .077), SRMR=.0224 As shown in the results, pa >0.05 represents a non-significant difference between the model and the data. Although it is desirable for the factor loadings to be higher, the threshold factor loading in the CFA stage was based on the minimum level of .30 for interpretation of structure (Hair et al. 2006; Peterson 2000). Hair et al. (2006) suggest that unidimensionality means a set of measured items with only one underlying latent construct. The Eignevalues (1.409 0.521 0.454 0.331 0.298) shows a unidimensional characteristic with the re-specified measurement model. The GFI and AGFI represent a good approximation of the data, with the RMSEA <0.08. The value of the CFI is close to one, which suggests that it is a well-fitting model. The construct reliability (α) for the product-related construct was 0.715. The regression weights for the statements in the final product-related measurement model are shown in Table 6.3. 158 Table 6.3 Regression weights for the product-related construct Estimate Product_ 1 Product_ 2 Product_ 3 Product_ 5 Product_ 6 <-<-<-<-<-- PRODUCT_RELATED PRODUCT_RELATED PRODUCT_RELATED PRODUCT_RELATED PRODUCT_RELATED SE CR P 1.000 Logo Pleasant texture 1.011 .074 13.694 *** Looks nice 1.013 .074 13.766 *** Smells nice .913 .075 12.181 *** Is trendy .775 .073 10.568 *** High nutritional value Note: ***p<.001 There are five statements that indicate a significant relationship. One is the weight assigned to statement 1 in order to obtain a solution. The weights for the last two statements were less than one, but still significant at p< 0.001 level (details are shown in Appendix 5, Table A). 6.4.4.2 CFA for the regulatory construct The statements in this construct were used to form a one factor model as shown in Figure 6.12a. 159 Figure 6.12a A one factor CFA model of the regulatory construct Chi-square=38.127, df=9, p =.000, CMIN/DF=4.236, GFI=.987, AGFI=.970, TLI=.895, CFI=.937, RMSEA=.059, 90%CI=(.040, .079), SRMR=0.345 As can be seen from the results above, the data did not fit the model well as evidenced by the results CMI/DF=4.236, TLI (0.895) and CFI (0.937) being less than 0.95. Hence statement 3.11 (I don’t trust organic food certification bodies) was removed in order to improve the model fit, and the final best fit congeneric measurement model is shown in Figure 6.12b. 160 Figure 6.12b A re-specified one factor CFA model for the regulatory construct Chi-square=8.065, df=5, p=.153, CMIN/DF=1.613, GFI=.997, AGFI=.990, TLI=.986, CFI=.993, RMSEA=.026, 90%CI=(.000,.057), SRMR=.0183 As shown in the results above, both the GFI and AGFI represent a good approximation of the data, with RMSEA <0.05. The value of CFI is close to one, which suggests that it is a well-fitting model. The construct reliability (α) for the regulatory construct was 0.601. The regression weights for the statements in the final regulatory measurement model are shown in Table 6.4. Table 6.4 Regression weights for regulatory measures Regulatory_8 <--- REGULATION Estimate 1.000 S. CR P Regulatory_7 <--- REGULATION .844 .098 8.649 *** Regulatory_6 <--- REGULATION 1.242 .122 10.182 *** Regulatory_5 <--- REGULATION .976 .103 9.493 *** Regulatory_1 <--- REGULATION .544 .078 6.990 *** Logo Market is chaotic More land should be organic Logo should be control Lack government control Read the label Note: ***p<.001 There are five statements that indicate a significant relationship. One is the weight assigned to statement 1 in order to obtain a solution. The weights for the other four variables are significant at p< 0.001 level (details are shown in Appendix 5, Table B). 161 6.4.4.3 CFA for lifestyle construct The three factors of this construct, i.e. variety seeking, self-indulgence and opinionleadership, were initially linked to form a three factor CFA measurement model as shown in Figure 6.13a. Figure 6.13a Initial three factor model for the lifestyle construct Chi-square=221.323, df=41, pa= .001, CMIN/DF=5.398, GFI=.958, AGFI=.932, TLI=.923, CFI=.943, RMSEA=.068, 90%CI=(.060,.077), SRMR=.0448 As can be seen from the results, the data did not fit the model well. Consequently statement 3.35 ‘It does not hurt to be trendy if I feel like it’ was deleted in order to improve the model fit, and the final best fit congeneric measurement model is shown in Figure 6.13b. 162 Figure 6.13b A re-specified three factor model for the lifestyle construct Chi-square=125.621, df=31, pa = .001, CMIN/DF=3.926, GFI=.973, AGFI=.954, TLI=.952, CFI=.966, RMSEA=.056, 90%CI=(.046,.067), SRMR=.0424 In CFA when the resulting latent variables are moderately to highly correlated, higherorder factors might be hypothesised as an explanation of the correlations that exist amongst the lower-order factors (Cunningham 2008). The three lower-order factors of variety seeking, self-indulgence and opinion-leadership were moderately intercorrelated with values ranging from a low of 0.33 to a high of 0.52. Related theory suggests that this model might be a higher-order model. Kline (2005) recommends that that a higher-order CFA model must have at least three first-order factors, and each first-order factor should 163 comprise at least two indicators. The model in Figure 6.13c satisfies all these requirements, therefore a higher-order factor is suggested. Figure 6.13c Higher-order CFA for the lifestyle construct Chi-square=125.621, df=31, pa = .001, CMIN/DF=3.926, GFI=.973, AGFI=.954, TLI=.952, CFI=.966, RMSEA=.056, 90%CI=(.046,.067), SRMR=.0424 Cunningham (2008) suggests that lower-order and higher-order models are equivalent; however, the higher-order factors must be established by relevant theory, which in this case is done using Yang’s (2004) study. The reliabilities of the three factors were: variety seeking 0.782, self-indulgence 0.766 and opinion-leadership was 0.630. Figure 6.13c depicts the results of the higher-order CFA. The two models have the same results and produce the same fit statistics. The regression weights for the statements in the final lifestyle measurement model are shown in Table 6.5. 164 Table 6.5 Regression weights for lifestyle measures Lifestyle_9 Lifestyle_7 Lifestyle_6 Lifestyle_5 Lifestyle_3 Lifestyle_2 <--<--<--<--<--<--- VARIETY-_SEEKING VARIETY-_SEEKING VARIETY-_SEEKING VARIETY-_SEEKING SELF-_INDULGENCE SELF-_INDULGENCE Estimate 1.000 1.425 1.443 1.341 1.000 1.300 Lifestyle_1 <--- SELF-_INDULGENCE 1.220 .072 16.866 *** OPINION-_LEADERSHIP OPINION-_LEADERSHIP OPINION-_LEADERSHIP 1.000 1.158 .802 .092 .076 12.593 10.530 *** *** Lifestyle_11 <--Lifestyle_10 <--Lifestyle_12 <--- SE CR P .093 .094 .095 15.312 15.301 14.144 *** *** *** .077 16.878 *** Logo Influenced Love trendy Try new Impulse purchase Spend it Do feel like it Without deliberation Consult me Influence people Successful Note: ***p<.001 There are ten statements which are significant. One was the weight assigned to three statements (lifestyle 9/lifestyle 3/lifestyle 11) in order to obtain a solution. The weight for the last statement was less than one, but still significant at p<0.001 level. Weights for the rest of the statements were all significant at p<0.001 level (details are shown in Appendix 5, Table C). 6.4.4.4 CFA for the ethnocentrism construct A one factor model should ideally have at least three or four indicators, with the minimum requirement being three indicators (Bollen 1989; Kline 2010). Three indicators (statements) were taken from the EFA results of Study One, and all these have been retained as shown in Figure 6.14. The result χ2 (1) =0.132, p=0.717 indicates that the data fits the model well. The model also generated other indicators, GFI=1.000, AGFI=0.9999, TLI=1.000, CFI=1.000, all of which exceed 0.95, indicating that this is a well-fitting model. The reliability of the ethnocentrism construct was 0.812 indicating good internal consistency. 165 Figure 6.14 CFA for ethnocentrism measures Chi-square =.132, df=1, p= .717, CMIN/DF=.132, GFI=1.000, AGFI=.9999, TLI=1.000, CFI=1.000, RMSEA=.000, 90%CI=(.000,.062), SRMR=.0020 Table 6.6 shows the regression weights for the statements in the ethnocentrism measurement model. There are three statements that are significant. One is the weight assigned to statement 1 in order to obtain a solution. The weights for the last two statements were greater than one, but still significant at p< 0.001 level (details are shown in Appendix 5, Table D). Table 6.6 Regression weights for ethnocentrism measures Estimate SE CR P Logo Should not buy Levy tariff Buy domestic Ethnocentric_5 <--- ETHNOCENTRISM 1.000 Ethnocentric_2 <--- ETHNOCENTRISM 1.021 .046 22.260 *** Ethnocentric_1 <--- ETHNOCENTRISM 1.021 .046 22.260 *** Note: ***p<.001 6.4.4.5 CFA for the beliefs/attitudes construct The ten statements of this construct were initially linked to form a one factor CFA measurement model as shown in Figure 6.15a. The beliefs and attitudes construct 166 included statements related to consumers’ personal beliefs and attitudes towards organic food purchase. Figure 6.15a: One factor CFA model for the beliefs/attitudes construct Chi-square =282.557, df=35, pa = .001, CMIN/DF=8.073, GFI=.941, AGFI=.907, TLI=.853, CFI=.886, RMSEA=.087, 90%CI=(.078, .097), SRMR=.0522 As can be seen from the results, the data did not fit the model well. Hence, five statements were removed (there were large residuals in these five statements) in order to improve the model fit, and the final best fit congeneric measurement model is shown in Figure 6.15b. 167 Figure 6.15b: A re-specified CFA for the beliefs/attitudes construct Chi-square =22.781, Df=5, pa= .041, CMIN/DF=4.556, GFI=.990, AGFI=.970, TLI=.963, CFI=.981, RMSEA=.062, 90%CI=(.038,.089), SRMR=.0253 As shown in the results, the GFI and AGFI represent a good approximation of the data, with RMSEA <0.08. The value of CFI is close to one, which suggests that it is a wellfitting model. The construct reliability (α) for this construct was 0.748. The regression weights for the statements in the final measurement model are shown in Table 6.7. Table 6.7 Regression weights for beliefs/attitudes measures Personal_3 Personal_6 Personal_7 Personal_8 <--<--<--<--- BEL/ATT BEL/ATT BEL/ATT BEL/ATT Personal_10 <--- BEL/ATT Estimate 1.000 1.165 1.589 1.491 S.E. C.R. P .101 .123 .120 11.556 12.913 12.475 *** *** *** 1.179 .104 11.362 *** Logo Like organic Good for environment Believe the quality Should always buy Labels mean high quality Note: ***p<.001 There are five statements that are significant. One is the weight assigned to statement 3 in order to obtain a solution. The weights for the other variables were significant at p < 0.001 level (details are shown in Appendix 5, Table E). 168 6.4.4.6 CFA for the pre-purchase evaluation construct The five statements of the pre-purchase evaluation construct were initially linked to form a one factor CFA measurement model as shown in Figure 6.16a. Figure 6.16a: One factor CFA model for the pre-purchase evaluation construct Chi-square =48.233, df=5, pa= .001, CMIN/DF=9.647, GFI=.981, AGFI=.944, TLI=.900, CFI=.950, RMSEA=.096, 90%CI=(.072,.122), SRMR=.043 As can be seen from the results, the data did not fit the model well. Statement 4.5 (I can recall the brand names and labelling of some of the organic food products) was deleted in order to improve the model fit, and the final best fit congeneric measurement model is shown in Figure 6.16b. 169 Figure 6.16b A re-specified CFA for the pre-purchase evaluation construct Chi-square =13.953, df=2, pa= .027, CMIN/DF=6.977, GFI=.992, AGFI=.962, TLI=.952, CFI=.984, RMSEA=.080, 90%CI=(.044,.122), SRMR=.0255 As shown in the results, GFI and AGFI represent a good approximation of the data, with RMSEA ≤0.08. The value of CFI is close to one, which suggests that it is a well-fitting model. The construct reliability (α) for this construct is 0.695. The regression weights for the statements in the final measurement model are shown in Table 6.8. Table 6.8 Regression weights for pre-purchase evaluation measures Note: ***p<.001 There are four statements that are significant. One is the weight assigned to statement Pre _ purchase _1 in order to obtain a solution. The weights for the other three variables were significant at p<0.001 level (details are shown in Appendix 5, Table F). 170 6.4.4.7 CFA for the behavioural/purchase intention construct Three indicators (statements) were taken from the results of the EFA in Study One. As can be seen, all the statements were retained and the data fitted the model well. The final congeneric measurement model is shown in Figure 6.17. Figure 6.17 CFA for behaviour/purchase intention construct Chi-square =.343, df=1, p= .558, CMIN/DF=.343, GFI=1.000, AGFI=.999, TLI=1.004, CFI=1.000, RMSEA=.0000, 90%CI=(.000,.072), SRMR=.0049 As shown in the results, GFI and AGFI represent a good approximation of the data, with RMSEA <0.05. The value of CFI was one, which suggests that it is a well-fitting model. The construct reliability (α) for this construct was 0.696 which indicates good internal consistency. The regression weights for the statements in this measurement model are shown in Table 6.9. Table 6.9 Regression weights for behavioural/purchase intention Be_intention_3 <--- INTENTION Be_intention_2 <--- INTENTION Be_intention_1 <--- INTENTION Estimate SE 1.085 .078 1.085 .078 1.000 CR P 13.966 *** 13.966 *** Logo Make same choice Recommend Probably use Note: ***p<.001 171 There are three statements that are significant. One is the weight assigned to statement Be_intention_1 in order to obtain a solution. The weights for the other statements were significant at p < 0.001 level (details are shown in Appendix 5, Table G). 6.4.5 Summary of the results of CFA The factors obtained from the EFA in Study One were validated using CFA, and a total of ten statements were deleted due to model fit requirements. Thirty-five statements comprising seven measurement models (or ten factors) were retained (see Appendix 5). Most of the modifications contributed adequately towards model fit requirements in the re-specified models. Since the seven measurement models would later form a structural equation model, it was important to test their reliability. Cronbach’s alpha was adopted to test their internal consistencies. Table 6.10 summarises the reliability scores of the seven measurement models. The scores range from a high of 0.827 to a low of 0.570. These are acceptable and will be further tested in the final SEM. Table 6.10 Summary of reliability of the measurement models Measurement model Product-related Regulatory Lifestyle (three factors) Ethnocentrism Beliefs/attitudes Pre-purchase evaluation Behavioural/purchase intention 6.5 Number of statements retained in CFA 5 5 10 3 5 4 3 Reliability (Cronbach’s alpha) .715 .601 α 1=.782 α 2=.766 α 3=.630 .809 .747 .693 .689 Full structural equation model Before conducting the hypotheses tests, convergent validity and discriminant validity were ascertained as follows. 172 6.5.1 Convergent validity Convergent validity refers to the extent to which multiple attempts measuring the same concept with different methods are in agreement (Hair et al. 2006). It refers to the degree in which items strongly load on one single factor. In confirmatory factor analysis, researchers can establish convergent validity within the context of the measurement model (Steenkamp & van Trijp 1991). In this study, CFA was used to assess the convergent validity of the constructs in two ways. Firstly, measurement models for each construct were examined in terms of the significance and strength of the item loadings. All items had significant factor loadings above 0.30. Secondly, modification indices (which find where the fit in a model could be improved) were used to identify possible cross-loading items (Straub, Boudreau & Gefen 2004). Convergent validity is confirmed when items have significant factor loadings above 0.30, and load only on the a priori theoretical construct 6.5.2 Discriminant validity In order to confirm how well the results obtained from the use of measures fit the theories around which this study was designed, discriminant validity tests were conducted based on the measurement models (Cavana, Delahaye & Sekaran 2001). A set of variables presumed to measure different constructs shows discriminant validity if their intercorrelations are not too high (Kline 2010, p. 72). Pattern and structure coefficients are used in determining whether constructs in measurement models are empirically distinguishable (Cunningham 2008; Thompson 1997). The discriminant validity was assessed in two sections of the proposed conceptual model: Section one (Figure 6.18) examined the influencing stage which comprised endogenous constructs and the results are shown in Table 6.11. This test clearly identifies that there are five constructs, namely, product, regulatory, self-indulgence, variety seeking and ethnocentrism. Essentially, discriminant validity was established by eliciting five distinctly different constructs which are not highly correlated to each other (Cavana, Delahaye & Sekaran 2001). 173 Figure 6.18 Endogenous constructs of the influencing stage of the conceptual model 174 Table 6.11 Factor pattern and structure coefficients for the endogenous constructs of the influencing stage Indicator statements Product_1 Product_2 Product_3 Product_5 Product_6 Regulatory_5 Regulatory _6 Regulatory_7 Regulatory_8 Lifestyle_1 Lifestyle_2 Lifestyle_3 Lifestyle_6 Lifestyle_7 Lifestyle_9 Ethnocentric_1 Ethnocentric_2 Ethnocentric_5 Product P .616 .659 .641 .532 .444 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* S .616 .659 .641 .532 .444 .142 .173 .122 .146 .139 .145 .106 .182 .190 .115 .100 .099 .092 Regulatory P 0* 0* 0* 0* 0* .522 .635 .449 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* S .168 .179 .174 .145 .121 .522 .635 .449 .534 .034 .036 .026 .100 .104 .063 .036 .036 .033 Self_ indulgence P S 0* .108 0* .116 0* .113 0* .094 0* .078 0* .023 0* .028 0* .020 0* .023 .788 .788 .826 .826 .602 .602 0* .365 0* .381 0* .230 0* .196 0* .194 0* .180 Variety seeking P S 0* .142 0* .152 0* .148 0* .123 0* .103 0* .066 0* .080 0* .057 0* .068 0* .365 0* .383 0* .279 .788 .788 .823 .823 .497 .497 0* .140 0* .139 0* .128 Ethnocentrism P 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* 0* .792 .782 .725 S .078 .083 .081 .067 .056 .024 .029 .020 .024 .195 .205 .149 .140 .146 .088 .792 .782 .725 Note: P=Pattern coefficients. S=Structure coefficients. Tabled values are standardised parameter estimates. Asterisked values are parameters fixed at reported levels to identify the model (Appendix 6: Table A) From Table 6.11, it is evident that the items purported to compose each individual construct have significantly higher loadings compared to the other items that are not believed to make up that construct. Section two (Figure 6.19) assessed the exogenous construct of the conceptual model, i.e. cognitive/affective, evaluation of alternatives and behavioural Intentions, and the results are shown in Table 6.12. The factor pattern and structure coefficients matrix demonstrate that there are three distinctly different latent constructs, namely, beliefs/attitudes, prepurchase evaluation, behavioural/purchase intention which display discriminant validity. Appendix 6 provides details of the results of discriminant validity for the entire structural equation model. 175 Figure 6.19 Exogenous construct of the cognitive/affective, evaluation of alternatives and behavioural/purchase intention stages of the conceptual model A higher-order construct is suggested when the correlations between constructs are high and when discriminate validity is lacking (Cunningham 2008). Although the correlations between the three constructs are high, discriminate validity was evident as shown in Table 6.12, hence it was deemed unnecessary to test for a higher-order construct. 176 Table 6.12 Factor pattern and structure coefficients for the exogenous constructs of beliefs/attitudes, evaluation of alternatives and behavioural/purchase intention Indicator statements Beliefs/attitudes Pre-purchase evaluation Behavioural intention P S P S P S BEL/ATT_6 .606 .606 0* .293 0* .347 BEL/ATT_7 .760 .760 0* .368 0* .435 ATT/BEL_8 .634 .634 0* .307 0* .363 ATT/BEL_10 .554 .554 0* .268 0* .317 Prepurchase _2 0* .316 .652 .652 0* .503 Prepurchase _3 0* .337 .696 .696 0* .537 Prepurchase _4 0* .371 .767 .767 0* .591 Be_intention_1 0* .383 0* .515 .668 .668 Be_intention_2 0* .374 0* .503 .652 .652 Be_intention_3 0* .366 0* .493 .640 .640 Note: P=Pattern coefficients. S=Structure coefficients. Tabled values are standardised parameter estimates. Asterisked values are parameters fixed at reported levels to identify the model (Appendix 6: Table B). 6.5.3 Establishing a full structural model Some studies (Grunert & Juhl 1995; McEachern & Willock 2004) suggest that it would be advantageous to adopt SEM in future research relating to organic consumers’ attitudes and motivations, as it would create the capacity to scrutinise a series of dependent relationships concurrently. Previous studies (Saba & Messina 2003; Tarkiainen & Sundqvist 2005) have demonstrated that this is a relatively strong method to analyse consumers’ purchase behaviour of organic food. It allows simultaneous analysis of the relationships between dependent (DV) and independent variables (IV) in the organic food behavioural intention model (de Magistris & Gracia 2008). Tabachnick and Fidell (2007, p. 676) describe the process of SEM as: Structural equation modelling (SEM) is a collection of statistical techniques that allow a set of relationships between one or more IVs, either continuous or discrete, and one or more DVs, either continuous or discrete, to be examined. 177 Confirmatory factor analysis is actually a special type of Structural equation modelling. SEM is a technique used to evaluate the goodness-of-fit of the hypothesised model to the sample data in order to offer support for the theoretical model (Cunningham 2008). It confirms the interrelationship between sets of observed indicator variables and factors of the proposed theoretical model, assesses the convergent and discriminate validities of various statements used in this study, and provides suitable indicators to test consumers’ purchase intention associated with organic food in urban China. Table 6.13 summarises the techniques used to test the hypotheses developed in Chapter Three. Table 6.13 Summary of techniques used to test hypotheses Research questions Research question one Related hypotheses Hypothesis 1a/b/c What are the factors that influence Hypothesis 2a/b/c urban Chinese consumers’ Hypothesis 3a/b/c beliefs/attitudes towards the purchase of organic foods, and how Hypothesis 4a/b/c they impact on the pre-purchase Hypothesis 5 Techniques used to test Study One: Exploratory factor analysis (EFA) Study Two: Confirmatory factor analysis (CFA) and structural equation modelling (SEM) evaluation and behavioural/purchase intention? Hypothesis 6 Research question two Hypothesis 7a Evaluate the role of demographic Hypothesis 7b variables and usage patterns in the Study Two : T-test and analysis of variance (ANOVA) Hypothesis 7c behavioural /purchase intentions of urban Chinese consumers towards Hypothesis 7d organic food. Hypothesis 7e 6.5.4 The output of the final model A structural model aims to specify which latent variables directly or indirectly influence the other latent variables in a model (Bollen 1989). The structural model was tested and 178 presented as the final stage of analysis. Post-hoc modifications were performed in an attempt to develop a better fitting and possibly more parsimonious model. The hypothesised structural model was tested based on validated measures derived from CFA in Study Two, and guided by the proposed conceptual framework developed in Chapter Three (Figure 3.2). Causal flows linked the influencing stage, namely, product, regulatory, lifestyle, and ethnocentrism to consumers’ beliefs/attitudes, pre-purchase evaluation and later to behavioural/purchase intention. Meanwhile a causal path linked beliefs/attitudes to pre-purchase evaluation and later to behavioural intentions. Furthermore, there was a direct path between beliefs/attitudes and behavioural intentions. These paths occur with multiple regression relationships corresponding to the hypotheses developed in the proposed conceptual framework (details are shown in Appendix 7, Tables A to D). Three extra links were added to improve the model fit. Firstly, the product-related construct significantly influenced the regulatory construct as evidenced by the standardised direct effect of 0.282 (p value was significant). This essentially means that when the product-related construct increased by 1 standardised deviation, the regulatory construct increased by 0.282 standardised deviation. Secondly, the lifestyle construct significantly influenced the product-related construct (p value was significant) as evidenced by the standardised direct effect of 0.288. This essentially means that when the lifestyle construct increased by 1 standardised deviation, the product-related construct increased by 0.288 standardised deviation. Thirdly, the beliefs/attitudes construct significantly influenced behavioural/purchase intention as evidenced by the standardised direct effect of 0.232 (p value was significant). One statement from the regulatory construct, i.e. ‘Regulatory_1’(loading =.32), and one from pre-purchase evaluation construct, i.e. ‘Pre-purchase evaluation_1’(loading = .35) were deleted due to low factor loadings. The overall factor loadings of the remaining statements ranged from 0.480 to 0.825 (Appendix 7, Table E). Statement Lifestyle_5 and beliefs/attitudes statement personal _3 were removed due to large residuals. The entire factor of opinion-leadership (three statements) was removed as it was distorting the whole model, and its Cronbach’s alpha was relatively low (α=0.633). Hence, in the final best-fit model four statements and one factor (comprising three statements) were removed 179 from the original seven measurement models. Figure 6.20 illustrates the final best-fit model. Latent variables are represented by circles and measured variables are represented by squares. In this study, all latent variables of the measurement models were validated and the re-specified models fitted well. The output of the Squared Multiple Correlations (Appendix 7, Table F) shows that the final model explains 41.3% of the variance in ‘attitudes’, 39.5% of the variance in ‘pre-purchase evaluation’ and 66.4% of the variance in ‘behavioural intention’. 180 CMIN/DF=2.813, p= .001, GFI=.931, AGFI=.915, TLI=.905, CFI=.916, RMSEA=.044, 90%CI=(.041,.048), SRMR=.0497 Figure 6.20 Final best-fit model of consumer purchase intention of organic food in urban China 181 6.5.5 Assessing the reliability of the final model Table 6.14 summarises the progression of how the 52 statements of the original survey were reduced to 28 statements in the final best-fitting model. It demonstrates that the highest Cronbach’s alpha was 0.809, and the lowest 0.611. Although ideally, the Cronbach’s alpha should be above 0.7, they are quite sensitive to the number of items in the scale. With scales of fewer than ten items, it is common to find low Cronbach values (Pallant 2007). This is the case for the ‘regulatory’ construct which has a Cronbach’s alpha of 0.611. Therefore, the internal consistencies of all scales in this study are acceptable and adequate. Table 6.14 Summary of reliability of the final model scales Original number of statements Number of statements deleted in EFA Number of statements deleted in CFA Number of statements deleted in final model Number of statements retained in final model Scale reliability in final model (Cronbach’s alpha) Productrelated Regulatory 9 2 2 Nil 5 .710 8 2 1 1 4 .611 Lifestyle (two factors) 12 1 1 4 6 α 1=.733 α 2=.775 Ethnocentrism 5 2 Nil Nil 3 .809 Beliefs/ attitudes Pre-purchase evaluation 10 Nil 5 1 4 .730 5 Nil 1 1 3 .750 Behavioural/ purchase intention Total statements 3 Nil Nil Nil 3 .689 52 7 10 7 28 Construct 6.5.6 Hypotheses tests Figure 6.21 illustrates a simplified final validated best-fit model with relevant hypotheses. Lines with arrows indicate hypothesised direct relationship between the different constructs (latent variables). This simplified model depicts only the hypotheses 182 which were eventually accepted. Although the items associated with ‘Ethnocentrism’ were retained in the final best-fit model, the hypothesis associated with this construct was not accepted, hence it is not included in the simplified model. Figure 6.21 Simplified final best-fit model Table 6.15 summarises the results (Appendix 7, Table A and Table D) of the hypotheses tests. A version of ‘t’ test was employed which uses critical ratios from the SEM. Standard errors are shown in the SE column, and the CR column stands for critical ratio (magnitude > 2 indicates statistical significance at the .05 level). P value indicates statistical significance at levels of 0.001, 0.01 and 0.05 respectively. The direction and importance of the relationships is determined by the magnitude of Beta weights. 183 REGULATORY <--<--<--<--<--<--<--<--<--<--<--- PRE_-PURCHASE BEHAVIOUR_INTENTION BEL/ATT PRE_-PURCHASE BEHAVIOUR_INTENTION BEL/ATT PRE_-PURCHASE BEHAVIOUR_INTENTION PRE_-PURCHASE BEHAVIOUR_INTENTION H2a H2b H2c H3a H3b H3c H4a H4b H4c H5 H6 Note: ***p<.001; **p <.01; *p <.05 PRODUCT <--- BEHAVIOUR_INTENTION BEL/ATT H1c PRE_-PURCHASE BELIEFS/ATTITUDES ETHNOCENTRISM ETHNOCENTRISM ETHNOCENTRISM LIFESTYLE LIFESTYLE LIFESTYLE REGULATORY REGULATORY PRODUCT <--- PRE_-PURCHASE H1b PRODUCT <--- BEL/ATT H1a Hypothesis No. .566 .223 .011 .027 .024 .222 .119 .404 .075 .464 .294 -.085 .264 .450 Estimate .058 .068 .026 .029 .026 .095 .105 .098 .061 .066 .053 .062 .071 .066 SE 9.729 3.257 .420 .922 .926 2.347 1.131 4.111 1.234 7.066 5.567 -1.363 3.695 6.820 CR Table 6.15 Summary of hypotheses tests (H1a-H6) *** ** .674 .356 .355 * .258 *** .217 *** *** .173 *** *** P Accepted Accepted Rejected Rejected Rejected Accepted Rejected Accepted Rejected Accepted Accepted Rejected Accepted Accepted Accepted/ rejected .621 .196 .016 .037 .038 .127 .062 .239 .067 .379 .274 -.069 .197 .383 Beta 184 6.5.7 Discussion relating to the hypotheses tests of SEM Hypothesis one H1a The product-related construct has a positive influence on the beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products The results (refer to Table 6.15) revealed that the product-related construct significantly influenced consumers’ beliefs/attitudes (p value was significant). The standardised direct effect of the product-related construct on beliefs/attitudes was 0.383, which means that, when this construct increases by 1 standardised deviation, beliefs/attitudes increases by 0.383 standardised deviation (Cunningham 2008). Thus, hypothesis H1a is accepted. H1b Organic food product-related construct is positively correlated to Chinese consumers’ pre-purchase evaluation The results (refer to Table 6.15) revealed that the product-related construct had a significant influence on pre-purchase evaluation (p value was significant). The standardised direct effect of this construct on pre-purchase evaluation was 0.197 which means that when this construct increases by 1 standardised deviation, pre-purchase evaluation increases by 0.197 standardised deviation. Again, hypothesis H1b is accepted. H1c Organic food product-related construct is positively correlated to Chinese consumers’ behavioural/purchase intention The results (refer to Table 6.15) revealed that the product-related construct did not have a significant influence on behavioural/purchase intention (p value was not significant). It was also negatively associated with the behavioural/purchase intention of urban Chinese consumers. This means that it had no significant bearing on consumers’ purchase intention of organic food. Thus, hypothesis H1c is rejected. However, the results (Appendix 7, Table B) also revealed that the product-related construct had an indirect effect (0.371) on behavioural/purchase intention, which means 185 that the product-related construct was associated with behavioural/purchase intention through the mediatory affect of beliefs/attitudes and pre-purchase evaluation. Hypothesis two H2a The regulatory construct associated with organic food has a positive influence on beliefs/attitudes of urban Chinese consumers The results (refer to Table 6.15) revealed that the regulatory construct had a significant influence on beliefs/attitudes (p value was significant). The standardised direct effect of the regulatory construct on beliefs/attitudes was 0.274, which means that when this construct increases by 1 standardised deviation, beliefs/attitudes increases by 0.274 standardised deviation. Thus, hypothesis H2a is accepted. H2b The regulatory construct associated with organic food is positively correlated to Chinese consumers’ pre-purchase evaluation The results (refer to Table 6.15) revealed that the regulatory construct had a significant influence on beliefs/attitudes (p value was significant). The standardised direct effect of this construct on pre-purchase evaluation was 0.379, which means that when this construct increases by 1 standardised deviation, pre-purchase evaluation increases by 0.379 standardised deviation. Again, hypothesis H2b is accepted. H2c The regulatory construct associated with organic food is positively correlated to Chinese consumers’ behavioural/purchase intention The results (refer to Table 6.15) revealed that the regulatory construct did not have a significant influence on behavioural/purchase intention (p value was not significant). Thus, hypothesis H2c is rejected. However, the results (Appendix 7, Table B) also revealed that the regulatory construct had an indirect effect (0.332) on behavioural/purchase intention, which means that the regulatory construct was associated with behavioural/purchase intention through the mediatory affect of beliefs/attitudes and pre-purchase evaluation. 186 Hypothesis three H3a The lifestyle construct has a positive influence on the beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products The results (refer to Table 6.15) revealed that the lifestyle construct had a significant influence on beliefs/attitudes (p value was significant). The standardised direct effect of the lifestyle construct on beliefs/attitudes was 0.239 which means that when this construct increases by 1 standardised deviation, beliefs/attitudes increases by 0.239 standardised deviation. Thus, hypothesis H3a is accepted. H3b The lifestyle construct is positively correlated to consumers’ pre-purchase evaluation of organic food products in urban China The results (refer to Table 6.15) revealed that the lifestyle construct did not have a significant influence on pre-purchase evaluation (p value was not significant). Thus, hypothesis H3b is rejected. However, the results (Appendix 7, Table B) also revealed that the lifestyle construct had an indirect effect (0.161) on pre-purchase evaluation, which means the lifestyle construct was associated with pre-purchase evaluation through the mediatory affect of beliefs/attitudes. H3c The lifestyle construct is positively correlated to Chinese consumers’ behavioural/purchase intention of organic food products The results (refer to Table 6.15) revealed that the lifestyle construct had a significant influence on behavioural/purchase intention (p value was significant). The standardised direct effect of the lifestyle construct on behavioural/purchase intention was 0.127, which means that when the lifestyle construct increases by 1 standardised deviation, behavioural/purchase intention increases by 0.127 standardised deviation. Again, hypothesis H3c is accepted. 187 Hypothesis four H4a The ethnocentrism construct has a significant influence on beliefs/attitudes of urban Chinese consumers towards the purchase of organic food products H4b The ethnocentrism construct is significantly correlated to the pre-purchase evaluation of organic food products in urban China H4c The ethnocentrism construct is significantly correlated to Chinese consumers’ behavioural/purchase intention of organic food products The results (refer to Table 6.15) revealed that the ethnocentrism construct did not have a significant influence on beliefs/attitudes, pre-purchase evaluation and behavioural/ purchase intention (p values were not significant). Thus, hypotheses H4a H4b and H4c are all rejected. Hypothesis five H5: Chinese consumers’ beliefs/attitudes are directly and positively correlated to their pre-purchase evaluation The results (refer to Table 6.15) revealed that the beliefs/attitudes related construct had a significant influence on pre-purchase evaluation (p value was significant). The standardised direct effect of the beliefs/attitudes on pre-purchase evaluation was 0.196, which means that when beliefs/attitudes construct increases by 1 standardised deviation, pre-purchase evaluation increases 0.196 standardised deviations. Thus, hypothesis H5 is accepted. Hypothesis six H6: The pre-purchase evaluation of Chinese consumers to organic food is directly and positively related to their behavioural/purchase intention The results (refer to Table 6.15) revealed that the pre-purchase evaluation construct had a significant influence on behavioural/purchase intention (p value was significant). The 188 standardised direct effect of the pre-purchase evaluation on behavioural/purchase intention was 0.621, which means that when pre-purchase evaluation increases by 1 standardised deviation, behavioural/purchases intentions increases by 0.621 standardised deviations behavioural/purchase intention. Thus, hypothesis H6 is accepted. 6.6 Analysis of additional data Besides the analysis of the 52 statements of the survey (EFA, CFA and SEM) reported thus far, analysis of data obtained from other sections of the survey was also undertaken and is reported in this section. 6.6.1 Analysis of demographic control variables as shown in the conceptual framework This section analyses the influences of demographic control variables, that is, gender, age, education and income levels, towards the purchase intention of organic food in China. The analyses were conducted using t-tests and ANOVA statistical techniques. H7a: Women are more likely to purchase organic food than men in urban China Independent samples t-tests are often used to compare the mean scores on continuous variables for two different groups of participants, that is, male and female (Allen & Bennett 2010; Pallant 2007). Hence an independent samples t-test was conducted, and the results are shown in Table 6.16. 189 Table 6.16 Independent samples t test Levene’s test for equality of variances t-test for equality of means 95% confidence interval of the difference Total_ Equal purchase_ variances intention assumed Equal variances not assumed F Sig. t df Sig. (2tailed) Mean Std. error difference difference Lower .210 .647 .726 921 .468 .028 .039 -.048 .104 .730 812.267 .466 .028 .038 -.047 .103 Upper Table 6.16 illustrates that Levene’s test was non-significant (sig. =.647). This result was greater than the cut-off of .05, thus equal variances can be assumed. The value in the sig. (2-tailed) column was .486, which is also above the required cut-off of .05. Therefore, there were no significant differences (p>.05) in the purchase intention of organic food between females and males in urban China. Thus, hypothesis H7a is rejected. H7b: Younger consumers are more likely to purchase organic food than older consumers ANOVA is used to test hypotheses about means when there are two or more groups of one independent variable (Pallant 2007). In this case, age group was considered to be the independent variable, which included four age groups: (a) between 18-30; (b) between 31-45; (c) between 46-60; (d) greater than 60. A factorial between groups ANOVA was used to compare the mean intention scores of different age groups. A Levene’s test for homogeneity of the four age groups was conducted. The results demonstrated that the differences of the variance between the groups was not significant at p = .05 [F (3, 915) = 2.162, p=.091]. The results of the ANOVA are depicted in Table 6.17. 190 Table 6.17 ANOVA (purchase intention, age groups) Sum of squares df Mean square F Sig. 2.484 3 .828 2.511 .057 301.750 304.234 915 918 .330 Between groups Within groups Total Figures in Table 6.17 reveals that a statistical value of sig (0.057) was greater than 0.05, indicating that there were no significant differences in the mean scores for purchase intention of organic food across the four age groups. Thus, hypothesis H7b is rejected. H7c: Highly educated consumers are more likely to purchase organic food as compared to those who are less educated ANOVA is used to test hypotheses about means when there are two or more groups of one independent variable. In this case, educational group was considered to be the independent variable, which included five levels of educational groups: (a) below high school; (b) high school; (c) two years college or associate degree; (d) bachelor degree; (e) postgraduate or above. The Levene’s test for homogeneity of the five different educational groups was conducted. The results demonstrate that the differences of the variances between the groups was not significant at p=0.05 [F (4, 913) =0.552, p=0.697]. The results of the ANOVA depicted in Table 6.18 reveals that a statistical value of sig (.000) is less than 0.05. It indicates that there were significant differences at the p<0.05 level in the mean scores for purchase intention of organic food across the five education groups [F (4, 913) = 7.229, p=0.000]. Table 6.18 ANOVA (purchase intention, education levels) Sum of squares df Mean square F Sig. Between groups 9.310 4 2.328 7.229 .000 Within groups 293.965 913 .322 Total 303.275 917 191 Despite reaching statistical significance, the actual difference in mean scores between the groups was quite small. The effect size calculated using eta squared7 was: Eta squared = Sum of squares between groups/Total sum of squares = 9.310/303.272 = 0.03 Post-hoc comparisons (Appendix 8, Tables A and B) of results indicate that the mean score for group 1 ‘below high school’ (M = 3.47, SD =0.575) was significantly different from group 2 ‘high school’ (M = 3.68, SD =0.533), group 3 ‘Two years college or associate degree’ (M =3.84, SD =0.591), group 4 ‘bachelor degree’ (M = 3.73, SD = 0.568), and group 5 ‘postgraduate or above’ (M = 3.76, SD = 0.564). However, group 2 did not significantly differ from groups 3, 4 or 5. Therefore, hypothesis H7c was partially accepted. H7d: Consumers earning higher income are more likely to purchase organic food as compared to those earning less ANOVA was used to test whether the higher the Chinese consumers’ income level, the more they intend to purchase organic food. The four different income level groups included: (1) Less than RMB5,000 (2) Between RMB5,001-10,000 (3) Between RMB10, 001-20,000 (4) Greater than RMB20, 000. Levene’s test for homogeneity of the four different income groups was conducted. The results demonstrated that the differences of the variances between the groups was not significant at p=.05 [F (3, 913) =1.606, p=0.186]. The results of the ANOVA are depicted in Table 6.19 which revealed that a statistical value of sig (0.000) was less than 0.05. It indicates that there were statistically significant differences at the p<0.05 level in purchasing intentions scores across the four income groups [F (3, 913) = 6.870, p=0.000]. 7 Eta squared: a rough estimate of effect size is available for any ANOVA through eta squared. It reflects the proportion variance in the dependent variable that is associated with the level of an independent variable (Tabachnick & Fidell 2007). 192 Table 6.19 ANOVA (purchase intention, income levels) Sum of squares df Mean square F Sig. Between groups 6.703 3 2.234 6.870 .000 Within groups Total 296.971 303.674 913 916 .325 Despite reaching statistical significance, the actual difference in mean scores between the groups was quite small. The effect size calculated using eta squared was: Eta squared = Sum of squares between groups/Total sum of squares = 6.703/303.674 =0.02 Post-hoc comparisons (Appendix 8, Tables C and D) of results indicate that the mean score for group 1 (consumers’ income ‘less than RMB5,000’) (M=3.63, SD=0.545) was significantly different from groups 2 (‘Between RMB5,001-10,000’) (M=3.81, SD=0.572) and group 4 (‘Greater than RMB20,000’) (M=3.84, SD=0.540), but did not differ significantly from group 3 (‘Between RMB10,001-20,000’) (M=3.74, SD=0.643). Group 2 did not differ significantly from either group 3 or 4. Hence, hypothesis H7d is partially accepted. 6.6.2 Important attributes relating to purchase of organic food This section analyses the important attributes associated with the purchase of organic food in China. Twenty-one of the 23 statements of section two of the survey were given mean ratings above ‘3’ (neutral rating) as shown in Appendix 8, Table E. This suggests that on the whole, the ratings given by the respondents to the various statements of important attributes for the purchase of organic food were generally significant. Table 6.20 illustrates the top five important attributes of these statements, ‘certification relating to the quality of organic food’, and the ‘overall quality of organic food’ were rated most important, followed by ‘enforcement relating to the quality of organic food’, ‘information about the nutritional value of organic food’, and ‘food safety in relation to 193 organic food’. This confirms the general concern and poor confidence that respondents have about certification and quality of organic food in China. Table 6.20 Top five important attributes No. 2.19 2.4 2.20 2.23 2.18 Statements Certification relating to the quality of organic food The overall quality of organic food Enforcement relating to the quality of organic food Information about the nutritional value of organic food Food safety in relation to organic food Mean 4.24 4.24 4.22 4.15 4.14 Std. deviation .842 .829 .868 .851 .890 Table 6.21 shows the five least important attributes relating to the purchase of organic food in China. These include ‘the idea of saving face when purchasing organic food’, and ‘social status of people purchasing organic food’. Attributes relating to status are least important in the purchase of organic food. This is in contrast to results obtained by Sun and Collins (2006) which indicate that organic food consumption is not an everyday activity practised in public, and organic food is used only for private consumption. China is described as being a highly collectivist society with high power distance, and status is important as an indicator of power and achievement. However, face-saving is important only when purchasing a socially visible product, such as an expensive watch or prestigious car, which can demonstrate the owner’s status. The pragmatic Chinese tend to value function over form for private-use products and quality is considered more important than appearance (Schütte & Ciarlante 1998). Table 6.21 Five least important attributes No. Statements Mean 2.3 2.14 2.7 2.11 2.10 The appearance of organic food The country of origin of the organic food The promotion and advertising of organic food The social status of people purchasing organic food The idea of face saving (mianzi) when purchasing organic food 3.34 3.30 3.27 2.35 2.30 Std. deviation .947 1.043 .977 1.103 1.084 As shown above, the least important attributes in respondents’ choice also include ‘promotion and advertising’, ‘country of origin’ and ‘appearance’. Attributes relating to status are not so important in the purchase of organic food. This finding is in contrast to 194 that obtained by Sun and Collins (2006) who pointed out that Chinese consumers purchase imported fruit not only to attain a high quality product, but also expect to gain some pleasure and symbolic benefit such as ‘superior appearance’, ‘expensive’ and ‘high prestige’. Consumers in developing countries buy foreign products which are generally considered social symbols and also very trendy, particularly those whose origins are from prestigious countries (Batra et al. 2000). Another study argued that consumers from developing countries, such as China, highlight the impact of ethnocentrism by their willingness to buy domestic products. However, when these consumers hold higher conspicuous consumption values, they tend to perceive domestic products as being lower quality (Wang & Chen, 2004). These findings are consistent with the results obtained after testing hypothesis four which was associated with ethnocentrism. 6.6.3 Usage pattern of organic food Essentially, there were five intended uses for respondents’ most recent purchase of organic food: for their own consumption, for children, for elderly people, as a gift and for family consumption. The importance of each attribute might be affected by its single intended use as compared to multiple intended usages (Sun & Collins 2006). Since many respondents in this survey indicated multiple intended usages, the results would be dissimilar to those of consumers with a single intended use. Respondents who selected more than one reason were eliminated in this analysis. Therefore, in this section, the analysis is only based on examining responses relating to a single intended use and their evaluation of important attributes of organic food. A total of 710 respondents (Appendix 8, Table F) chose a single reason for their latest organic food purchases. The main reasons were ‘for own consumption’ (n=254), ‘for children’ (n=90), ‘for elderly people’ (n=27), ‘for gift’ (n=11) and ‘for family consumption’ (n =328) (details are shown in Appendix 8, Table F). 195 6.6.4 Important attributes versus usage pattern ANOVA between groups was used to investigate whether there were significant differences between intended uses in consumers’ evaluations of important attributes for the purchase of organic food. Table 6.22 reports that there were statistically significant differences at p<0.05 level in (1) social status [F (4, 702) = 4.877, p=0.001]; (2) the awareness of organic food products [F (4, 702) = 6.502, p=0.000]; (3) food safety [F (4, 700) = 3.948, p=0.004] (details are shown in Appendix 8, Table G). Table 6.22 Means of important attributes as applicable to the usage pattern Important attributes Taste Smell Appearance Quality Price Availability Promotion Value Environmental Face saving Social status Knowledge Awareness Country of origin Produce in China Brand Regulation Food safety Certification Enforcement Packaging Labelling Information Own consumption 3.83 3.86 3.25 4.32 3.88 3.73 3.20 3.94 4.01 2.16 2.14 3.66 3.38 3.17 Children Elderly Gift Family ANOVA(F) 3.82 3.58 3.40 4.09 4.04 3.81 3.27 3.84 3.69 2.44 2.58 3.71 3.81 3.36 3.63 3.67 3.48 4.39 3.81 4.00 3.37 3.59 3.96 2.19 2.48 3.70 3.85 3.30 3.45 3.64 3.36 4.27 3.82 3.80 3.64 3.73 3.91 2.64 3.18 3.73 4.09 3.36 3.83 3.71 3.41 4.24 3.93 3.79 3.27 3.88 3.94 2.30 2.30 3.74 3.63 3.37 .853 2.365 1.204 1.612 .765 .675 .748 1.088 1.922 1.675 4.877* .248 6.502* 1.424 3.33 3.38 3.26 3.36 3.40 .246 3.35 3.87 4.27 4.30 4.28 3.36 3.99 4.20 3.51 3.72 3.87 4.08 4.16 3.50 4.14 4.08 3.52 3.93 3.93 4.22 4.04 3.56 3.96 3.81 3.55 4.18 4.27 4.27 4.36 3.36 4.18 4.18 3.56 3.88 4.10 4.27 4.22 3.52 4.10 4.15 1.994 .835 3.948* 1.208 .786 1.068 1.006 1.406 *p <0.05 196 Hypothesis 7e: tests whether the important attributes of organic food products have positive influences on their intended usage pattern in China In ANOVA, if the significance value is equal to or less than 0.05, it means that there is a significant difference somewhere among the mean scores on the dependent variable for the groups. The post-hoc test tells exactly where the differences among the groups occur. The statistical significance of the differences between each pair of groups is examined in multiple comparisons, which give the results of the post-hoc test (Pallant 2007). The results show that social status, awareness and food safety were significantly different from the rest of the twenty important attributes, in terms of the five intended usages. To investigate where these differences existed among the five intended use groups, posthoc test were conducted on the data. Despite reaching statistical significance, the actual differences in mean scores between groups were quite small. The effect sizes, calculated using eta squared were: Eta squared = Sum of squares between groups/Total sum of squares. Hence, Eta squared of social status =22.929/848.028=.0027 Eta squared of awareness=20.527/574.549=0.035 Eta squared of food safety=12.555/569.016=0.022 According to Cohen’s term (in Pallant 2007), 0.01 is a small effect, 0.06 is a medium effect and 0.14 is a large effect. Therefore, despite reaching statistical significance, the actual difference in mean scores between the five different intended use groups was quite small. The results (Appendix 8, Tables G, H and I) of the post-hoc comparisons for the evaluation of ‘the social status of people purchasing organic food’ revealed that differences in the mean score for the group of ‘own consumption’ (M=2.14, SD=1.078) was significantly different from the group ‘use for children’ (M=2.58, SD=1.085), and ‘gift’ (M=3.18, SD=1.079). Social status was more important for gift giving, and least important for own consumption. 197 In terms of evaluation of ‘the awareness of organic food products’, the mean score for own consumption (M=3.38, SD=0.975) was significantly different from that for children (M=3.81, SD=0.748) and for families (M=3.86, SD=0.867). Awareness of organic food was most important for gift giving (M=4.09, SD=0.701) and least important for own consumption (M=3.38, SD=0.975), which seems logical in the Chinese context. In terms of evaluation of ‘food safety in relation to organic food’, the mean score for own consumption (M=4.27, SD=0.859) was significantly different for children (M=3.87, SD=.919). Food safety was important for own consumption (M=4.27, SD=0.859), as well as for gift giving (M=4.27, SD=0.786). Hence, some support for hypothesis H7e was found. The results suggest that the above statements and associated factors are important to consumers when deciding about the purchase of organic food. Table 6.23 summarises the acceptance/rejection of the hypotheses. The results revealed that the ethnocentrism construct did not influence beliefs and attitudes, pre-purchase evaluation or behavioural/purchase intention of urban Chinese consumers towards organic food. Table 6.23 Summary of hypotheses testing No. H1a H1b H1 c H2 a H2b H2c H3a H3b Hypothesis Product-related construct has positive influence on the personal beliefs/attitudes of urban Chinese consumers Product-related construct is positively correlated to the prepurchase evaluation Product-related construct is positively correlated to Chinese consumers’ behavioural/purchase intention Regulatory construct has positive influence on personal beliefs/attitudes of urban Chinese consumers Regulatory construct is positively correlated to the pre-purchase evaluation Regulatory construct is positively correlated to Chinese consumers’ behavioural/purchase intention Lifestyle construct has positive influence on the personal beliefs/attitudes of urban Chinese consumers Lifestyle construct is positively correlated to the pre-purchase evaluation Accepted/ rejected Accepted Accepted Rejected Accepted Accepted Rejected Accepted Rejected 198 H3 c H4 a H4 b H4 c H5 H6 H7 a H7 b H7 c H7 d H7 e Lifestyle construct is positively correlated to the Chinese consumers’ behavioural/purchase intention Ethnocentrism construct has a negative influence on the personal beliefs/attitudes of urban Chinese consumers Ethnocentrism construct is negatively correlated to the pre-purchase evaluation Ethnocentrism construct is negatively correlated to Chinese consumers’ behavioural/purchase intention Personal beliefs/attitudes is directly and positively correlated to Pre-purchase evaluation Pre-purchase evaluation construct is directly and positively related to behavioural/purchase intention Women are more likely to purchase organic food The younger the consumers’ age, the more they intend to purchase organic food The higher the Chinese consumers’ level of education, the more they intend to purchase organic food The higher the Chinese consumers’ income, the more they intend to purchase organic food The important attributes of organic food products have positive influences on their intended usage pattern Accepted Rejected Rejected Rejected Accepted Accepted Rejected Rejected Partially Accepted Partially Accepted Partially Accepted 6.6.5 Comparative analysis of responses from the four selected cities Previous studies suggest that urban China does not constitute a homogeneous market; there are significant differences between various regions and cities (Fram, Le & Reid 2005; Yusuf & Brooks 2010). To determine whether there were differences in the final structural model for responses from the four selected cities, invariance tests were conducted. 6.6.5.1 Invariance tests Multigroup comparisons are commonly used to test the invariance of measurement instruments across groups (Jones-Farmer, Pitts & Rainer 2008). Members of different populations who have the same standing on the construct being measured receive the same observed score on the test. The measure can be tested invariantly (Schmitt & Kuljanin 2008). Invariant testing is an assessment of the relationships between latent 199 variables and their indicator statements, across different groups, e.g. culture, age, gender, location (Vandenberg & Lance 2000). In the context of invariance tests, Byrne (2010, p. 209) suggests that: Testing for multigroup invariance provides two important functions. First, it allows for invariance tests to be conducted across the two groups simultaneously. Second, the fit of this configural model provides the baseline value against which all subsequently specified invariance models are compared. Despite the multigroup structure of this and subsequent models, analyses yield only one set of fit statistics for overall model fit. The respondents from four cities (Beijing, Shanghai, Shenzhen and Chengdu) were used to carry out these invariance tests. The aim was to check whether their views were similar or different, and whether the final model was adequately valid for each city. Byrne (2004) suggests that establishing a baseline model for each group separately is often a prerequisite for testing multigroup invariance. Therefore the invariance analyses were performed using a two-stage routine. The first step was to investigate the data from each city to ascertain its fit with the model. The second step was to investigate the data from all the four cities simultaneously in a single analysis. 6.6.5.2 Step one of the invariance tests The initial step involved the application of the final structural model to the data from each city. The sample size and fit indices for each group and the whole data set are presented in Table 6.24, which shows goodness-of-fit indices for the four cities. 200 Table 6.24 Goodness-of-fit indices showing comparative analysis of data from the four cities Note: p= Bollen-Stein p values. The model served as a baseline for comparison. The chi-square test statistics showed that the data did not fit the model well for each of the four cities (p=0.001) and hence did not support configural invariance across the four groups. Since all of SRMR values were greater than 0.05, it may suggest that outliers were present in the four groups. However, the results did not improve even after deleting the potential outliers. Consequently, no outlying cases were removed. The components of influencing dimensions (product, regulatory, lifestyle) would accommodate different statements. For all these models, all parameter estimates were not significant, and the pattern of relationships between influencing dimensions with personal attitudes, pre-purchase evaluation and purchase intention were very different. It possibly suggests that individual models would apply for each city. According to model fit indices used in this study (Table 6.2), and based on value of CMIN/DF, the results suggested that in comparative terms, the model fitted the data best for respondents from Chengdu (CMIN/DF=1.576), followed by Beijing (CMIN/DF=1.771), Shenzhen (CMIN/DF=1.772) and Shanghai (CMIN/DF=2.912). 6.6.5.3 Step two of the invariance tests The most efficient way of conducting measurement invariance tests in AMOS is to run a number of simultaneous multiple-group testing in a single analysis (Byrne 2010; Cunningham 2008). This involves invariance tests across the four groups simultaneously. Structural covariances were chosen, which means equating the variances and covariances of the measured variables across all groups (Cunningham 201 2008). Hence analysis was performed based on the final model by applying it to the data obtained from respondents in the four cities simultaneously. The idea of invariance tests is to progressively test increasingly constrained models that are nested in previously estimated less constrained models. The difference of chi-square (∆ χ2) test is required (Cunningham 2008), and can be calculated from results shown in Table 6.25. This result suggests that the variance and covariances patterns were not consistent across four cities. Table 6.25 Goodness-of-fit indices for data from all four cities Model Unconstrained Measurement weights Structural weights Structural covariances Structural residuals Measurement residuals Saturated model Independence model NPAR 296 239 185 182 158 74 1624 112 CMIN 2841.312 2953.846 3046.691 3059.356 3120.014 3730.999 .000 9354.366 DF 1328 1385 1439 1442 1466 1550 0 1512 P .000 .000 .000 .000 .000 .000 CMIN/DF 2.140 2.133 2.117 2.122 2.128 2.407 .000 6.187 The difference in chi-square between the nested models is also distributed as a chisquare distribution with degrees of freedom equal to the difference in degrees of freedom ∆ x 2/∆ df =3059.36-2841.312/1442-1328 = 1.61E-08. Obviously this is significant at p<0.001 level. The above analysis indicates that the data was not a good fit to the structural equation model as there were significant differences across the four cities. Therefore, the invariance tests indicate that the structural equation model did not apply homogeneously across data from all four cities. Although the model fits well with the entire data, it does not work with data from each dividual city. Hence, we cannot assume that the data was drawn from the same population. In other words, respondents from the north had different opinions to those from the south, east and west. 202 6.6.5.4 Step three of the invariance tests The final step was to locate where the differences occurred in the data from the four cities. There are a number of ways to operationalise this. One way is to examine the standardised factor coefficients (standardised regression weights) across data from the four cities (Vandenberg & Lance 2000). Table 6.26 depicts the AMOS output of the data from respondents in each city, showing significant differences in the standardised regression weight. Table: 6.26 Standardised regression weights (four cities) Beijing Shenzhen Chengdu Shanghai <--- PRODUCT_RELATED REGULATORY LIFESTYLE ETHNOCENTRISM BEL/ATT REGULATORY LIFESTYLE PRODUCT_RELATED ETHNOCENTRISM <--- PRE_-PURCHASE .740 .827 .582 .416 <--- LIFESTYLE .417 -.047 .201 -.056 <--- PRODUCT_RELATED -.213 -.164 .010 .035 <--- REGULATORY -.005 -.137 -.042 .183 <--- ETHNOCENTRISM -.160 .220 -.060 .088 BEL/ATT <--- BEL/ATT BEL/ATT BEL/ATT PRE_-PURCHASE PRE_-PURCHASE PRE_-PURCHASE <--<--<--<--<--<--- PRE_-PURCHASE <--PRE_-PURCHASE BEHAVIOUR_INTENTION BEHAVIOUR_INTENTION BEHAVIOUR_INTENTION BEHAVIOUR_INTENTION BEHAVIOUR_INTENTION .284 .407 .319 .442 .422 .173 -.019 .077 .597 .267 .160 .096 .120 .372 .359 .054 .267 .383 .007 .188 .348 .304 .273 -.003 .032 .157 .114 .022 .009 .222 .049 .367 -.027 -.223 .013 .119 (Chi-squared= 2841.321, df =1328, p<0.001) Data from respondents in Shanghai had the worst model fit (details are shown in Table 6.25), such as χ2/DF equivalent 2.912 in Shanghai compared to 1.576 in Chengdu. Thus, six hypotheses were tested again based on data obtained from respondents of Shanghai. Table 6.28 summarises the results of the hypotheses tests for Shanghai. 203 <--<--<--<--<--<--<--<--- BELIEFS/ATTITUDES PRE_PURCHASE BEHAVIOUR_INTENTION BELIEFS/ATTITUDES PRE_PURCHASE BEHAVIOUR_INTENTION BELIEFS/ATTITUDES PRE_-PURCHASE BEHAVIOUR_INTENTION PRE_-PURCHASE BEHAVIOUR_INTENTION H2a H2b H2c H3a H3b H3c H4a H4b H4c H5 H6 Note: ***p<.001; **p <.01; *p <.05 <--- BEHAVIOUR_INTENTION <--- <--- <--- PRE_PURCHASE BELIEFS/ATTITUDES ETHNOCENTRISM ETHNOCENTRISM ETHNOCENTRISM LIFESTYLE LIFESTYLE LIFESTYLE REGULATORY REGULATORY REGULATORY PRODUCT .455 .160 .180 .222 .059 -.046 .017 -.002 .214 .122 .287 .051 .489 H1c PRODUCT PRE_PURCHASE H1b <--- .579 PRODUCT BELIEFS/ATTITUDES H1a <--- Estimate Hypothesis no. .092 .096 .144 .148 .137 .054 .054 .050 .092 .093 .089 .128 .138 .132 S.E. Table 6.27 Evaluated hypotheses tests for Shanghai 4.929 1.668 1.251 1.501 .431 -.852 .312 -.044 2.336 1.304 3.237 .398 3.554 4.369 C.R. *** .095 .211 .133 .667 .394 .755 .965 * .192 ** .691 *** *** P .416 .157 Rejected Accepted .088 .119 .032 -.056 .022 -.003 .183 .114 .273 .035 .367 .442 Beta Rejected Rejected Rejected Rejected Rejected Rejected Accepted Rejected Accepted Rejected Accepted Accepted Accepted/ rejected 204 A comparison of Tables 6.25 and 6.27 reveals that the relationships between constructs were different as compared to similar relationships in the whole data set applied to the model, i.e. regulatory construct to pre-purchase and behavioural/purchase intention, lifestyle construct to beliefs/attitudes and behavioural/purchase intention, beliefs/attitudes to pre-purchase evaluation. The results also suggest that the hypotheses tests (H1–H6) were quite different for data obtained from Shanghai and from four cities as a whole. There are a total 14 series of hypotheses, five of them demonstrating different results. The details are as follows: (1) Regulatory construct had direct relationship with pre-purchase evaluation in the whole data set, but did not show direct influence effect in the data obtained from Shanghai. (2) Regulatory construct did not have direct relationship with behavioural/purchase intention in the data obtained from the four cities as a whole, but had direct relationship with behavioural/purchase intention in the data obtained from Shanghai. (3) Lifestyle construct had direct relationship with beliefs and attitudes in the data obtained from the four cities as a whole, but did not have direct relationship with beliefs and attitudes in the data obtained from Shanghai. (4) Lifestyle construct had direct relationship with behavioural/purchase intention in the data obtained from the four cities as a whole, but did not have direct relationship with behavioural/purchase intention in the data obtained from Shanghai. (5) There was direct relationship between beliefs/attitudes and pre-purchase evaluation in the data obtained from the four cities as a whole, but the results from data obtained from Shanghai did not show the direct relationship. These results therefore would suggest that the relationships between factors and associated latent variables were not invariant across the data obtained from each of the four cities. They reveal that China is not a homogeneous country, a finding which is similar to those in 205 previous studies (Cui & Liu 2000; Fram, Le & Reid 2005; Wong & Yu 2003; Yusuf & Brooks 2010). 6.7 Chapter summary This chapter has presented a detailed analysis of data obtained in Study Two. A series of CFAs were performed to validate the findings from the EFA results of Study One. This process confirmed and modified the factors and the associated constructs in the measurement models, and assessed their validity and reliability. The re-specified models indicated reasonable model fit for the data. Finally, the measurement models were linked using a structural equation to obtain a final best-fit model. A series of hypotheses were tested. Invariance tests were performed on the data obtained from the four selected cities. 206 CHAPTER SEVEN: DISCUSSION OF FINDINGS, RECOMMENDATIONS AND CONCLUSION 7.1 Introduction This study was conducted to explore and investigate the various determinants which influence consumers’ purchase behaviour of organic food products in urban China. The three underpinning theories used in developing the conceptual framework for this research study are: the Consumer Decision Process model (CDP), the Hierarchy Effects model and the Theory of Planned Behaviour (TPB). The five stages of the proposed conceptual framework as explained in Chapter Three were operationalised and investigated by analysing data obtained in the pilot study (Chapter Five) and the main study (Chapter Six). This chapter engages in a discussion of the relationships between the various stages of the conceptual framework as evidenced by the final best-fit model (Section 7.2). The two research questions are addressed separately and answers are provided. Section 7.3 discusses the implications of the findings regarding the first research question which relates to the factors that influence urban Chinese consumers’ beliefs and attitudes, pre-purchase evaluation and behavioural/purchase intention towards the purchase of organic food. Section 7.4 addresses the second research question which relates to the role played by the purchase behaviour of organic food. The seven hypotheses presented in this study were empirically tested, and the resulting outcomes provide key findings as conclusive evidence of consumer buyer behaviour of organic food products in urban China. Section 7.5 includes contributions and recommendations both for the theory and practice of understanding consumer buyer behaviour of organic food in urban China. It becomes evident that there are implications for the various stakeholders involved in this industry. In addition this chapter highlights the limitations of the study (Section 7.6) and suggests future research directions (Section 7.7). Finally Section 7.8 provides concluding remarks to this research study. Figure 7.1 diagrammatically illustrates the structure of Chapter Seven. 207 Figure 7.1 Structure of Chapter Seven Source: developed for this research 7.2 Implications of the best-fit SEM model The main aim of this research study was to investigate the determinants that influence urban Chinese consumers’ decision-making towards organic food consumption. A five stages conceptual framework was proposed to achieve this aim. The five stages were named as influencing, cognitive/affective, evaluation of alternatives, behavioural/purchase intention and purchase. 208 As the antecedent of purchase and consumption of organic food, the influencing stage plays a vital role in the total consumer decision-making process. The emphasis was placed on the influencing aspects and dimensions of this process in the Chinese context. This stage comprised four constructs: product, regulatory, lifestyle and ethnocentrism. The final bestfit structural equation model (Figure 6.20) revealed that three influencing constructs, namely, those related to product, regulatory and lifestyle, directly and significantly influence urban Chinese consumers’ beliefs and attitudes towards the purchase of organic food. They also influence stages two (cognitive/affective), three (evaluation of alternative) and four (behavioural/purchase intention). The fourth construct of the influencing stage, ethnocentrism, was not found to have a significant effect on consumers’ beliefs/attitudes, pre-purchase evaluation and purchase intention. The findings of this study reveal that the influencing dimension constructs (product and regulatory) are associated with pre-purchase evaluation, and also with behavioural/purchase intention through the mediatory affects of beliefs/attitudes and pre-purchase evaluation. There is a positive direct influence of the lifestyle construct on organic food behavioural/ purchase intention. The results of the final best-fit model regarding beliefs/attitudes reveal that Chinese consumers believe that organic food is good for the environment, and has superior quality. Also organic food labels denote high quality of product, and consumers indicated their preference for organic food as compared to conventional food. In addition, three new linkages between (i) the product and lifestyle constructs, (ii) regulatory and product constructs and (iii) personal attitudes and behavioural intentions were made in the final best-fit model. These linkages indicate direct and significant relationships between these factors and constructs in the context of this research study. These findings assist in the accurate identification of determinants that influence Chinese consumers’ purchase intentions with regard to organic food, and have significant implications for both the theoretical study and management of the industry. A set of action plans is suggested, as summarised in Table 7.1. 209 Research question one: What are the factors that influence urban Chinese consumers’ beliefs and attitudes, pre-purchase evaluation and behavioural/purchase intention towards the purchase of organic food? Research questions Lifestyle related construct indirectly and positively influences on urban Regulatory related construct influences on urban Chinese consumers’ beliefs/attitudes towards the purchase of organic food Product related construct influences on urban Chinese consumers’ beliefs/attitudes towards the purchase of organic food. Key research findings Consumption of organic food is part of the lifestyle in urban China. The findings suggest that consumers who are more self-indulgent and more desirous of variety are more likely to have 1. This study suggests that sensory appeal (appearance, texture, smell) as well as nutritional value are part of a complex multitude of influences that shape Chinese consumers' purchase intentions with regard to organic food. The freshness of food is a key dietary consideration of Chinese consumers’ and a central element of Chinese cuisine. 2. Price sensitivity seems less important in the purchase of organic food. 1. Chinese urban consumers believe that there is lack of sufficient government involvement in scrutinising certifications and labels on organic food. 2. The terms 'organic' and 'green' are used interchangeably by many in the industry and are, in consequence, confusing for consumers. Implication of the current study 210 1. The organic food producers and marketers should develop technologies and methods of storage that will keep the products fresh for longer periods without resorting to the use of chemicals. Marketing organic food according to this concept means not only the method of production, but also the whole process of distribution from the farm to the table by means of organic methods. 2. Organic marketers may capitalise on the image of organic food as fresh and environmentally friendly, even though it is more expensive than conventional food. 1. Government policy makers and the organic industries should ensure the credibility of inspection and labelling of organic food to enhance Chinese consumers' trust in overall food safety. 2. There is a need for the industry to clearly explain the meaning of the term ‘organic’ and all that it implies with regard to foods. Educational campaigns need to be organised to raise awareness of, and clarify, the term 'organic'. The organic marketers should endeavour to focus on the ‘variety-seeking’ and ‘self-indulgent’ consumers. New marketing strategies should promote the additional attributes of organic food and how they meet the different concerns of these Action plans for the decision makers Table 7.1 Action plans for the decision makers Research question two: Evaluate the role of demographic variables and usage patterns in the behavioural /purchase intentions of urban Chinese consumers towards organic food 1. Organic food consumptions in urban China are partially influenced by the consumers' education and income levels. 2. There are significant differences in terms of social status, product awareness and food safety issues related to purchasing organic food in urban China Chinese consumers beliefs/attitudes towards organic food 1. This finding suggests the most affluent consumers in urban China who purchase organic food were also well educated. 2. There are five intended uses for organic food in urban China, indicating that the important attributes of organic food have exerted an influence on the different usage patterns. positive attitudes towards organic food. 211 1. The organic industry should anticipate targeting consumers who are highly educated and are high income earners. Providing additional knowledge of the concept of 'organic' is a key to motivating consumers in the higher demographic. 2. Organic marketers should improve the packaging of their products, in order to emphasise the benefits of consuming organic food, as well as provide certified organic logos that are signs of quality and prestige. consumers. These include such things as environmental and health benefits, and the need to possess novelty icons or follow trends. 7.3 Discussion of findings relating to research question one The first research question as previously articulated in Chapter One read as follows: What are the factors that influence urban Chinese consumers’ beliefs and attitudes, pre-purchase evaluation and behavioural/purchase intention towards the purchase of organic food? Research question one has been addressed through the influencing, cognitive/affective, evaluation of alternatives and behavioural/purchase intention stages of the proposed conceptual model. This model was tested and validated using a series of hypotheses (H1H6). The hypotheses tests performed on the constructs of the final best-fit model revealed that product, regulatory and lifestyle influenced urban Chinese consumers’ beliefs and attitudes towards the purchase of organic food. The subsequent sections address details of how the various constructs of the influencing stage impact on pre-purchase evaluation and purchase intention of urban Chinese consumers of organic food. A detailed discussion on each of these significant constructs and hypotheses follows: 7.3.1 Product H1a which hypothesises that the organic food product-related construct has a positive influence on urban Chinese consumers’ beliefs/attitudes was accepted. The results of the final best-fit structural equation model (Figure 6.20) demonstrate that the product-related construct has a total effect of 0.468 on beliefs and attitudes. The total effect comprises a direct effect of 0.38 plus an indirect effect of 0.08 (0.38+0.08=0.46). The evidence of this outcome is clearly indicated in Appendix 7. Kline (2005) suggests that the higher the standardised regression weights in the structural equation model, the more important the construct in predicting the results. This indicates that the product-related 8 All statistical analysis results in this chapter will only provide a two decimal showing. 212 construct is the most important construct positively influencing urban Chinese consumers’ beliefs and attitudes towards the purchase of organic foods. The texture, appearance, smell, nutritional value and trendiness are important attributes associated with this construct. These findings are consistent with previous studies (Magnusson et al. 2001; Torjusen et al. 2001), which suggest that aspects such as sensory appeal (e.g. odour, texture) are important to organic food consumers, and are part of the complex function of a multitude of influences associated with the consumption of food. The items related to the product construct are herewith discussed in a logical sequence as enumerated in the survey instrument shown in Appendix 1. Previous studies (Veeck 2000; Veeck & Burns 2005) suggested that serving fresh food to Chinese consumers is important, and even more important is their mind-set that food is ‘fresh’ only when it has been purchased that day and has not been refrigerated. Frequent shopping trips to purchase the freshest, tastiest and nutritious meals for family members remain the custom in China. The results of this study further confirm that Chinese consumers hold in high regard food sensory characteristics (Veeck 2000; Zhao et al. 2000). The freshness and texture are of high value on decision-making of purchasing food products (Zhao et al. 2000). As stated by Ennis (2007), consumers are possibly reluctant to purchase products which have no visual signs of quality. Importantly, in the process of obtaining the final best-fit structural equation model, the item relating to ‘taste’ was deleted. Hence, it is likely that ‘taste’ is less important for organic consumers in China. Confirming this view was a Taiwanese study (Chen 2007) which reported that taste evaluation (sensory appeal) did not contribute in influencing consumers’ purchase attitudes towards organic food. Zhao et al. (2007) also suggest that US customers perceive no significant taste difference in conventional and organic produced vegetables. The findings of this study suggest that organic food is perceived as being of high nutritional value. This view is supported by several studies conducted in different countries (Botonaki et al. 2006; Krystallis & Chryssohoidis 2005). Generally organic food consumers perceive that organic food is more nutritious than conventional food, even though this perception is 213 sometimes overstated, especially as there is lack of extensive scientific evidence to support such a claim (Kluger 2010). Respondents in this study perceive organic food to be ‘trendy’. This finding is similar to that reported by Lockie et al. (2002) who found that stereotype images of organic food consumers seem to be more interested in fashion than anything else. A study conducted in Thailand by Roitner-Schobesberger et al. (2008) revealed that the second most important motive for the purchase of organic food was its attraction as new and ‘fashionable’. The way organic food is marketed in the media reinforces its perception amongst some consumers as being fashionable and trendy (Hill & Lynchehaun 2002). As the data for this research study was obtained in early 2010, a period following the most severe global economic downturn since the Great Depression, it was expected that the price of organic food would be an important consideration to consumers. However, in the process of obtaining the final best-fit structural equation model, items relating to price were removed. It was expected that price sensitivity would be important for Chinese consumers, nevertheless the findings of this study suggest that price has no significant influence on their beliefs and attitudes or on their purchase intention. Previous literature focusing on global studies suggests that the higher cost of organic food is one of the barriers to its consumption. Moreover, price premiums tend to negatively affect organic food purchasing (O’Donovan & McCarthy 2002; Shepherd, Magnusson & Sjödén 2005; Tregear, Dent & McGregor 1994). On the other hand, Chang and Zepeda (2005) observe that consumers who are more knowledgeable about organic foods are willing to pay higher prices. In contrast, the findings of this study provide a somewhat different outcome. Chinese consumers do not have great understanding about organic food, and they seem to accept and are willing to pay higher prices. Only 15.7% of respondents in this study believed that they knew a lot about organically produced food, while the others were unsure as to what ‘organic’ meant and knew little about it. This indicates that organic food is still a new 214 concept in China, and its awareness is low. Chinese consumers are generally not knowledgeable about organic foods, even though they tolerate higher prices and are willing to pay extra. Some other studies (Fotopoulos, Krystallis & Ness 2003; Tarkiainen & Sundqvist 2005) also suggest the reduced importance of price. Consumers did not see that the higher price of organic food would affect their intention to buy. As was evidenced in a Greek study (Krystallis, Fotopoulos & Zotos 2006), organic consumers are less price sensitive and more interested in the quality of the food. This may provide evidence that, in a relatively affluent urban Chinese society, price is less important than food safety and health related attributes. This may support the argument that food safety and not price is one of the more important issues influencing consumers’ purchase of organic food products in urban China. In this respect, Eves and Cheng (2007) report that product appearance and quality are more important than price when evaluating factors driving Chinese consumers’ purchase intention of new food products. H1b which hypothesises that the product-related construct is positively correlated to the pre-purchase evaluation was accepted. The output of the final best-fit model (Appendix 7) demonstrates that the product-related construct has a total effect of 0.40 on pre-purchase evaluation which comprises a direct effect of 0.20 plus an indirect effect of 0.20 (0.20+0.20=0.40). This reveals that the product-related construct is positively correlated to the pre-purchase evaluation. H1c which hypothesises that the product-related construct is positively correlated to urban Chinese consumers behavioural/purchase intention was not supported. Furthermore, the results (Appendix 7) also demonstrate that the product-related construct has a total effect of 0.30 on purchase intention which comprises a direct effect of -0.07 plus an indirect effect of 0.37 (-0.07+0.37=0.30). The very weak (-0.07) direct effect suggests that there is no significant relationship between the product-related construct and purchase 215 intention. However, the product-related construct is associated with purchase intention through the mediatory construct of beliefs/attitudes and pre-purchase evaluation. These findings are similar to those reported by Chryssochoidis (2000) where the standardised coefficient between purchase intention and organic food physical product appearance is weak (0.08). 7.3.2 Regulatory H2a which hypothesises that the regulatory construct has a positive influence on urban Chinese consumers’ beliefs/attitudes was accepted. The results (Appendix 7) of the final best-fit model (Figure 6.20) demonstrate that the regulatory construct in the influencing stage has a total direct effect of 0.27 on beliefs/attitudes. This suggests that it is the second most important construct influencing consumers’ beliefs/attitudes after the product construct. Although the relationship between the regulatory construct and beliefs/attitudes is relatively weak (0.27), it is statistically significant (p<0.05), which suggests that it plays an important role in directly and positively influencing Chinese urban consumers’ beliefs and attitudes. Consumers perceive that there is lack of adequate government control on organic production. Moreover, organic logos and labels need to be controlled and monitored, including allocating more land for organic production. In addition, the findings of the study in relation to certification of the logos for organic food suggest that consumers believe that the organic food market is quite chaotic, and the terms ‘organic’ and ‘green’ are confusing. The items related to the regulatory construct are herewith discussed in a logical sequence enumerated in the survey instrument shown in Appendix 1. The findings reveal that, as in many western countries, Chinese consumers are motivated by food safety issues related to personal health and also the environment. They are suspicious regarding the quality of food purchased and have high quality expectations of organic food certification, inspections and policy enforcement. They are also eager to understand more about organic food and what this industry is all about. An organic food 216 logo is the consumer’s guarantee that the product has been produced organically. This finding is similar to that obtained by a study conducted by the United States Department of Agriculture (2008). It reveals that Chinese consumers may doubt whether labelled organic products are really organic or not and whether they meet all declared standards. On the matter of market regulation, the results support previous studies (Giannakas 2002; Radman 2005) which suggest that consumers believe that the organic market is not regulated by appropriate laws. They demonstrate a low trust in the regulations of organic food and they possess little awareness of organic systems. Similar results are reflected in Table 6.20 which lists the top five most important attributes relating to the purchase of organic food. From the foregoing discussion it is evidenced that Chinese consumers regard certification, quality, enforcement, information about nutritional value and food safety related to organic food as the key issues when they consider their purchase. H2b which hypothesises that the regulatory construct is positively correlated to prepurchase evaluation was accepted. The results of the final best-fit SEM reveal that the regulatory construct has a total effect of 0.43 on pre-purchase evaluation (Appendix 7). This comprises a direct effect of 0.38 plus an indirect effect of 0.05 (0.38+0.05=0.43). These findings indicate that the regulatory construct is positively correlated with the pre-purchase evaluation. H2c which hypothesises that the regulatory construct is positively correlated to Chinese consumers’ behavioural/purchase intention was not supported. The regulatory construct has a total effect of 0.40 on purchase intention which comprises a direct effect of 0.07 plus an indirect effect of 0.33 (0.07+0.33=0.40). The direct effect is very weak, showing its relative unimportance. This suggests that the regulatory construct is associated with behavioural/purchase intention through the mediatory affect of beliefs/ attitudes and pre-purchase evaluation. 217 Depending on the pre-purchase evaluation, Chinese consumers would purchase or give up purchasing organic food. Government policy or regulations may directly influence their beliefs/attitudes and pre-purchase evaluation, and further indirectly influence purchase intention. From a related regulatory standpoint, the results confirm that consumers searching for information about organic food products during the pre-purchase phase are interested to know where they are produced. They ascertain whether the organic logo can be trusted, and where they can buy available products. Krystallis, Arvanitoyannis and Chryssohoidis (2006) suggest that consumers’ purchase of meat is based on the evaluation of the pre-purchase phase which includes visual quality. Organic buyers show more interest in signs of quality guaranteed by the label. This implies that regulatory attributes significantly influence consumers’ pre-purchase evaluation process. The final best-fit SEM (Figure 6.20) also demonstrates that the product-related construct has a direct effect of 0.28 (Appendix 7 Table B) on the regulatory construct. This supplementary finding suggests that information on organic food products could make them more attractive and influence consumers’ perceived evaluation. This is also found in a European study (Wier & Calverley 2002) where product specific characteristics are one of the major motivations for consumers to purchase organic food. 7.3.3 Lifestyle H3a which hypothesises that the lifestyle related construct has a positive influence on the urban Chinese consumers’ beliefs/attitudes was accepted. The results (Appendix 7) of the final best-fit SEM (Figure 6.20) demonstrate that the lifestyle construct has a total effect of 0.37 on beliefs and attitudes, which comprises a direct effect of 0.24 plus an indirect effect of 0.13 (0.24+0.13=0.37). These findings reveal that the lifestyle construct plays a relatively important role in directly and positively influencing consumers’ personal beliefs and attitudes. The final best-fit SEM retained two lifestyle factors: self-indulgence and variety seeking. The original proposed conceptual framework (Figure 3.2) which contained the ‘opinion leadership’ factor was removed. This 218 category of consumers are opinion leaders among their friends and their community. They do not seek novel and trendy products and are less affected by advertisement of new products. They are also very traditional, less experimental and are not risk takers (Yang 2004). The findings of this study revealed that consumers’ beliefs and attitudes are influenced partially by the two lifestyle segments: variety seeking and self-indulgence. The ‘self-indulgence’ category comprises those who make spontaneous decisions without deliberate thinking, and are those who do what and when they feel like. This category also considers the purpose of ‘making money is to spend it’. Variety seekers are those seeking information, who want variety and the novelty of doing things. They always try something new and unique and enjoy a challenge. They are prepared to take the risk to experience new products and are quite influenced by advertisement. The findings suggest that organic food consumers in urban China are less traditional and are looking for experimental experiences. They are risk takers who are individually centred, who seek novelties and new lifestyles. These findings are in contrast to those of Gil, Gracia and Sanchezs (2000) who suggest that Spanish consumers who display negative attitudes towards the purchase of organic food product generally perceive those products to be fashionable. This study demonstrates that the two lifestyle segments of self-indulgent and variety seeking influence consumers’ beliefs and attitudes towards organic food, suggesting that people who are more self-indulgent and more desirous of variety are more likely to have a positive attitude towards organic food. Previous studies suggest that consumption of organic food reflects a ‘greening of consumer lifestyles’ (Lockie et al. 2002, p. 24). For regular organic food buyers, consumption of organic food is part of a lifestyle, with related interests in nature, society and the environment (Schifferstein & Oude Ophuis 1998). Lockie et al. (2002) suggest stereotypical images of organic consumers as ‘greenies’, ‘health nuts’ or ‘yuppies’, who are more interested in fashion than anything else, hence consumption of organic food reflects a ‘greening’ lifestyle. Organic food consumers in China may belong to this kind of lifestyle. 219 H3b which hypothesises that the Lifestyle construct is positively correlated to the prepurchase evaluation was not supported. The lifestyle-related construct has a total effect of 0.22 on pre-purchase evaluation which comprises a direct effect of 0.06 plus an indirect effect of 0.16 (0.06+0.16=0.22). These findings show a direct relationship between lifestyle and pre-purchase evaluation as being weak, and relatively unimportant. It reveals that the lifestyle construct impacts on the consumer pre-purchase evaluation of organic food through the mediating effects of consumers’ attitudes. However, H3c which hypothesises that the lifestyle construct is positively correlated to the Chinese consumers behavioural/purchase intention was accepted. The results (Appendix 7) demonstrates that the lifestyle construct has a total effect of 0.34 on purchase intention which comprises a direct effect of 0.13 plus an indirect effect of 0.21 (0.13+0.21=0.34). It reveals that the lifestyle construct impacts on the consumer purchase intention of organic food. This is in agreement with studies of Gil, Gracia and Sanchez (2000) and Sanjuán et al. (2003) who concur that the lifestyle attribute is a key factor in explaining organic food consumption. Consumers who are concerned about balanced and healthy lifestyles are more likely to have positive attitudes towards organic food. This then enhances consumers’ purchase intention of organic food. In addition, the results of the final best-fit SEM also demonstrate that the lifestyle construct has a direct effect of 0.29 on the Product-related construct (Appendix 7 Table A). This suggests that the lifestyle related attributes significantly influence the Product-related construct. This result is in line with a Chinese lifestyle study (Zhu et al. 2009) which suggests that there are significant interactive effects between lifestyle influences and product attributes. As such product attributes are a critical factor for understanding consumers with different lifestyles. People with certain lifestyles will be drawn particularly to a set of product attributes because of the perceived functional or emotional relevance of these product attributes to the consumers’ needs. Reid et al. (2001) also report that food 220 related lifestyles are displayed when influenced by novelty and social events. Consumers like to experiment with new flavours and new food styles. 7.3.4 Ethnocentrism H4a which hypothesises that the ethnocentrism construct has a significant influence on the beliefs and attitudes of urban Chinese consumers towards the purchase of organic food was not supported. Noticeably the results (Appendix 7) of the final SEM (Figure 6.20) reveal that there is no significant relationship between the fourth influencing construct, ethnocentrism, and urban Chinese consumers’ beliefs/attitudes. It has a total effect of 0.16 on consumers’ beliefs/attitudes which comprise a direct effect of 0.04 plus an indirect effect of 0.12 (0.04+0.12=0.16). It is evident that there is a very weak relationship (0.04) between ethnocentrism and beliefs/attitude, meaning that it is relatively unimportant. The ethnocentrism construct as such cannot impact urban Chinese consumers’ beliefs/attitudes towards organic food. H4b which hypothesises that the Ethnocentrism construct is significantly correlated to the pre-purchase evaluation was not supported. The results (Appendix 7) of the final best-fit SEM demonstrate that the ethnocentrism construct has a total effect of 0.12 on pre-purchase evaluation which comprises a direct effect of 0.04 plus an indirect effect of 0.08 (0.04+0.08=0.12). This therefore displays a very weak relationship between ethnocentrism and pre-purchase evaluation which is not supported by the hypothesis. H4c which hypothesises that the Ethnocentrism construct is significantly correlated to Chinese consumers’ behavioural/purchase intention was not supported. Furthermore, the ethnocentrism construct has a total effect of 0.16 on purchase intention which comprises a direct effect of 0.02 plus an indirect effect of 0.14 (0.02+0.14=0.16). 221 The direct relationship between ethnocentrism and behavioural/purchase intention is still very weak and hence relatively unimportant. Overall, the foregoing results demonstrate that the ethnocentrism construct does not significantly influence consumers’ beliefs/attitudes, pre-purchase evaluation or purchase intention of organic food in urban China. In contrast, some previous studies suggest that organic food is perceived as being environmentally friendly, hence consumers may prefer to buy less travelled organic products from local sources rather than shipped from overseas (Williams & Hammitt 2000; Bonti-Ankomah & Yiridoe 2006). This has a bearing on support for the locally produced organic foods. Therefore ethnocentric attitudes may develop in organic food consumers owing to their desire of supporting local farmers. ‘Ethnocentric’ category consumers prefer to purchase local and domestically produced organic food products rather than imported ones (Fotopoulos & Krystallis 2002a). Chinese consumers in this study tended to show weak ethnocentric beliefs and attitudes towards organic food. These findings are supported by Hsu and Nien (2008) who suggest that ethnocentrism attitudes towards imported products in big cities tend to be diluted, particularly with those consumers who are more exposed to foreign products. In relation to this, consumers from developing countries such as China sometimes tend to perceive domestic products as being of lower quality (Wang & Chen 2004). It may also suggest that Chinese consumers are not too concerned about the country of origin which features in the five least important attributes of purchasing organic food (Table 6.21). Noticeably, nearly half of the respondents did not mind which country or region the organic food came from. Beyond this, interestingly Australia and New Zealand products had the highest acceptance rates (17.8%) as preferred importing countries compared to the rest of the world (Figure 6.7). This could possibly be due to them being perceived as clean and green as well as the successful presence of New Zealand dairy products in the Chinese market (Sedgman 2011). At the same time the survey results indicate a low awareness and 222 recognition (Figure 6.2) of the Australian Certificated Organic logo (ACO), in comparison to the better recognised US logo. The findings of this study are supported by a recent McKinsey research survey (St-Maurice, Süssmuth-Dyckerhoff & Tsai 2008) which reported that, in 2008, only 30% of respondents trusted Chinese brands, down sharply from 44% in 2007. In the bigger cities, only 13% expressed a clear preference for Chinese brands. This proves that in Chinese first and second tier cities, consumers tend to show no ethnocentrism towards organic food. Although there are alternative organic equivalent products produced in China, the recent spate of food scandals has damaged consumers’ confidence in terms of domestically produced food products. The result from this study on the role of ethnocentrism is also supported by a recent report from Euromonitor International (2011) which concludes that Chinese consumers are losing confidence in locally produced milk after a series of scandals involving food safety. In 2008, melamine-tainted milk power killed six infants and saw over 300,000 children unwell across the country. This food scandal involved one of China’s largest dairy firms SanLu (Euromonitor International 2011; Shumei 2009). The scandal caused Chinese consumers to lose faith in the country’s dairy industry. Another more recent incident involving China’s largest meat processor Shuanghui provoked the government to suspend its activities due to the presence of the illegal additive Clenbuterol9 which was found in the pork meat products (Wu & Zhao 2011; Zhou 2011). The Clenbuterol pork meat scandal led consumers to express concern and disappointment in the meat processor giant. Both SanLu and Shuanghui were considered to be amongst the most reliable brands in China, but now Chinese consumers are unsure which brand they can trust. These issues relating to food safety seem to be paramount in the minds of consumers in the light of the above incidents of food poisoning and food scandals. 9 Clenbuterol is a chemical that can be used to prevent pork meat from accumulating fat. It is banned as an injected additive in pig feed in China as it can end up in the flesh of pigs and is poisonous to humans. 223 7.3.5 Pre-purchase evaluation and behavioural/purchase intention H5 which hypothesises that urban Chinese consumers’ beliefs/attitudes are directly and positively correlated to pre-purchase evaluation was accepted. H6 which hypothesises that the pre-purchase evaluation construct is directly and positively related to behavioural/purchase intention was accepted. The results (Appendix 7 Table A) of the final best-fit SEM (Figure 6.20) demonstrates that (i) consumers beliefs and attitudes have a total effect of 0.20 on pre-purchase evaluation which is both a direct effect and a positive relationship and (ii) that pre-purchase evaluation has a total effect of 0.62 on purchase intention which demonstrates there is a direct effect and strong relationship between pre-purchase evaluation and behavioural/purchase intention. These findings support expectations that consumers’ beliefs/attitudes play an important role in the pre-purchase evaluation. Pre-purchase evaluation ultimately influences the purchase intention. Consumers with a positive attitude to organic foods are more likely to have a positive pre-purchase evaluation. Finally, potential consumers’ positive pre-purchase evaluations will likely translate into purchase intention. After searching and obtaining adequate information related to organic food, potential consumers evaluate this information and then select the best option. This study confirms previous findings, for example, Tarkiainen and Sundqvist (2005) in which attitudes were found to significantly contribute to purchase intention among Finnish consumers. In conclusion, the findings (Table 6.15) of this study reveal that consumers’ attitudes are positively and directly influenced by product-related, regulatory and lifestyle constructs (especially those in the category of variety seeking and self-indulgence). The productrelated construct (Beta=0.38) made the strongest contribution to explain beliefs/attitudes towards organic food. This was followed by regulatory (Beta=0.27) and lifestyle (Beta=0.24) effects. The results also reveal that both product-related and regulatory 224 constructs have significant influences on pre-purchase evaluation. In addition these constructs are associated with behavioural/purchase intention through the mediatory affect of personal beliefs/attitudes and pre-purchase evaluation. Furthermore, personal beliefs and attitudes partially mediated the relationship between lifestyle construct and pre-purchase evaluation. However, the lifestyles construct directly influences behavioural/purchase intention. 7.4 Discussion related to research question two The second research question as previously articulated in Chapter One reads as follows: Evaluate the role of demographic variables and usage patterns in the behavioural /purchase intentions of urban Chinese consumers towards organic food. This research question has been answered by analysis of the control variables of the proposed conceptual model, which are addressed in hypotheses seven. Hypotheses tests were performed on gender, age group, education and income levels, and usage patterns. A detailed discussion on each of these control variables and hypotheses follows: 7.4.1 Gender H7a which hypothesises that women are more likely to purchase organic food was not supported. The findings of this study reveal that consumer’ intentions to purchase organic food do not seem to be influenced by gender. This means that there are no differences in purchase intention towards organic food between women and men in urban China. This finding is similar to previous findings from a Greek (Tsakiridou et al. 2008) and a US (Dahm, Samonte & Shows 2009) study. Noticeably, even though this study sought out demographic variations where possible, the gender percentage of final respondents was 59.7% females and only 40.3% males. The greater participation of females reflecting the same results was 225 evidenced in other studies (Janssen, Heid & Hamm 2009; Krystallis & Chryssohoidis 2005), in which the sample size indicated that more than 70% were female and only 30% male, suggesting that women were predominantly responsible for the purchase of food products in the family. 7.4.2 Age H7b which hypothesises that younger consumers are more likely to purchase organic food than older consumers was not accepted. There is much debate on the importance of age as an influencing factor on consumers’ purchase intention. Young consumers may purchase organic food for reasons of environmental concern, and older consumers due to personal health concerns (Magnusson et al. 2001, 2003; Tsakiridou et al. 2008). The findings of this study indicate that there are no significant differences between age groups in terms of purchase intention relating to organic food. These findings concur with a Swedish study by Magnusson et al. (2001, 2003). Furthermore, an Indian study (Chakrabarti & Baisya 2007) suggests that consumers’ attitudes and motivation towards the purchase of organic food are not influenced by different age groups. In this study, a small proportion of respondents were older than 60 years of age, with the overwhelming number in the younger age category. This may indicate a propensity of younger consumers having money to spend and preferring to shop in supermarkets where organic food products are readily available while older consumers may prefer to shop in other traditional outlets. Almost half of the respondents belonged to the youngest age group (18-30), followed by the 31-45 years old category accounting for 35%, with only 12% in the 46-60 years old category. While recent economic reforms have been implemented and the economic wellbeing of many Chinese has improved significantly, many older consumers aged over 60 still largely depend on retirement pensions and subsidies which places them in a lower income category compared to other age groups. 226 7.4.3 Education and Income levels H7c which hypothesises that the higher the Chinese consumers’ level of education, the more likely they are to purchase organic food was partially accepted. H7d which hypothesises that the higher the Chinese consumers’ income, the more likely are to purchase organic food was partially accepted. The findings of this study reveal that levels of education and income are partially correlated with organic food purchase intention. Consumers with at least high school levels of education tended to have greater intentions to purchase organic food. Meanwhile, higher levels of income earners also indicated a stronger preference in purchase intention towards organic food. These results are supported by previous studies (Roddy, Cowan & Hutchinson 1996; Tsakiridou et al. 2008). Higher education and higher income are an influence on organic consumers’ consumption. It also can be noted that higher educational and higher income groups show high awareness and high purchase intention of organic food. Equally, non-organic users exhibit low education levels and lower income levels as compared to organic users (Fotopoulos & Krystallis 2002a, b; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006). The findings of this and most previous studies suggest that the most affluent people who purchased organic food were well educated. Previous studies (Connor & Douglas 2001; Lea & Worsley 2005; O’Donovan & McCarthy 2002; Tsakiridou et al. 2008) suggest that demographic variables such as gender, age, level of education and income influence consumers’ organic food purchase intention. Partially in agreement with these studies, the current study indicates that gender and age did not influence Chinese consumers’ purchasing intentions towards organic food, but the level of education and income did impact the purchase intention of organic food. 227 7.4.4 Usage patterns H7e which hypothesises that the important attributes of organic food products have positive influences on their intended usage pattern was partially accepted. The results (Table 6.22) of this study reveal that urban Chinese consumers have five intended uses for organic food: own consumption, for children, elderly people, for gift giving and for family consumption. Analysis of the intended usage patterns reveals that there are significant differences in the attributes of social status, awareness and food safety when urban Chinese consumers buy organic food products for their own consumption, for children, for the elderly, for gifts or for their family. Social status is considered to be the most important attribute for the purchase of organic food for gift giving, while own consumption is the least important attribute. Previous studies (Qian, Razzaque & Keng 2007) suggest that social status is important for Chinese gift giving. Personal favours or gift giving can be used as a medium of social exchange for people in reciprocal relationships. Sun and Collins (2006) also reveal that a gift not only reflects the receiver’s social status, but also reflects the giver’s good-will and self-image. An expensive gift offered by Chinese consumers seeks to also imitate superior attributes. Sun and Collins suggest that Chinese consumers seek to benefit from expensive imported fruit as gifts which convey symbolic meaning such as ‘clean’, ‘superior appearance’ and ‘expensive and high prestige’. These attributes project imported fruits as being consumed by wealthy consumers who apparently have high social status. Food can be a symbol of social status (Bareham 1995) and consumers with a high social status may obtain satisfaction because of its novelty and style of food consumption (Fotopoulos, Krystallis & Ness 2003). The awareness of organic food is the most important attribute for gift giving and least important attribute for own consumption as per Tables 6.22. Awareness of organic food means purchasers and consumers understand the differences between organic and nonorganic foods, and appreciate their benefits. As Chinese consumers engage in gift giving as 228 reciprocal favours, the awareness of organic food could improve the receiver’s understanding of the product, and perceive organic food as an expensive product. Obviously, food safety is considered an important attribute for both own consumption and gift giving usage patterns (the means of own consumption and for gift giving are both 4.27 as shown in Table 6.22). This finding is consistent with the study by Krystallis, Fotopoulos and Zotos (2006) who suggest that both organic and non-organic food buyers share the same perception towards food safety issues. The results of this study of intended usages pattern reveal that social status is more important for gift giving and less important for own consumption in China. If organic food purchase for gift giving is the intended usage, then the package appearance and description of the product are essential. The importance of giving gifts in Chinese culture is quite different from western countries. It is of utmost importance to offer gifts in a reciprocal arrangement in undertaking a relationship (business or otherwise). The prestigious nature of the gift can also represent the respect and financial strength of the donor. The gift reflects both the status of the donor and respect towards the receiver (Huliyeti, Marchesini & Canavari 2008). Sun and Collins (2006) report the high price of imported fruit, for gift purposes as it is perceived as a luxury product. As gifts, the high price of products may signify prestige (sign of respect) to the receiver and demonstrates the generosity of the giver (Huliyeti, Marchesini & Canavari 2008). In this study it could be suggested that respondents could not identify significant differences between organic and conventional or even GM food from the appearance. It is recommended that marketers need to be aware that consumers would only understand organic food appearance by improved packaging, identification of the logos and the labels. Most consumers who purchase organic food for their own consumption and family use are not gift givers. Because of the private nature of consumption, quality thus assumes greater importance than packaging. China is described as a highly collectivist society with high power distance and status as important indicators of power and achievement (Hofstede 229 1994). However, the face-saving exercise is important only when purchasing a socially visible product, such as an expensive watch or a luxury car, all of which can demonstrate the owner’s status. The pragmatic Chinese tend to value functionality over appearance for private use of products. Quality in this context is considered as more important than appearance (Schütte & Ciarlante 1998). The presence of children in the household in this study does not have a significant impact on the purchase of organic food products. Even the one child policy in China, which was apparently directed at giving more attention to children from four grandparents and parents, has not resulted in greater levels of consumption of organic food. This finding seems to contradict the UK study (Hill & Lynchehaun 2002) which offered this nexus of organic food with feeding children and with the perceived increase of organic food as a result of the arrival of babies. In the UK study there were increasing numbers of new parents who were buying organic baby food. However, this research study indicates that food safety concern is not just for children, it is for families. Therefore the organic food purchase is not only for the one child, but for the entire family. 7.5 Contribution of this study China experienced dramatic economic development in the last twenty years and its entry into the WTO in 2001 further opened up its huge retail markets to the rest of the world. This study has provided a systematic and extensive analysis and findings of urban Chinese consumers’ purchase behaviour of organic food. As best as can be ascertained, very few studies on organic foods have been undertaken in China. The Chinese economy is experiencing phenomenal growth in the organic food sector. Little is known or understood about Chinese consumers’ attitudes towards organic food products or their associated purchase behaviour. This study is one of the very few dealing with consumers’ buyer behaviour of organic food in urban China. Moreover this study, the first of its kind in obtaining data from the major cities of China, has made significant contributions to both academics and practitioners. 230 7.5.1 Theoretical contributions This study has examined the main determinants which influence consumption of organic food in mainland China, and has developed and tested a distinctive conceptual model of organic food consumption intentions in the Chinese context. The unique conceptual model (Figure 7.2) applied in this study is the integration of three theoretical models, namely, Consumer Decision Process (CDP), Hierarchy of Effects and Theory of Planned Behaviour (TPB). The results provide empirical evidence to verify the applicability of these three theories for purchase intention of organic food in urban China. It examined the influencing stage, which comprised constructs related to product, regulatory, lifestyle and ethnocentrism. These four constructs influence consumers’ beliefs and attitudes, their prepurchase evaluation and finally their organic food purchase intention. The final best-fit structural equation model obtained in this study suggests that: (i) urban Chinese consumers’ beliefs/attitudes towards organic food consumption are positively and directly influenced by product-related, regulatory and lifestyle constructs (especially variety seeking and selfindulgence); (ii) both product-related and regulatory constructs had significant influences on pre-purchase evaluation, and these constructs were associated with behavioural/purchase intention through the mediatory affect of beliefs/attitudes and pre-purchase evaluation; and (iii) some demographic variables such as level of income and education partially influence organic food consumption. Also some important attributes of organic food product influence consumers’ intended usage pattern. 231 232 This study has contributed a new body of knowledge to the growing research in the field of organic food purchase behaviour and adds to the current literature, which is well developed in Western European and North American countries. It has also provided valuable insights into current studies of global and Chinese consumer behaviour towards organic food. Most importantly, it has addressed consumers’ purchase intention towards organic food in the Chinese context. It provides a new understanding of organic food consumers, and has related implications for global and Chinese organic food studies. China has developed urban districts over the decades and the move from the countryside continues insistently. Understanding Chinese urban consumers is crucial to understanding Chinese society. This study has examined organic food product-related attributes and Chinese consumers’ lifestyle-related aspects. In addition it has examined the regulatory factors related to government policies and their impact on consumers’ attitudes towards organic food purchase. This study provides a political, economic, and social-cultural understanding of Chinese urban consumers and their consumption behaviour, especially towards food products. Hence, it provides a strong empirical contribution to research in the Chinese context. Theoretical contributions are often a primary objective of academic research, but practitioners may develop or propose a set of relationships that are as complex and interrelated as any academically based theory. Thus, researchers from both academia and industry can benefit from the unique analytical tools provided by the Structural Equation Model (SEM). A final best-fit model was developed which examined various factors and their interrelationships and this may prove beneficial to both academics and practitioners. 7.5.2 Business implications The findings of this research have important implications for marketers of organic food in China. They should attempt to leverage on these findings by educating their target audience (both existing and potential customers) and by promoting trials of organic food products. 233 The rise of globalisation and the importance of foreign markets for the business environment requires an understanding of purchase and consumption behaviour. This applies to all sectors, including the organic food industry. Valuable insights into global consumer behaviour have emerged in this study in terms of organic foods building on a new body of knowledge with regard to the potential of organic food business in China. Beneficiaries of this study include various stakeholders in China and globally such as consumers, vendors both local and international and government agencies. The results of this study provide some valuable marketing implications, revealing a number of factors that affect the purchase intention of organic food consumers in China. Organic food sectors, including food marketers who want to develop their market in China, will find the analysis and findings of this study valuable. It has identified important factors and their elements which influence consumers’ purchase behaviour of organic food in China. As such these findings may give future directions to policy marketers in developing strategies aimed at promoting such products. It is evident from the results of this study that supermarkets in the larger Chinese cities are the main retail distribution outlets of organic food. The convenience of shopping in a supermarket is especially attractive for younger Chinese consumers and also for potential new buyers. The future of increased organic food availability, of distribution channels, particularly with speciality stores, as well as organic farms is becoming more prevalent in China. China has developed one of the largest online shopping markets – known as taobao.com – which is gaining importance. It is suggested that online shopping could be a future direct market strategy. This study highlights the increasing awareness of organic food products, and their availability through different types of retail outlets. There is also a need to enhance the inspection and certification of organic food labelling as well as to ensure that labelling and logos are a sign of actual quality. 234 7.5.2.1 Importance of the product-related construct to marketers One of the important business implications of this study is to provide information to marketers about organic food product attributes. The results suggest that consumers perceive organic food to appear and taste better than conventional food. Some organic food products, as a result of less chemical use during production, transportation and storage, seemingly do not appear as good as conventionally grown food products. Consumers may not purchase organic food if it does not appear fresh. Finding ways of preserving organic food for a longer shelf life is a challenge for growers and marketers. Organic food is categorised as fresh or processed produce. It is suggested that with fresh organic food, such as vegetables, fruit and meat, a fresher looking product is perceived as being of higher quality in urban China. This study reveals that the higher price of organic food as compared to conventional food does not deter Chinese consumers from purchasing. Producers and marketers of organic food products can use this information in establishing and setting their prices, generally in the Chinese context the higher price signifies an image of safe, clean and high quality. The price differentiator might be used to maintain organic products’ niche marketing strategy. In China, prices of certified organic food and green food can be up to 700% higher than those for conventional products (Kluger 2010). Overall price sensitivity seems less important in terms of purchase of organic food as a result of food safety issues. On the other hand, the results of this study indicate that retail price premiums remain higher than what most consumers are willing to pay. More than half of the respondents (60.8%) were willing to pay an extra 20% to 50%, nearly 10% were willing to pay an extra 51% to 100%, and 1% were willing to pay an extra premium greater than 100%. Approximately one quarter of respondents did not wish to pay extra for the organic products. Emerging from the study was the consumers’ preferred premium payment of an extra 20% to 50%. Therefore an organic food premium of an extra 20% to 50% is suggested as the acceptance price range. These findings are supported by Yin et al. (2010) who suggest that Chinese 235 consumers are willing to pay an extra 35.3% for organic food. The organic food consumers in China have similar views on price as compared to those in Western Europe and North America (Lockie, Halpin and Pearson (2006). Although Chinese consumers traditionally seek value for money, they are prepared to be more tolerant of the higher cost of organic food. The very public food scandals in China have made consumers tolerate paying a higher price for organic food where it stands as a symbol of safety. Organic food products are strategically important to some retailers, enabling them to practice and implement a differentiation marketing strategy. This is due to consumers’ perceived view of organic food as being environmentally friendly and of high quality (Aertsens, Mondelaers & Huylenbroeck 2009). The findings of this study are in line with previous studies, which suggest that organic food is perceived as being safe and of higher quality. Organic food marketers may well emphasise these beneficial attributes, such as higher nutritional values, freshness and being environmentally friendly. The perception of organic food being safe and of high quality might be reflected in packaging and promotion material. 7.5.2.2 Importance of the regulatory construct to policy makers The term ‘organic’ has many different interpretations, and previous studies indicate that consumers seem to lack knowledge about organic standards. The ‘Chinese green food’ label had the highest rate of recognition by Chinese consumers, followed by ‘Chinese nonpolluted food’ and ‘Chinese organic food’ logo. Organic food logos in this study were chosen from both China and overseas. The overseas organic food logos were less recognised by Chinese consumers. Li, Cheng and Ren (2005) suggest that Chinese consumers have less awareness of the term ‘organic’ as evidenced in this study. Organic food is still a new concept and not very well- known in China (Thøgersen & Zhou 2012). Rising personal income and a growing abundance of material goods in China has significantly improved consumers’ standard of living and quality of life. A series of food 236 scandals has triggered higher consumer demand for ‘safe food’ (including organic food, green food and non-polluted food). There is a clear need to educate consumers regarding the differences between organic, conventional and GM types of food in the market place. Overall there is a lack of trust in the quality credentials of organic food as well as doubt in the enforcement of quality by Chinese authorities. Government agencies and industry need to be aware of the necessity to continually regulate the organic food market. They also need to enhance the inspection and certification of organic food labelling as well as ensure that the labelling and logos guarantee quality. The findings of this study reveal that the regulatory-related construct influences consumers’ attitudes and their pre-purchase evaluation towards organic food. Organic producers and government agencies should make a concerted effort towards improving the awareness and benefits of consuming organic food. Relevant stakeholders of this industry need to implement market strategies aimed at enhancing consumers’ confidence and trust in organic labelling. The findings of this research may also be useful to public policy makers who are interested in identifying strategies to increase awareness and demand for organic food. The recent food scandals in China have made consumers more sceptical of manufacturers’ claims, and they doubt the reliability of certification bodies and government agencies. They are more suspicious of organic food products in comparison to consumers from many other countries. The Chinese government and producers should improve certification and inspection systems to regain consumers’ trust and confidence. Certainly, there is a need for government agencies to give clear and safe advice on food safety issues, as well as a need for organic food certification bodies and marketers to ensure that the certification labelling and explanations are clear and comprehensive. 7.5.2.3 Lifestyle As regards lifestyle, the findings of this study reveal that Chinese organic food consumers tend to be more self-indulgent and variety seeking and prefer product hedonic attributes. 237 They love to try something new and trendy, and often show impulse buying behaviour. Therefore they would be prepared to try more and enjoy the organic food product attributes, such as fragrance, texture and nutrition. For regular organic food consumers, their purchase and consumption are part of today’s modern lifestyle. The traditional Chinese diet which primarily includes rice and vegetables and only moderate amounts of pork, poultry and seafood provides for a relatively healthy lifestyle. However, many newly affluent urban consumers are doing away with this traditional diet and are becoming Western adopters. Looking at today’s China, there are more than 169 million overweight adults, and an additional 26 million who are obese (Zhang et al. 2008). Lifestyle segmentation is a critical step to better understand existing and potential organic food consumers, and also nonorganic food buyers. It is recommended that organic food marketers emphasise both sensory and non-sensory organic food attributes to the different segments of the market based on Chinese consumers’ lifestyle categories. Organic food marketers should endeavour to focus on the variety seeking and self-indulgent consumers with enhanced market strategies. Additionally they could possibly convert consumers of the opinionleadership lifestyle category to be buyers of organic food. This category is more traditional and capable of influencing others. Marketing strategies such as word-of-mouth promotion and direct marketing could be used to promote and distribute organic food products. 7.5.2.4 Country of origin and impact of ethnocentrism Chinese consumers’ ethnocentric attitudes towards imported goods have undergone changes in the recent past, especially in relation to goods originating from developed western countries. When China opened its door to the world in 1978, imported goods were scarce and unknown. Over time this reality has changed. Consumers expressed positive attitudes towards imported goods after many decades in which products were unavailable in China (Sun and Collins 2002). Over the decades, China’s economy has soared, with joining the WTO in 2001 and hosting the 2008 Olympic Games. In turn consumers are gaining back their national pride back. Today China is the second largest economy in the world even though its per capita GDP is relatively low. Chinese consumers no longer perceive 238 imported products as symbols of high quality, as they may have done in the past. However, in the last few years, owing to a series of food scandals and environmental disasters, they have lost some confidence in purchasing domestically produced organic food. This study reveals that ethnocentrism does not influence Chinese consumers’ organic food purchase behaviour. This may suggest that locally produced organic food would not necessarily have the upper hand. Hence the finding may provide significant export business opportunities for international organic firms. Clearly, there is room for marketing international organic food products in the Chinese market. The main exporters of organic foods to China include the United States, Australia and the European Union countries (International Trade Centre 2011). This research reveals that nearly half of the respondents did not mind which country or region the organic foods came from. However, if they were purchasing imported organic food, they indicated their preference for Australian and European products rather than Japanese or American. This might reflect in part Chinese consumers’ animosity towards Japanese products and possibly the rivalry which exists with the United States (Klein, Ettenson & Morris 1998). The results of this study also reveal that Chinese consumers are more familiar with American certified logos rather than European and Australian ones. It seems to suggest that Australian and European organic food producers need to increase the familiarity of their products and brandings within the Chinese market. Australia is often regarded as a ‘lucky country’ with vast spaces, a cleaner environment and has the image of being ‘green and clean’. Not surprisingly it has the largest organic area under cultivation in the world and offers the ideal image for promoting organic products to overseas markets. With this in mind, Australian organic producers should give due importance to the burgeoning Chinese market. It may suggest that in order to gain access to a larger share of this market, they must address the relatively low understanding of the Australian certification procedures. 239 7.5.2.5 China is not a single market The findings of this study concur with previous studies that China is not simply one big uniform market. It is heterogeneous and diverse from north to south and from east to west. Companies that attempt to develop the Chinese organic product market need to understand the varying aspects of consumer behaviour across the country. Consumers’ knowledge and awareness is crucial to expansion of organic marketing, but such awareness in China is relatively low. There needs to be an increase in the availability of distribution channels and there is a need to publicise the overall quality process of organic food related to food safety issues. Also there is a need to educate consumers to understand the value of organic food relative to its price. This study concurs with the findings of another study by Cui and Liu (2000) which reports that Chinese consumers from various regions differ significantly in terms of attitudes, lifestyles and consumption patterns. Attitudes tend to be influenced more by productrelated attributes in Shanghai (eastern region) and Shenzhen (southern region) where consumers place greater emphasis on organic food texture, fragrance, nutritional value as well as its fashionable trend setting. On the other hand, consumers from Beijing tend to be influenced more by regulatory related items. As the capital city in China, consumers with access to key government agencies tend to be more aware of government regulations and policies. These consumers generally show more concern regarding the certification and regulatory authorisation of logos. On the other hand, consumers in Chengdu (western region) tend to associate lifestyle with the purchase behaviour of organic food. Therefore, it is suggested that international companies which attempt to enter the Chinese market must understand the consumer purchase behaviour patterns in the different regions. As Fairbank and Goldman (1998) observe, economic and social differences are deepened by the accelerating geographic disparities between the coastal area involved in international trade versus the poorer inland provinces. Cui and Liu (2000) suggest that regional markets 240 in the south and east represent China’s ‘growth markets’, as they are more advanced economically and have more affluent consumers than the hinterland provinces. 7.6 Limitations of this study There is no scientific evidence which clearly and unambiguously demonstrates that organic food is scientifically healthier than the so-called conventional food. This is an inhibiting element in all organic food studies, and one that this study could not address. Some people in China have never heard of the term ‘organic’ although the respondents in this study were generally aware of it. Respondents were divided into four consumer segments according to their buying frequency: frequent buyers, medium buyers, occasional buyers and non-buyers. Frequent buyers are those who purchase organic food weekly; medium buyers are those who purchase every fortnight or monthly. Nearly half of the respondents indicated they were occasional buyers. Noticeably, nearly 23% of the respondents were in the category of ‘non-buyers but aware’ of organic food. This market segment should be targeted in the future to investigate how they could be converted. The results of this study indicate that 77.5% of the respondents had purchased organic food at some point in their life, but it was unclear whether they had actually purchased ‘organic’ products or some other Chinese categorised safe food (such as green food or non-polluted food). Some consumers did not recognise the logos of the organic foods and instead indicated they had purchased food with logos indicating they were ‘green’ or ‘non-polluted’. It is abundantly clear that confusion still exists among many Chinese consumers. A need for further clarification between ‘organic’, ‘green’ and ‘non-polluted’ food has been identified in the Chinese market. Due to time and financial constraints, this study was only able to collect a limited amount of responses (964 valid responses). The data was based primarily on that collected from four selected cities. These included cities in the north, south, west and east which has allowed insights into the understanding of Chinese consumers’ organic food purchase 241 behaviour as a cross-national phenomenon. The findings from the invariant tests applied to data from the four cities confirm that China is not a homogenous market. This implies that consumers living in the north, south, west and east of China show significantly different influencing dimensions and attitudes towards purchasing organic food. China has become the largest manufacturing hub in the world and as a result of its intense economic growth and development has sought to limit the activities around environmental controls. In the west there is a clearer concern for climate change, pollution levels and other environmental issues. Another limitation for further studies in the Chinese market is the different level of understanding of environmental concerns, as evidenced by the serious pollution levels in the larger cities. This study does not take into account the environmental concerns of the Chinese government and consumers and how these impact the purchase behaviour of organic food. 7.7 Future research directions As organic food has only been recently introduced in China, and mainly available in supermarkets of the larger cities, its awareness among the general consumers is relatively low. As economic development continues unabated in China, future research in this market and this sector could result in consumers being ‘more aware’, thus enhancing the results and findings of such a study. Secondly, China is an organic innovator. The three categories which can be summarised as organic, green and non-polluted food are generally perceived as being safe in China. As awareness of these terms becomes more widespread, future research may focus on the differences of those consumer groups and examine the switching behaviour of the Chinese consumers between these three categories of food. Thirdly, China has developed one of the largest consumer markets with a population of 1.3 billion and annual GDP growth of around 8% to 9% in an area covering 9,600,000 square kilometres. Future research should endeavor to collect data from a greater number of 242 locations in order to enhance the geographical generalisations. It would be ideal to choose more cities from each region as there are over 70 large and medium sized cities in China (National Bureau of Statistics of China 2011). Moreover, China has demonstrated itself to be a heterogeneous market, with both similarities and differences between north, south, east and west. Future studies may develop different best-fit SEM to suit the diversification of the Chinese market. Fourthly, organic consumers believe that organic food is an environmentally related product. As the awareness is increased and there is greater emphasis on environmental issues in China, a future study may investigate how environmental factors impact consumers’ purchase of organic food. Fifthly, actual purchase behaviour was evaluated by the purchase frequency and usage patterns through consumers’ past experience. Longitudinal studies to ascertain the changes which occur in time with Chinese consumer preferences, attitudes and behaviour in relation to organic food would be recommended. Finally, this study has evidenced the dissimilarity between Chinese organic consumers from the western literature. As Chinese consumers are becoming more affluent, their buying behaviour is evolving alongside economic development. The future of research direction should also include studying the cross-cultural comparison of organic food purchasing behaviour. Data would be collected in China and other emerging markets as well as the developed countries. It would be interesting to know what the differences between those consumers’ attitudes towards food attributes, regulatory, lifestyle, ethnocentrism factors are as well as their intentions related to purchase of organic foods. This would elicit those factors that exert the strongest influence on their behaviour with regard to purchasing organic food. 243 7.8 Concluding remarks In the last two decades, organic food has assumed greater global awareness. Organic products have been perceived as being more nutritious, healthier, safer and environmentally friendly. They contain fewer chemical residues and are considered to taste better than conventional food and as such consumers might consider paying a premium price for them. In contrast to the above perceptions, other studies have revealed that there are also barriers to consumers purchasing organic food. China has experienced dramatic economic growth which has accelerated the pace of its urbanisation with more than half a billion consumers living in urban areas. With the increasing quality of life in urban China, consumers also desire to pursue the choice for quality food. On the other hand, rapid economic development has seen food standards suffer, thus triggering a series of food scandals with calls for ‘safer food’. Consumer attitudes towards organic food and decision-making in choosing organic food have been well documented across much of the developed world. Only a handful of studies have been conducted in Asian countries, and even fewer in China. A five stages conceptual framework was proposed to investigate consumer decision-making towards organic food in urban China. This framework was underpinned in three models: CDP, Hierarchy of Effects and TPB. The five stages of the conceptual framework were influencing, cognitive/affective, evaluation of alternatives, behavioural intentions and purchase. The emphasis was placed on the influencing aspects and dimensions of this process in the Chinese context. The four constructs, namely product, regulatory, lifestyle and ethnocentrism, were operationalised in the influencing stage. This is the first study of its kind in the organic food consumption research, which has integrated three well known models to investigate consumer buyer behaviour. The findings of this study have revealed that Chinese consumers’ attitudes towards organic food are influenced by the dimensions of product, regulatory and partially lifestyle (variety 244 seeking and self-indulgence). Additionally it was revealed that all the dimensions and their components mentioned above significantly impacted the behavioural intentions for the purchase of organic food in urban China through the mediating effects of consumers’ attitudes and their pre-purchase evaluation. The influence of demographic and segmentation variables was also examined. Chinese urban consumers now contribute to being one of the largest consumer markets in the world. Studying their buying behaviour is critical for understanding the overall Chinese consumer profile. Results obtained by the final best-fit model have demonstrated that Structural Equation Modelling is an efficacious statistical technique to address various issues associated with the purchase behaviour of organic food products. This model has the capacity to simultaneously examine various factors and their interrelationships which influence consumer decision making towards the purchase of organic food. The theoretical and practical business implications significantly contributed to a further understanding of urban Chinese consumers’ purchase behaviour of organic food products. This study has sought to contribute to this understanding. 245 REFERENCES Aarset, B, Beckmann, S, Bigne, E, Beveridge, M et al. 2004, ‘The European consumers’ understanding and perceptions of the “organic” food regime: the case of aquaculture’, British Food Journal, vol. 106, pp. 93-105. Abbey, K 2005, Going organic: your guide to a healthier life, revised edn, New Holland, Sydney. Abrams, K, Meyers, C & Irani, T 2010, ‘Naturally confused: consumers’ perceptions of allnatural and organic pork products’, Agriculture and Human Values, vol. 27, no. 3, pp. 365-374. Aertsens, J, Mondelaers, K & Huylenbroeck, GV 2009, ‘Differences in retail strategies on the emerging organic market’, British Food Journal, vol. 111, no. 2, pp. 138-154. Aertsens, J, Verbeke, W, Mondelaers, K & Huylenbroeck, GV 2009, ‘Personal determinants of organic food consumption: a review’, British Food Journal, vol. 111, no. 10, pp. 1140-1167. Agbonifoh, BA & Elimimian, JU 1999, ‘Attitudes of developing countries towards ‘country-of-origin’ products in an era of multiple brands’, Journal of International Consumer Marketing, vol. 11, no. 4, pp. 97-116. Aguirre, JA 2007, ‘The farmer’s market organic consumer of Costa Rica’, British Food Journal, vol. 109, no. 2, pp. 145-154. Ahmad, S & Juhdi, N 2010, ‘Organic food: a study on demographic characteristics and factors influencing purchase intention among consumers in Klang Valley, Malaysia’, International Journal of Business and Management, vol. 5, pp. 105-118. Ajzen, I 1991, ‘The theory of planned behavior’, Organizational Behavior and Human Decision Processes, vol. 50, no. 2, pp. 179-211. Ajzen, I 2001, ‘Nature and operation of attitudes’, Annual Review of Psychology, vol. 52, pp. 27-58. Ajzen, I 2005, Attitudes, personality, and behavior, 2nd edn, Open University Press, New York. 246 Ajzen, I & Albarracin, D 2007, ‘Predicting and changing behavior: a reasoned action approach’ in I Ajzen, D Albarracίn and R Hornik (eds) Prediction and change of health behavior: applying the reasoned action approach, Lawrence Erlbaum Associates, Mahwah, NJ. Ajzen, I & Fishbein, M 1969, ‘The prediction of behavioral intentions in a choice situation’, Journal of Experimental Social Psychology vol. 5, no. 4, pp. 400-416. Ajzen, I & Fishbein, M 1980, Understanding attitudes and predicting social behavior, Prentice Hall, Englewood Cliffs, NJ. Allen, P & Bennett, K 2010, PASW statistics by SPSS: a practical guide: version 18.0, CENGAGE Learning, Melbourne. Anders, S & Moeser, A 2008, ‘Assessing the demand for value-based organic meats in Canada: a combined retail and household scanner-data approach’, International Journal of Consumer Studies, vol. 32, no. 5, pp. 457-469. Anderson, JC & Gerbing, DW 1988, ‘Structural equation modeling in practice: a review and recommended two-step approach’, Psychological Bulletin, vol. 103, no. 3, pp. 411423. Anić, I-D 2010, ‘Attitudes and purchasing behavior of consumers in domestic and foreign food retailers in Croatia’, Preliminary communication, vol. 28, no. 1, pp. 113-133. Armitage, CJ, Conner, M & Norman, P 1999, ‘Differential effects of mood on information processing: evidence from the theories of reasoned action and planned behaviour’, European Journal of Social Psychology, vol. 29, pp. 419-433. Arvanitoyannis, IS, Krystallis, A & Kapirti, A 2003, ‘Health and environmental consciousness: Greek consumers’ attitudes toward the organic, HACCPand ISO14000 certifications on food’, Journal of International Food & Agribusiness Marketing, vol. 15, no. 1/2, pp. 93-136. Atsmon, Y, Ding, J, Dixit, V, Leibowitz, G, Magni, M & Zipser, D 2009, 2009 Annual Chinese consumer study, McKinsey Asia, viewed 15 August 2010, 247 <http://www.mckinsey.com/locations/greaterchina/McKinsey_Annual_Consumer_Rep ort_Downturn_part2.pdf>. Atsmon, Y & Dixit, V 2009, Understanding China’s wealthy, viewed 15 August 2010, <http://www.mckinseyquarterly.com/Understanding_Chinas_wealthy_2392>. Austrade 2011, China country overview: economic climate, viewed 15 March 2011, <http://www.austrade.gov.au/China-Market-Overview/default.aspx>. Baer, N 2007, The spread of organic food in China, Woodrow Wilson International Center for Scholars, Washington, DC. Baker, S, Thompson, KE & Engelken, J 2004, ‘Mapping the values driving organic food choice: Germany vs the UK’, European Journal of Marketing, vol. 38, no. 8, pp. 9951012. Bao, Y, Zhou, KZ & Su, C 2003, ‘Face consciousness and risk aversion: do they affect consumer decision-making?’, Psychology and Marketing, vol. 20, no. 8, pp. 733-755. Bareham, J 1995, Consumer behaviour in the food industry: a European perspective, Butterworth-Heinemann, Oxford. Barnes, AP, Vergunst, P & Topp, K 2009, ‘Assessing the consumer perception of the term “organic”: a citizens’ jury approach’, British Food Journal, vol. 111, no. 2, pp. 155164. Barry, TF & Howard, DJ 1990, ‘A review and critique of the Hierarchy of Effects in advertising’, International Journal of Advertising, vol. 9, no. 2, pp. 121-135. Batra, R, Ramaswamy, V, Alden, DL, Steenkamp, J-BEM & Ramachander, S 2000, ‘Effects of brand local and nonlocal origin on consumer attitudes in developing countries’, Journal of Consumer Psychology, vol. 9, pp. 83-95. Beharrell, B & MacFie, JH 1991, ‘Consumer attitudes to organic foods’, British Food Journal, vol. 93, no. 2, pp. 25-30. Beijing Municipal Government 2010, Residing in Beijing, viewed 10 August 2010, <http://www.ebeijing.gov.cn/feature_2/ResidinginBeijing/>. 248 Bellows, AC, Onyango, B, Hallman, WK & Diamond, A 2008, ‘Understanding consumer interest in organics: production values vs. purchasing behavior’, Journal of Agricultural & Food Industrial Organisation, vol. 6, no. 1, pp. 1-31. Bennett, R 1999, ‘Sports sponsorship, spectator recall and false consensus’, European Journal of Marketing, vol. 33, no. 3-4, pp. 291-313. Bhaskaran, S & Hardley, F 2002, ‘Buyer beliefs, attitudes and behaviour: foods with therapeutic claims’, Journal of Consumer Marketing, vol. 19, no. 7, pp. 591-606. Bhaskaran, S, Polonsky, M, Gary, J & Fernandez, S 2006, ‘Environmentally sustainable food production and marketing: opportunity or hype?’, British Food Journal, vol. 108, no. 8, pp. 677-690. Blackwell, RD, Miniard, PW & Engel, JF 2006, Consumer behavior, 10th edn, Thomson South-Western, Boston. Bollen, KA 1989, Structural equations with latent variables, John Wiley, Hoboken, NJ. Bonini, SMJ & Oppenheim, JM 2008, ‘Helping “green” products grow’, McKinsey Quarterly, Oct., pp. 1-7. Bonti-Ankomah, S & Yiridoe, EK 2006, Organic and conventional food: a literature review of the economics of consumer perceptions and preference, Organic Agriculture Centre of Canada, Nova Scotia Agriculture College, Truro, NS. Botonaki, A, Polymeros, K, Tsakiridou, E & Mattas, K 2006, ‘The role of food quality certification on consumers’ food choices’, British Food Journal, vol. 108, no. 2/3, pp. 77-90. Bower, AB 2001, ‘Highly attractive models in advertising and the women who loathe them: the implications of negative affect for spokesperson effectiveness’, Journal of Advertising, vol. 15, no. 3, pp. 51-63. Boyce, J 2002, Market research in practice, McGraw-Hill, Sydney Bredahl, L, Grunert, KG & Frewer, LJ 1998, ‘Consumer attitudes and decision-making with regard to genetically engineered food products: a review of the literature and a 249 presentation of models for future research’, Journal of Consumer Policy, vol. 21, no. 3, pp. 251-277. Brennan, C, Gallagher, K & McEachern, M 2003, ‘A review of the “consumer interest” in organic meat’, International Journal of Consumer Studies, vol. 27, no. 5, pp. 381-394. Browne, MW & Cudeck, R 1992, ‘Alternative ways of assessing model fit’, Sociological Methods & Research, vol. 21, no. 2, pp. 230-258. Brunsø, K & Grunert, K 1995, ‘Development and testing of a cross-culturally valid instrument: food-related life style’, Advances in Consumer Research, vol. 22, pp. 475480. Bryman, A & Bell, E 2007, Business research methods, 2nd edn, Oxford University Press, Oxford. Burton, S, Lichtenstein, DR, Netemeyer, RG & Garretson, JA 1998, ‘A scale for measuring attitude toward private label products and an examination of its psychological and behavioral correlates’, Journal of the Academy of Marketing Science, vol. 26, no. 4, pp. 293-306. Byng, J 1993, ‘EC organic food standards’, British Food Journal, vol. 95, no. 1, pp. 16-17. Byrne, BM 2004, ‘Testing for multigroup invariance using AMOS graphics: a road less traveled’, Structural Equation Modeling, vol. 11, no. 2, pp. 272-300. Byrne, BM 2010, Structural equation modeling with AMOS: basic concepts, applications, and programming, 2nd edn, Taylor & Francis, London. Cai, H-J, Lin, Y-J, Wu, Q-P, Yan, L & Huang, X-F 2008, ‘Examining the measurement invariance between paper-and-pencil and Internet-administered tests in China’, Acta Psychologica Sinica, vol. 40, no. 2, pp. 228-239. Callick, R 2011, ‘Special report: China: growing partnership’, Australian, 6 June. Canavari, M, Bazzani, GM, Spadoni, R & Regazzi, D 2002, ‘Food safety and organic fruit demand in Italy: a survey’, British Food Journal, vol. 104, no. 3-5, pp. 220-232. Carroll, JB (ed.) 1974, Language, thought and reality: selected writings of Benjamin Lee Whorf, MIT Press, Cambridge, Mass. 250 Carson, R 1962, Silent spring, Houghton Mifflin, Boston. Cavana, RY, Delahaye, BL & Sekaran, U 2001, Applied business research qualitative and quantitative methods, John Wiley, Brisbane. CGFDC 2009, About China Green Food Development Center, viewed 30 March 2009, <http://www.greenfood.org.cn/sites/GREENFOOD/List_3675_3811.html>. Chakrabarti, S & Baisya, RK 2007, ‘Purchase motivations and attitudes of organic food buyers’, Decision, vol. 34, no. 1, pp. 1-22. Chan, RYK 2001, ‘Determinants of Chinese consumers’ green purchase behaviour’, Psychology & Marketing, vol. 18, no. 4, pp. 389-413. Chang, H-S & Kristiansen, P 2006, ‘Selling Australia as “clean and green”’, Australian Journal of Agricultural & Resource Economics, vol. 50, no. 1, pp. 103-113. Chang, H-S & Zepeda, L 2005, ‘Consumer perceptions and demand for organic food in Australia: focus group discussions’, Renewable Agriculture and Food Systems, vol. 20, no. 3, pp. 155-167. Chang, H-S, Zepeda, L & Griffith, G 2005, ‘The Australian organic food products market: overview, issues and research needs’, Australian Agribusiness Review, vol. 13, paper no. 15. Chen, M-F 2007, ‘Consumer attitudes and purchase intention in relation to organic foods in Taiwan: moderating effects of food-related personality traits’, Food Quality & Preference, vol. 18, pp. 1008-1021. Chen, M-F 2009, ‘Attitude toward organic foods among Taiwanese as related to health consciousness, environmental attitudes, and the mediating effects of a healthy lifestyle’, British Food Journal, vol. 111, no. 2, pp. 165-178. Chesneaux, J 1979, China: the People’s Republic, 1949-1976, Harvester, Brighton. China Internet Network Information Center 2009, China Internet Statistics Report (in Chinese), viewed 20 July 2010, <http://www.cnnic.cn/uploadfiles/pdf/2009/7/16/125126.pdf>. 251 Chinnici, G, D’Amico, M & Peorino, B 2002, ‘A multivaritate statistical analysis on the consumers of organic products’, British Food Journal, vol. 104, no. 3/4/5, pp. 187-199. Chryssochoidis, G 2000, ‘Repercussions of consumer confusion for late introduced differentiated products’, European Journal of Marketing, vol. 34, no. 5/6, pp. 705-722. Chryssochoidis, G, Krystallis, A & Perreas, P 2007, ‘Ethnocentric beliefs and country-oforigin (COO) effect: impact of country, product and product attributes on Greek consumers’ evaluation of food products’, European Journal of Marketing, vol. 41, no. 11/12, pp. 1518-1544. Chryssochoidis, GM & Krystallis, A 2005, ‘Organic consumers’ personal values research: testing and validating the list of values (LOV) scale and implementing a value-based segmentation task’, Food Quality & Preference, vol. 16, pp. 585-599. Cicia, G, Giudice, TD & Scarpa, R 2002, ‘Consumers’ perception of quality in organic food: a random utility model under preference heterogeneity and choice correlation from rank-orderings’, British Food Journal, vol. 104, no. 3/4/5, pp. 200-213. CMP Information Ltd 2006, ‘Global view: Chinese “green food” revolution is growing’, Arable Farming, 10 July, p. 40. Cohen, J & Cohen, P 1983, Applied multiple regression/correlation analysis for the behavioral sciences, 2nd edn, Lawrence Erlbaum Associates, Hillsdale, NJ. Collis, J & Hussey, R 2009, Business research: a practical guide to undergraduate and postgraduate students, 3rd edn, Palgrave Macmillan, Basingstoke, UK Connor, R & Douglas, L 2001, ‘Consumer attitudes to organic foods’, Nutrition and Food Science, vol. 31, no. 4/5, pp. 254-264. Courville, S 2006, ‘Organic standards and certification’ in P Kristiansen, A Taji and J Reganold (eds) Organic agriculture: a global perspective, CSIRO Publishing, Melbourne. Creswell, JW 2009, Research design: qualitative, quantitative, and mixed method approaches, 3rd edn, Sage, Thousand Oaks, Calif. 252 Croll, E 2006, China’s new consumers: social development and domestic demand, Routledge, London. Cronbach, LJ 1951, ‘Coefficient alpha and the internal structure of tests’, Psychometrika, vol. 16, no. 3, pp. 297-334. Cronin Jr, JJ, Brady, MK & Hult, GTM 2000, ‘Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments’, Journal of Retailing, vol. 76, no. 2, pp. 193-218. Cui, G & Liu, Q 2000, ‘Regional market segments of China: opportunities and barriers in a big emerging market’, Journal of Consumer Marketing, vol. 17, no. 1, pp. 55-72. Cunningham, E 2008, A practical guide to structural equation modelling using Amos, Statsline, Melbourne. Dahm, MJ, Samonte, AV & Shows, AR 2009, ‘Organic foods: do eco-friendly attitudes predict eco-friendly behaviors?’, Journal of American College Health, vol. 58, no. 3, pp. 195-202. Dai, Y-c, Zhu, B & Ying, R-y 2006, ‘Consumers’ choice on food safety: a case study of organic vegetable purchasing behaviour in Nanjing’, Journal of Nanjing Agricultural University (Social Science Edition), vol. 6, no. 1, pp. 47-52 (in Chinese). Davies, A, Titterington, AJ & Cochrane, C 1995, ‘Who buys organic food ? A profile of the purchasers of organic food in Northern Ireland’, British Food Journal, vol. 97, no. 10, pp. 17-23. Davis, D 2005, Business research for decision-making, 6th edn, Thomson/Brooks/Cole, Belmont, Calif. de Magistris, T & Gracia, A 2008, ‘The decision to buy organic food products in Southern Italy’, British Food Journal, vol. 110, no. 9, pp. 929-947. DeVellis, RF 2003, Scale development: theory and applications, Applied Social Research Methods Series, vol. 26, Sage, Thousand Oaks, Calif. DFAT 2012, People’s Republic of China country brief, viewed 25 March 2012 <http://www.dfat.gov.au/geo/china/china_brief.html>. 253 Denver, S, Christensen, T, Jensen, JD & Jensen, KOD 2012, 'The Stability and Instability of Organic Expenditures in Denmark, Great Britain, and Italy', Journal of International Food & Agribusiness Marketing, vol. 24, no. 1, pp. 47-65. Dickson, MA, Lennon, SJ, Montalto, CP, Shen, D & ZhangˈLi 2004, ‘Chinese consumer market segments for foreign apparel products’, Journal of Consumer Marketing, vol. 21, no. 5, pp. 301-317. Donovan, R & Henley, N 2003, Social marketing: principles and practice, IP Communications, Melbourne. Dou, W, Wang, G & Zhou, N 2006, ‘Generational and regional differences in media consumption patterns of Chinese generation X consumers’, Journal of Advertising, vol. 35, no. 2, pp. 101-110. Drichoutis, AC, Lazaridis, P & Nayga Jr, RM 2007, ‘An assessment of product class involvement in food-purchasing behavior’, European Journal of Marketing, vol. 41, pp. 888-914. Dubé, L, Cervellon, M-C & Han, J 2003, ‘Should consumer attitudes be reduced to their affective and cognitive bases? Validation of a hierarchical model’, International Journal of Research in Marketing, vol. 20, no. 3, pp. 259-272. Easterby-Smith, M, Thorpe, R & Lowe, A 2002, Management research: an introduction, 2nd edn, Sage, London. Engel, J, Miniard, PW & Blackwell, R 1993, Consumer behavior, Dryden, Fort Worth, Tex. Engel, JF, Kollat, DT & Blackwell, RD 1968, Consumer behavior, Holt, Rinehart and Winston, New York. Ennis, S 2007, ‘Competition and regulation in agriculture: background note’, OECD Journal of Competition Law and Policy, vol. 9, no. 2, pp. 93-165. Essoussi, LH & Zahaf, M 2008, ‘Decision-making process of community organic food consumers: an exploratory study’, Journal of Consumer Marketing, vol. 25, no. 2, pp. 95-104. 254 Euromonitor International 2005, ‘Food, beverages and tobacco forecast’, Food & Drinks Forecast World, June, pp. 1-486. Euromonitor International 2008a, China income & expenditure, viewed 7 October 2009, <http://www.portal.euromonitor.com/passport/DocumentView.aspx>. Euromonitor International 2008b, Consumer lifestyles – China, Country Market Insight, viewed 26 July 2010, <http://www.portal.euromonitor.com/Portal/accessPDF.ashx?c=13\PDF\&f=F-10752114730713.pdf&code=mAXXAFKUOH%2fHUrT62gtynszoG%2fA%3d>. Euromonitor International 2010, Organic packaged food in China, Country Market Insight, viewed 3 February 2010, <http://www.portal.euromonitor.com/Portal/Pages/Search/SearchResultsList.aspx>. Euromonitor International 2011, China: country pulse, Country Market Insight, viewed 30 March 2011, <http://www.portal.euromonitor.com/Portal/accessPDF.ashx?c=97\PDF\&f=F-4878419583197.pdf&code=1VdjTdtOG3UMlwZUR%2f7%2b2Gcj5vA%3d>. Eves, A & Cheng, L 2007, ‘Cross-cultural evaluation of factors driving intention to purchase new food products-Beijing, China and South-East England’, International Journal of Consumer Studies, vol. 31, no. 4, pp. 410-417. Fairbank, JK & Goldman, M 1998, China: a new history, Belknap, Cambridge, Mass. Farndon, J 2008, China rises: how China’s astonishing growth will change the world, Virgin, London. Farrell, D, Gersch, UA & Stephenson, E 2006, ‘The value of China’s emerging middle class’, McKinsey Quarterly, special edition, pp. 61-68. Field, A 2009, Discovering statistics using SPSS, 3rd edn, Sage, London. Finch, JE 2005, ‘The impact of personal consumption values and beliefs on organic food purchase behavior’, Journal of Food Products Marketing, vol. 11, no. 4, pp. 63-76. First, I & Brozin, S 2009, ‘Cultural influences on motives for organic food consumption’, EuroMed Journal of Business, vol. 4, no. 2, pp. 185-199. 255 Fishbein, M (ed.) 1967, Attitude and the prediction of behavior, John Wiley, New York. Fishbein, M & Ajzen, I 1975, Belief, attitude, intention, and behavior: an introduction to theory and research, Addison-Wesley, Reading, Mass. Food and Agriculture Organization of the United Nations 2002, Market developments for organic meat and dairy products: implications for developing countries, Committee on Commodity Problems, Intergovernmental Group on Meat and Dairy Products. Fotopoulos, C & Chryssochoidis, G 2000, ‘Factors affecting the decision to purchase organic food’, Journal of Euromarketing, vol. 9, no. 3, pp. 45-66. Fotopoulos, C & Krystallis, A 2002a, ‘Organic product avoidance: reasons for rejection and potential buyers’ identification in a countrywide survey’, British Food Journal, vol. 104, no. 3-5, pp. 233-260. Fotopoulos, C & Krystallis, A 2002b, ‘Purchasing motives and profile of the Greek organic consumer: a countrywide survey’, British Food Journal, vol. 104, no. 8/9, pp. 730-765. Fotopoulos, C, Krystallis, A & Ness, M 2003, ‘Wine produced by organic grapes in Greece: using means--end chains analysis to reveal organic buyers’ purchasing motives in comparison to the non-buyers’, Food Quality and Preference, vol. 14, no. 7, pp. 549566. Fram, EH, Le, L & Reid, DM 2005, ‘Consumer behavior in China: an exploratory study of two cities’, Journal of Asia-Pacific Business, vol. 5, no. 4, pp. 25-42. Furst, T, Connors, M, Bisogni, CA, Sobal, J & Falk, LW 1996, ‘Food choice: a conceptual model of the process’, Appetite, vol. 26, no. 3, pp. 247-266. Gao, Z, Zhao, K, xiao, X & Tai, c 2009, Organic agriculture and organic food, China Environmental Science Publishing House, Beijing (in Chinese). Gendall, P, Betteridge, K & Bailey, B 1999, ‘The Japanese market for organic fruit and vegetables’, Marketing Bulletin, vol. 10, pp. 24-37. Getz, C & Shreck, A 2006, ‘What organic and fair trade labels do not tell us: towards a place-based understanding of certification’, International Journal of Consumer Studies, vol. 30, no. 5, pp. 490-501. 256 Giannakas, K 2002, ‘Information asymmetries and consumption decisions in organic food product markets’, Canadian Journal of Agricultural Economics, vol. 50, no. 1, pp. 3550. Gil, JM, Gracia, A & Sanchez, M 2000, ‘Market segmentation and willingness to pay for organic products in Spain’, International Food and Agribusiness Management Review, vol. 3, no. 2, pp. 207-226. González, JAA 2009, ‘Market trends and consumer profile at the organic farmers market in Costa Rica’, British Food Journal, vol. 111, no. 5, pp. 498-510. Google Image 2010, China map, viewed 8 March 2010, <http://www.google.com.au/search?q=China+map&ie=utf-8&oe=utf8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a>. Gorsuch, RL 1983, Factor analysis, 2nd edn, Lawrence Erlbaum Associates, Hillsdale, NJ. Gorsuch, RL 1997, ‘Exploratory factor analysis: its role in item analysis’, Journal of Personality Assessment, vol. 68, no. 3, pp. 532-560. Gotschi, E, Vogel, S, Lindenthal, T & Larcher, M 2010, ‘The role of knowledge, social norms, and attitudes toward organic products and shopping behavior: survey results from high school students in Vienna’, Journal of Environmental Education, vol. 41, pp. 88-100. Govindasamy, R, DeCongelio, M & Bhuyan, S 2005, ‘An evaluation of consumer willingness to pay for organic produce in the Northeastern US’, Journal of Food Products Marketing, vol. 11, no. 4, pp. 3-20. Grankvist, G, Lekedal, H & Marmendal, M 2007, ‘Values and eco- and fair-trade labelled products’, British Food Journal, vol. 109, no. 2, pp. 169-181. Grannis, J, Hine, S & Thilmany, D 2001, ‘Marketing premium food products in emerging economies: the case of Macedonian cheese’, Journal of International Food & Agribusiness Marketing, vol. 13, no. 2/3, pp. 59-76. 257 Grønhøj, A 2006, ‘Communication about consumption: a family process perspective on “green” consumer practices’, Journal of Consumer Behaviour, vol. 5, no. 6, pp. 491503. Grunert, SC & Juhl, HJ 1995, ‘Values, environmental attitudes, and buying of organic foods’, Journal of Economic Psychology, vol. 16, no. 1, pp. 39-62. Guido, G, Prete, M, Peluso, A, Maloumby-Baka, R & Buffa, C 2010, ‘The role of ethics and product personality in the intention to purchase organic food products: a structural equation modeling approach’, International Review of Economics, vol. 57, no. 1, pp. 79-102. Haghiri, M & McNamara, ML 2007, ‘Predicting consumers’ acceptability of pesticide-free fresh produce in Canada’s maritime provinces: a probit analysis’, Journal of International Food & Agribusiness Marketing, vol. 19, pp. 45-59. Hair, JFJ, Black, WC, Babin, BJ, Anderson, RE & Tatham, RL 2006, Multivariate data analysis, 6th edn, Pearson Prentice Hall, Upper Saddle River, NJ. Halpin, D 2004, The Australian organic industry: a profile, Department of Agriculture, Fisheries and Forestry, Canberra. <http://www.daff.gov.au/agriculturefood/food/organic-biodynamic/industry>. Hamzaoui-Essoussi, L & Zahaf, M 2012, 'Canadian organic food consumers' profile and their willingness to pay premium prices', Journal of International Food & Agribusiness Marketing, vol. 24, no. 1, pp. 1-21. Hansen, K 2001, ‘Purchasing decision behaviour by Chinese supermarkets’, International Review of Retail, Distribution & Consumer Research, vol. 11, no. 2, pp. 159-175. Hansen, T 2005, ‘Understanding consumer perception of food quality: the cases of shrimps and cheese’, British Food Journal, vol. 107, no. 7, pp. 500-525. Harper, GC & Makatouni, A 2002, ‘Consumer perception of organic food production and farm animal welfare’, British Food Journal, vol. 104, no. 3, pp. 287-299. 258 Heckman, J 2005, ‘A history of organic farming: transitions from Sir Albert Howard’s war in the soil to USDA national organic program’, Renewable Agriculture and Food Systems, vol. 21, no. 3, pp. 143-150. Herche, J 1992, ‘A note on the predictive validity of the CETSCALE’, Journal of the Academy of Marketing Science, vol. 20, no. 3, pp. 261-264. Hill, H & Lynchehaun, F 2002, ‘Organic milk: attitidues and consumption patterns’, British Food Journal, vol. 104, no. 7, pp. 526-542. Ho, P, Vermeer, EB & Zhao, JH 2006, ‘Biotechnology and food safety in China: consumers’ acceptance or resistance?’, Development & Change, vol. 37, no. 1, pp. 227-254. Ho, S-c & Tang, F 2006, ‘An exploratory investigation into the food-shopping attitudes and behavior of Chinese shoppers’, Journal of International Consumer Marketing, vol. 19, no. 2, pp. 53-74. Hofstede, G 1994, ‘Management scientists are human’, Management Science, vol. 40, no. 1, pp. 4-13. Honkanen, P, Verplanken, B & Olsen, SO 2006, ‘Ethical values and motives driving organic food choice’, Journal of Consumer Behaviour, vol. 5, no. 5, pp. 420-430. Howard, A 1943, An agricultural testament, Oxford University Press, London. Howard, PH & Allen, P 2006, ‘Beyond organic: consumer interest in new labelling schemes in the Central Coast of California’, International Journal of Consumer Studies, vol. 30, no. 5, pp. 439-451. Hsu, JL & Nien, H-P 2008, ‘Who are ethnocentric? Examining consumer ethnocentrism in Chinese societies’, Journal of Consumer Behaviour, vol. 7, pp. 436-447. Hu, L-t & Bentler, PM 1998, ‘Fit indices in covariance structure modeling: sensitivity to underparameterized model misspecification’, Psychological Methods, vol. 3, no. 4, pp. 424-453. 259 Hu, W, Chen, K & Yoshida, K 2006, ‘Japanese consumers’ perceptions on and willingness to pay for credence attributes associated with canola oil’, Journal of Agricultural and Applied Economics, vol. 38, no. 1, pp. 91-103. Huang, CL 1996, ‘Consumer preferences and attitudes towards organically grown produce’, European Review of Agricultural Economics, vol. 23, no. 3, pp. 331-342. Hughner, RS, McDonagh, P, Prothero, A, Shultz, CJ, II & Stanton, J 2007, ‘Who are organic food consumers? A compilation and review of why people purchase organic food’, Journal of Consumer Behaviour, vol. 6, no. 2, pp. 94-110. Huliyeti, H, Marchesini, S & Canavari, M 2008, ‘Chinese distribution practitioners’ attitudes towards Italian quality foods’, Journal of Chinese Economic and Foreign Trade Studies, vol. 1, no. 3, pp. 214-231. Hutchins, RK & Greenhalgh, LA 1995, ‘Organic confusion: sustaining competitive advantage’, Nutrition and Food Science, vol. 95, no. 6, pp. 11-14. IFOAM 2006, International Federation of Organic Agriculture Movements annual reports 2005 and 2006, viewed 30 March 2008, <http://www.ifoam.org/about_ifoam/inside_ifoam/annual_reports.html>. IFOAM 2012, The history of IFOAM, viewed 20 March 2012, <http://www.ifoam.org/about_ifoam/inside_ifoam/history.html>. IFOAM 2009, The IFOAM organic guarantee system, viewed 2 February 2009, <http://www.ifoam.org/about_ifoam/standards/ogs.html>. IFOAM 2011a, About the International Federation of Organic Agriculture Movements (IFOAM), viewed 28 May 2011, <http://www.ifoam.org/about_ifoam/index.html>. IFOAM 2011b, Growth in organic continues, viewed 14 April 2011, <http://www.ifoam.org/IFOAMBioFachMediatext-e.pdf>. Institute of Food Science and Technology 2009, Information statement organic food, viewed 13 March 2009, <http://www.ifst.org/site/cms/contentcategoryview.asp?category=126>. 260 International Trade Centre 2011, Organic food products in China: market overview, International Trade Centre, Geneva, viewed 24 April 2011, <http://www.intracen.org/uploadedFiles/intracenorg/Content/Publications/Organicfood-products-in-China-market-overview.pdf>. Jain, S 2006, ‘Role of socio-demographics in segmenting and profiling green consumers’, Journal of International Consumer Marketing, vol. 18, no. 3, pp. 107-146. Jain, SK & Kaur, G 2004, ‘Green marketing: an attitudinal and behavioural analysis of Indian consumers’, Global Business Review, vol. 5, no. 2, pp. 187-205. Janssen, M, Heid, A & Hamm, U 2009, ‘Is there a promising market “in between” organic and conventional food? Analysis of consumer preferences’, Renewable Agriculture and Food Systems, vol. 24, no. 3, pp. 205-213. Jones-Farmer, LA, Pitts, JP & Rainer, RK 2008, ‘A note on multigroup comparisons using SAS PROC CALIS’, Structural Equation Modeling: A Multidisciplinary Journal, vol. 15, no. 1, pp. 154-173. Kahal, SE 2001, Business in Asia Pacific: text and cases, Oxford University Press, Oxford. Kalafatis, SP, Pollard, M, East, R & Tsogas, MH 1999, ‘Green marketing and Ajzen’s Theory of Planned Behaviour: a cross-market examination’, Journal of Consumer Marketing, vol. 16, no. 5, pp. 441-460. Kaplan, D 2009, Structural equation modeling: foundations and extensions, 2nd edn, Sage, Thousand Oaks, Calif. Keane, M 2007, Created in China: the Great New Leap Forward, Routledge, Abingdon. Kiesel, K & Villas-Boas, SB 2007, ‘Got organic milk? Consumer valuations of milk labels after the implementation of the USDA organic seal’, Journal of Agricultural & Food Industrial Organization, vol. 5, pp. 1-40. Kim, J-H & Lennon, SJ 2010, ‘Information available on a web site: effects on consumers’ shopping outcomes’, Journal of Fashion Marketing and Management, vol. 14, no. 2, pp. 247-262. 261 Kim, WG, Ma, X & Kim, DJ 2006, ‘Determinants of Chinese hotel customers’ esatisfaction and purchase intention’, Tourism Management, vol. 27, no. 5, pp. 890-900. Klein, JG, Ettenson, R & Morris, MD 1998, ‘The animosity model of foreign product purchase: an empirical test in the People’s Republic of China’, Journal of Marketing, vol. 62, no. 1, pp. 89-100. Kline, P 1994, An easy guide to factor analysis, Routledge, London. Kline, RB 2005, Principles and practice of structural equation modeling, 2nd edn, Guilford, New York. Kline, RB 2010, Principles and practice of structural equation modeling, 3rd edn, Guilford, New York. Klonsky, K & Greene, C 2005, ‘Widespread adoption of organic agriculture in the US: are market driven policies enough?’, American Agricultural Economics Association Annual Meeting, Providence, RI, 23-27 July. Kluger, J 2010, ‘What’s so great about organic food?’, Time, vol. 196, no. 10. Kortbech-Olesen, R 1998, ‘Exporting organic foods’, International Trade Forum, no. 3, pp. 4-9. Kotabe, M, peloso, A, Gregory, G, Noble, G, Macarthur, W, Neal, C, Riege, A & Helsen, K 2005, International marketing: an Asia Pacific focus, John Wiley, Brisbane. Kotler, P, Brown, L, Adam, S & Armstrong, G 2001, Marketing, 5th edn, Pearson Education, Sydney. Kraft, FB & Goodell, PW 1993, ‘Identifying the health conscious consumer’, Journal of Health Care Marketing, vol. 13, no. 3, pp. 18-25. Kristiansen, P 2006, ‘Overview of organic agriculture’ in P Kristiansen, A Taji and J Reganold (eds) Organic agriculture: a global perspective, CSIRO Publishing, Melbourne. Kristiansen, P, Mitchell, A, Bez, N & Monk, A 2010, Australian organic market report 2010, Biological Farmers of Australia, Brisbane. 262 Kristiansen, P & Smithson, A 2008, Australian organic market report 2008, Organic Research Centre, University of New England, Armidale, viewed 26 July 2009, <http://www.bfa.com.au/index.asp?Sec_ID=259>. Krystallis, A, Arvanitoyannis, I & Chryssohoidis, G 2006, ‘Is there a real difference between conventional and organic meat? Investigating consumers’ attitudes towards both meat types as an indicator of organic meat’s market potential’, Journal of Food Products Marketing, vol. 12, no. 2, pp. 47-78. Krystallis, A & Chryssohoidis, G 2005, ‘Consumers’ willingness to pay for organic food: factors that affect it and variation per organic product type’, British Food Journal, vol. 107, no. 5, pp. 320-343. Krystallis, A, Fotopoulos, C & Zotos, Y 2006, ‘Organic consumers’ profile and their willingness to pay (WTP) for selected organic food products in Greece’, Journal of International Consumer Marketing, vol. 19, no. 1, pp. 81-106. Krystallis, A, Petrovici, D & Arvanitoyannis, I 2006, ‘From commodities to the consumption of quality foods in Eastern European context: an empirical examination of the determinants of consumer behaviour towards honey’, Journal of East-West Business, vol. 12, no. 4, pp. 5-37. Kuhar, A & Juvancic, L 2010, ‘Determinants of purchasing behaviour for organic and integrated fruits and vegetables in Slovenia’, Agricultural Economics Review, vol. 11, no. 2, pp. 70-83. Kwok, S, Uncles, M & Huang, Y 2006, ‘Brand preferences and brand choices among urban Chinese consumers: an investigation of country-of-origin effects’, Asia Pacific Journal of Marketing and Logistics, vol. 18, no. 3, pp. 163-172. La Ferle, C & Lee, W-N 2003, ‘Attitudes toward advertising: a comparative study of consumers in China, Taiwan, South Korea and the United States’, Journal of International Consumer Marketing, vol. 15, no. 2, pp. 5-23. Larue, B, West, GE, Gendron, C & Lambert, R 2004, ‘Consumer response to functional foods produced by conventional, organic, or genetic manipulation’, Agribusiness, vol. 20, no. 2, pp. 155-166. 263 Lavidge, RJ & Steiner, GA 1961, ‘A model for predictive measurements of advertising effectiveness’, Journal of Marketing, vol. 25, no. 6, pp. 59-62. Lea, E & Worsley, T 2005, ‘Australians’ organic food beliefs, demographics and values’, British Food Journal, vol. 107, no. 11, pp. 855-869. Lee, JW, Jones, PS, Mineyama, Y & Zhang, XE 2002, ‘Cultural differences in responses to a Likert scale’, Research in Nursing & Health, vol. 25, no. 4, pp. 295-306. Leedy, PD & Ormrod, JE 2005, Practical research: planning and design, 8th edn, Merrill Prentice Hall, Upper Saddle River, NJ. Lennernas, M, Fjellstrom, C, Becker, W, Giachetti, I et al. 1997, ‘Influences on food choice perceived to be important by nationally-representative samples of adults in the European Union’, European Journal of Clinical Nutrition, vol. 51, pp. S8-15. Leu, A 2009, The benefits of organic food, viewed 2 August 2009, <http://www.ofa.org.au/papers//The%20Benefits%20of%20Organic%20Food%20pape r.htm>. Li, J, Zepeda, L & Gould, BW 2007, ‘The demand for organic food in the US: an empirical assessment’, Journal of Food Distribution Research, vol. 38, no. 3, pp. 54-69. Li, Y, Cheng, L & Ren, G 2005, ‘Product differences and purchasing decision-making on non-environmental pollution agricultural products: an analysis of consumer survey in Jiaxin’ (in Chinese), Zhejiang Statistics, vol. 2. Lichtenstein, DR, Ridgway, NM & Netemeyer, RG 1993, ‘Price perceptions and consumer shopping behavior: a field study’, Journal of Marketing Research, vol. 30, pp. 234-245. Lin, B-H, Smith, TA & Huang, CL 2008, ‘Organic premiums of US fresh produce’, Renewable Agriculture and Food Systems, vol. 23, no. 3, pp. 208-216. Lindeman, M & Vaananen, M 2000, ‘Measurement of ethical food choice motives’, Appetite, vol. 34, no. 1, pp. 55-59. Liu, F, Murphy, J, Li, J & Liu, X 2006, ‘English and Chinese? The role of consumer ethnocentrism and country of origin in Chinese attitudes towards store signs’, Australasian Marketing Journal, vol. 14, no. 2, pp. 5-16. 264 Lockie, S, Halpin, D & Pearson, D 2006, ‘Understanding the market for organic food’ in P Kristiansen, A Taji and J Reganold (eds) Organic agriculture: a global perspective, CSIRO Publishing, Melbourne. Lockie, S, Lyons, K, Lawrence, G & Grice, J 2004, ‘Choosing organics: a path analysis of factors underlying the selection of organic food among Australian consumers’, Appetite, vol. 43, no. 2, pp. 135-146. Lockie, S, Lyons, K, Lawrence, G & Mummery, K 2002, ‘Eating “green”: motivations behind organic food consumption in Australia’, Sociologia Ruralis, vol. 42, no. 1, pp. 23-40. Lodorfos, GN & Dennis, J 2008, ‘Consumers’ intent: in the organic food market’, Journal of Food Products Marketing, vol. 14, no. 2, pp. 17-38. Loureiro, M, L. & Hine, S 2002, ‘Discovering niche markets: a comparison of consumer willingness to pay for local (Colorado grown), organic, and GMO-free products’, Journal of Agricultural and Applied Economics, vol. 34, no. 3, pp. 477-487. Lu, PX 2008, Elite China: luxury consumer behavior in China, John Wiley, Singapore. Luo, S 2007, ‘An analysis of organic food demand and slow development in China’ (in Chinese), Rural Economy and Science-Technology, vol. 18, no. 7. Magkos, F, Arvaniti, F & Zampelas, A 2003, ‘Putting the safety of organic food into perspective’, Nutrition Research Reviews, vol. 16, pp. 211-221. Magkos, F, Arvaniti, F & Zampelas, A 2006, ‘Organic food: buying more safety or just peace of mind? A critical review of the literature’, Critical Reviews in Food Science and Nutrition, vol. 46, no. 1, pp. 23-56. Magnusson, M, K., Arvola, A, Koivisto Hursti, U-K, Aberg, L & Sjoden, P-O 2001, ‘Attitudes towards organic foods among Swedish consumers’, British Food Journal, vol. 103, no. 3, pp. 209-227. Magnusson, MK, Arvola, A, Hursti, U-KK, Åberg, L & Sjödén, P-O 2003, ‘Choice of organic foods is related to perceived consequences for human health and to environmentally friendly behaviour’, Appetite, vol. 40, no. 2, pp. 109-117. 265 Makatouni, A 2002, ‘What motivates consumers to buy organic food in the UK? Results from a qualitative study’, British Food Journal, vol. 104, no. 3-5, pp. 345-352. Malhotra, N 2008, Essentials of marketing research: an applied orientation, 2nd edn, Pearson Education, Sydney. Malhotra, N, Hall, J, Shaw, M & Oppenheim, P 2006, Marketing research: an applied orientation, 3rd edn, Pearson Education, Sydney. Margetts, BM, Martinez, JA, Saba, A, Holm, L & Kearney, M 1997, ‘Definitions of “healthy” eating: a pan-EU survey of consumer attitudes to food, nutrition and health’, European Journal of Clinical Nutrition, vol. 51, pp. S23-29. Mather, D, Knight, J & Holdsworth, D 2005, ‘Pricing differentials for organic, ordinary and genetically modified food’, Journal of Product & Brand Management, vol. 14, no. 6, pp. 387-392. Mathews, SW 2009, Internationalisation: the internet’s influence on international market growth in the firm’s outward internationalisation process, PhD thesis, School of Advertising, Marketing & Public Relations, Queensland University of Technology, Brisbane. McDonald, P 2001, The organic food market in Europe: an overview of Europe’s fastest growing niche food market, Queensland Government Trade & Investment Office Europe. McEachern, MG & McClean, P 2002, ‘Organic purchasing motivations and attitudes: are they ethical?’, International Journal of Consumer Studies, vol. 26, no. 2, pp. 85-92. McEachern, MG & Willock, J 2004, ‘Producers and consumers of organic meat: a focus on attitudes and motivations’, British Food Journal, vol. 106, pp. 534-552. McEwen, W, Fang, X, Zhang, C & Burkholder, R 2006, ‘Inside the mind of the Chinese consumer’, Harvard Business Review, vol. 84, no. 3, pp. 68-76. McGregor, J 2005, One billion customers: lessons from the front lines of doing business in China, Nicholas Brealey, London. 266 McGuire, W 1978, ‘An information processing model of advertising effectiveness’ in HL Davis and AJ Silk (eds) Behavioral and management science in marketing, Ronald Press, New York, pp. 156-180. McMurray, AJ, Pace, RW & Scott, D 2004, Research: a commonsense approach, Thomson Social Science Press, Melbourne. Millock, K, Wier, M & Andersen, LM 2004, ‘Consumer demand for organic foods: attitudes, values and purchasing behavior’, paper presented at Thirteenth Annual Conference of European Association of Environmental and Resource Economics, Budapest, 25-28 June. Mitsostergios, KT & Skiadas, CH 1994, ‘Attitudes and perceptions of fresh pasteurised milk consumer: a qualitative and quantitative survey’, British Food Journal, vol. 96, no. 7, pp. 4-10. Mooij, Md & Hofstede, G 2002, ‘Convergence and divergence in consumer behavior: implications for international retailing’, Journal of Retailing, vol. 78, no. 1, pp. 61-69. Moon, W, Florkowski, W, Resurreccion, AV, Beuchat, LR, Jordanov, J, Chinnan, M & Paraskova, P 1998, ‘Consumer concerns about nutritional attributes in a transition economy’, Food Policy, vol. 23, no. 5, pp. 357-369. Moore, O 2006, ‘Understanding postorganic fresh fruit and vegetable consumers at participatory farmers’ markets in Ireland: reflexivity, trust and social movements’, International Journal of Consumer Studies, vol. 30, no. 5, pp. 416-426. Mullen, MR 1995, ‘Diagnosing measurement equivalence in cross-national research’, Journal of International Business Studies, vol. 26, pp. 573-596. National Bureau of Statistics of China 2009, China Statistical Yearbook 2009, viewed 1 August 2010, <http://www.stats.gov.cn/english/whatsnew/t20090925_402590214.htm>. National Bureau of Statistics of China 2011, Statistical communiqué of the People’s Republic of China on the 2010 national economic and social development, viewed 30 267 March 2011, <http://www.stats.gov.cn/english/newsandcomingevents/t20110228_402705764.htm>. Nees, WP 2011, Chinese consumers and organic food, Master’s thesis, Department of Marketing and Statistics, Aarhus University, Denmark. Nunnally, JC 1967, Psychometric theory, McGraw-Hill, New York. Nunnally, JC & Bernstein, IH 1994, Psychometric theory, 3rd edn, McGraw-Hill, New York. O’Donovan, P & McCarthy, M 2002, ‘Irish consumer preference for organic meat’, British Food Journal, vol. 104, no. 34/55, pp. 353-370. Okun, J 2007, ‘A growing movement toward green food’, Buffalo News, 18 July. Onozaka, Y 2007, Three essays on consumers’ preferences for fresh organic produce, PhD thesis, Agricultural and Resource Economics, University of California, Davis. Onyango, BM, Hallman, WK & Bellows, AC 2007, ‘Purchasing organic food in US food systems: a study of attitudes and practice’, British Food Journal, vol. 109, no. 5, pp. 399-411. Oraman, Y & Unakitan, G 2010, ‘Analysis of factors influencing organic fruit and vegetable purchasing in Istanbul, Turkey’, Ecology of Food and Nutrition, vol. 49, no. 6, pp. 452-466. Organic Federation of Australia 2005, National standard for organic and bio-dynamic produce, viewed 2 August 2009, <http://www.ofa.org.au/pages/BecomingCertified.html>. Organic Food Development Center 2009, OFDC introduction, viewed 30 March 2009, <http://www.ofdc.org.cn/about/about.asp>. Özcelik, AÖ & Uçar, A 2008, ‘Turkish academic staffs’ perception of organic foods’, British Food Journal, vol. 110, no. 9, pp. 948-960. Ozretic-Dosen, D, Skare, V & Krupka, Z 2007, ‘Assessments of country of origin and brand cues in evaluating a Croatian, western and eastern European food product’, Journal of Business Research, vol. 60, no. 2, pp. 130-136. 268 Padel, S & Foster, C 2005, ‘Exploring the gap between attitudes and behaviour: understanding why consumers buy or do not buy organic food’, British Food Journal, vol. 107, no. 8, pp. 606-625. Pallant, J 2007, SPSS survival manual: a step by step guide to data analysis using SPSS for Windows (Version 15), 3rd edn, Allen & Unwin, Sydney. Paull, J 2008a, The greening of China’s food-green food, organic food, and eco-labelling, Sustainable Consumption and Alternative Agri-Food Systems Conference (SUSCONS), Liege University, Belgium Paull, J 2008b, ‘The lost history of organic farming in Australia’, Journal of Organic Systems, vol. 3, no. 2, pp. 2-17. Peng, H 2006, Consumer behaviour research in food safety management: an empirical study from Beijing (in Chinese), PhD thesis, Department of Economy and Management, China Agricultural University, Beijing. Perry, C 1995, A structured approach to presenting PhD theses: notes for candidates and their supervisors, paper presented to ANZ Doctoral Consortium, University of Sydney. Peterson, RA 2000, ‘A meta-analysis of variance accounted for and factor loadings in exploratory factor analysis’, Marketing Letters, vol. 11, no. 3, pp. 261-275. Petrovici, DA, Ritson, C & Ness, M 2004, ‘The theory of reasoned action and food choice: insights from a transitional economy’, Journal of International Food & Agribusiness Marketing, vol. 16, no. 1, pp. 59-87. Phau, I & Min, T 2009, ‘Devil wears (counterfeit) Prada: a study of antecedents and outcomes of attitudes towards counterfeits of luxury brands’, Journal of Consumer Marketing, vol. 26, pp. 15-27. Pilgrim, FJ 1957, ‘The components of food acceptance and their measurement’, American Journal of Clinical Nutrition, vol. 5, pp. 171-175. Piron, F 2006, ‘China’s changing culture: rural and urban consumers’ favorite things’, Journal of Consumer Marketing, vol. 23, pp. 327-334. 269 Prasso, S, Ting, W & Dan, Z 2007, ‘China’s new cultural revolution’, Fortune, vol. 155, no. 10, p. 28. Qian, W, Razzaque, MA & Keng, KA 2007, ‘Chinese cultural values and gift-giving behavior’, Journal of Consumer Marketing, vol. 24, no. 4, pp. 214-228. Qing, P, Lobo, A & Li, C 2012, 'The impact of lifestyle and ethnocentrism on consumers' purchase intentions of fresh fruit in China', Journal of Consumer Marketing, vol. 29, no. 1, pp. 43-51. Radman, M 2005, ‘Consumer consumption and perception of organic products in Croatia’, British Food Journal, vol. 107, no. 4, pp. 263-273. Reid, M, Li, E, Bruwer, J & Grunert, K 2001, ‘Food-related lifestyles in a cross-cultural context’, Journal of Food Products Marketing, vol. 7, no. 4, pp. 57-75. Rimal, A, Moon, W & Balasubramanian, SK 2006, ‘Perceived risks of agro-biotechnology and organic food purchases in the United States’, Journal of Food Distribution Research, vol. 37, no. 2, pp. 70-79. Rimal, A, P., Moon, W & Balasubramanian, S 2005, ‘Agro-biotechnology and organic food purchase in the United Kingdom’, British Food Journal, vol. 107, no. 2, pp. 84-97. Roddy, G, Cowan, C & Hutchinson, G 1994, ‘Organic food: a description of the Irish market’, British Food Journal, vol. 96, no. 4, pp. 3-10. Roddy, G, Cowan, CA & Hutchinson, G 1996, ‘Consumer attitudes and behavior to organic foods in Ireland’, Journal of International Consumer Marketing, vol. 9, no. 2, pp. 4163. Roitner-Schobesberger, B, Darnhofer, I, Somsook, S & Vogl, CR 2008, ‘Consumer perceptions of organic foods in Bangkok, Thailand’, Food Policy, vol. 33, no. 2, pp. 112-121. Saba, A & Messina, F 2003, ‘Attitudes towards organic foods and risk/benefit perception associated with pesticides’, Food Quality and Preference, vol. 14, no. 8, pp. 637-645. 270 Salleh, M, Ali, S, Harun, E, Jalil, M & Shaharudin, M 2010, ‘Consumers’ perception and purchase intention towards organic food products: exploring attitude among academician’, Canadian Social Science, vol. 6, no. 6, pp. 119-129. Sanders, R 2006, ‘A market road to sustainable agriculture? Ecological agriculture, green food and organic agriculture in China’, Development & Change, vol. 37, no. 1, pp. 201-226. Sanjuán, AI, Sánchez, M, Gil, JM, Gracia, A & Soler, F 2003, ‘Brakes to organic market enlargement in Spain: consumers’ and retailers’ attitudes and willingness to pay’, International Journal of Consumer Studies vol. 27, no. 2, pp. 134-144. Saunders, M, Lewis, P & Thornhill, A 2007, Research methods for business students, 4th edn, Pearson Prentice Hall Financial Times, Harlow, UK. Schifferstein, HNJ & Oude Ophuis, PAM 1998, ‘Health-related determinants of organic food consumption in the Netherlands’, Food Quality and Preference, vol. 9, no. 3, pp. 119-133. Schmitt, N & Kuljanin, G 2008, ‘Measurement invariance: review of practice and implications’, Human Resource Management Review, vol. 18, no. 4, pp. 210-222. Scholten, M 1996, ‘Lost and found: the information-processing model of advertising effectiveness’, Journal of Business Research, vol. 37, no. 2, pp. 97-104. Schütte, H & Ciarlante, D 1998, Consumer behavior in Asia, New York University Press, New York. Sedgman, P 2011, China imports of New Zealand milk jump five-fold, aiding quake-hit nation, Bloomberg, 9 March, viewed 20 September 2011, <http://www.bloomberg.com/news/2011-03-08/china-imports-of-new-zealand-milkjump-five-fold-aiding-quake-hit-nation.html>. Shaharudin, M, Pani, J, Mansor, S & Elias, S 2010a, ‘Factors affecting purchase intention of organic food in Malaysia’s Kedah State’, Cross-Cultural Communication, vol. 6, no. 2, pp. 105-116. 271 Shaharudin, M, Pani, J, Mansor, S, Elias, S & Sadek, D 2010b, ‘Purchase intention of organic food in Kedah, Malaysia; a religious overview’, International Journal of Marketing Studies vol. 2, no. 1, pp. 96-103. Shaharudin, MR, Pani, JJ, Mansor, SW & Elias, SJ 2010c, ‘Purchase intention of organic food: perceived value overview’, Canadian Social Science, vol. 6, no. 1, pp. 70-79. Shamdasani, P, Chon-Lin, GO & Richmond, D 1993, ‘Exploring green consumers in an oriental culture: role of personal and marketing mix factors’, Advances in Consumer Research, vol. 20, no. 1, pp. 488-493. Shaw, D & Clarke, I 1999, ‘Belief formation in ethical consumer groups: an exploratory study’, Marketing Intelligence & Planning, vol. 17, no. 2, pp. 109-120. Shenzhen government online 2009, Comprehensive economic strength, viewed 8 December 2009, <http://english.sz.gov.cn/lis/>. Shepherd, R, Magnusson, M & Sjödén, P-O 2005, ‘Determinants of consumer behaviour related to organic foods’, Ambio, vol. 34, no. 4/5, pp. 352-359. Shimp, TA & Sharma, S 1987, ‘Consumer ethnocentrism: construction and validation of the CETSCALE’, Journal of Marketing Research, vol. 24, pp. 280-289. Shumei, C 2009, ‘Sham or shame: rethinking the China’s milk powder scandal from a legal perspective’, Journal of Risk Research, vol. 12, no. 6, pp. 725-747. Slavicek, LC 2010, The Chinese Cultural Revolution, Chelsea House, New York. Smith, L 2007, ‘Think China, act local’, Australia China Connections, vol. 7, no. 4, p. 26. Smith, S & Paladino, A 2010, ‘Eating clean and green? Investigating consumer motivations towards the purchase of organic food’, Australasian Marketing Journal, vol. 18, no. 2, pp. 93-104. Smith, T, Huang, C & Lin, B 2009a, ‘Does price or income affect organic choice? Analysis of US fresh produce users’, Journal of Agricultural and Applied Economics, vol. 41, no. 3, pp. 731-744. Smith, T, Huang, C & Lin, B 2009b, ‘Estimating organic premiums in the US fluid milk market’, Renewable Agriculture and Food Systems, vol. 24, pp. 197-204. 272 Soler, F, Gil, JM & Sánchez, M 2002, ‘Consumers’ acceptability of organic food in Spain’, British Food Journal, vol. 104, no. 8/9, pp. 670-687. Sottoni, VA, Contini, C, Martini, A & Menghini, S 2002, ‘Possible development of organic production in a mountain area of the Florence province in Tuscany’, British Food Journal, vol. 104, no. 8/9, pp. 654-669. Sparks, P, Hedderley, D & Shepherd, R 1992, ‘An investigation into the relationship between perceived control, attitude variability and the consumption of two common foods’, European Journal of Social Psychology, vol. 22, pp. 55-71. Sparks, P & Shepherd, R 1992, ‘Self-identity and the theory of planned behavior: assessing the role of identification with “green consumerism”’, Social Psychology Quarterly, vol. 55, no. 4, pp. 388-399. Squires, L, Juric, B & Cornwell, TB 2001, ‘Level of market development and intensity of organic food consumption: cross-cultural study of Danish and New Zealand consumers’, Journal of Consumer Marketing, vol. 18, no. 4/5, pp. 392-409. St-Maurice, I, Süssmuth-Dyckerhoff, C & Tsai, H 2008, ‘What’s new with the Chinese consumer’, McKinsey Quarterly, Sept. Steenkamp, J-BEM 1997, ‘Dynamics in consumer behavior with respect to agricultural and food products’ in B Wierenga, Av Tilburg, K Grunert and J-BEM Steenkamp (eds) Agricultural marketing and consumer behavior in a changing world, Kluwer Academic Publishers, Boston, pp. 143-188. Steenkamp, J-BEM & van Trijp, HCM 1991, ‘The use of LISREL in validating marketing constructs’, International Journal of Research in Marketing, vol. 8, no. 4, pp. 283-299. Steptoe, A, Pollard, TM & Wardle, J 1995, ‘Development of a measure of the motives underlying the selection of food: the food choice questionnaire’, Appetite, vol. 25, no. 3, pp. 267-284. Stewart, B & Bean, R 2006, China, Peoples Republic of market development reports Chengdu: an emerging market in China’s heartland, GAIN Report, USDA Foreign 273 Agricultural Service, viewed 10 August 2010, <http://www.fas.usda.gov/gainfiles/200601/146176543.pdf>. Stobbelaar, DJ, Casimir, G, Borghuis, J, Marks, I, Meijer, L & Zebeda, S 2007, ‘Adolescents’ attitudes towards organic food: a survey of 15- to 16-year old school children’, International Journal of Consumer Studies, vol. 31, no. 4, pp. 349-356. Storstad, O & Bjørkhaug, H 2003, ‘Foundations of production and consumption of organic food in Norway: common attitudes among farmers and consumers?’, Agriculture and Human Values, vol. 20, pp. 151-163. Straub, D, Boudreau, M-C & Gefen, D 2004, ‘Validation guidelines for IS positivist research’, Communications of the Association for Information Systems, vol. 13, no. 1, pp. 380-427. Sun, X 2003, The dynamics of Chinese consumer behaviour in relation to the purchase of imported fruit, PhD thesis, School of Natural and Rural System Management, University of Queensland, Brisbane. Sun, X & Collins, R 2002, ‘Attitudes and consumption values of consumers of imported fruit in Guangzhou, China’, International Journal of Consumer Studies, vol. 26, no. 1, pp. 34-43. Sun, X & Collins, R 2006, ‘Chinese consumer response to imported fruit: Intended uses and their effect on perceived quality’, International Journal of Consumer Studies, vol. 30, no. 2, pp. 179-188. Tabachnick, BG & Fidell, LS 2007, Using multivariate statistics, 5th edn, Pearson Education, Boston Tai, SHC & Tam, JLM 1997, ‘A lifestyle analysis of female consumers in greater China’, Psychology and Marketing, vol. 14, no. 3, pp. 287-307. Tanner, C & Kast, SW 2003, ‘Promoting sustainable consumption: determinants of green purchases by Swiss consumers’, Psychology & Marketing, vol. 20, no. 10, pp. 883-902. 274 Tarkiainen, A & Sundqvist, S 2005, ‘Subjective norms, attitudes and intentions of Finnish consumers in buying organic food’, British Food Journal, vol. 107, no. 10/11, pp. 808822. Taylor, DA 2008, ‘Recovering the good earth: China’s growing organic market’, Environmental Health Perspectives, vol. 116, no. 8, pp.346-349. Thøgersen, J 2009, ‘Consumer decision-making with regard to organic food products’ in TdeN Vaz, P Nijkamp and J-L Rastoin (eds) Traditional food production and rural sustainable development: a European challenge, Ashgate, Farnham, UK. Thøgersen, J & Zhou, Y 2012, 'Chinese consumers’ adoption of a ‘green’ innovation – The case of organic food', Journal of Marketing Management, vol. 28, no. 3/4, pp. 313-333. Thompson, B 1997, ‘The importance of structure coefficients in structural equation modeling confirmatory factor analysis’, Educational and Psychological Measurement, vol. 57, no.1, pp. 5-19. Thompson, GD 1998, ‘Consumer demand for organic foods: what we know and what we need to know’, American Journal of Agricultural Economics, vol. 80, no. 5, pp. 11131118. Thompson, GD & Kidwell, J 1998, ‘Explaining the choice of organic produce: cosmetic defects, prices and consumer preferences’, American Journal of Agricultural Economics, vol. 80, no. 2, pp. 277-287. Tian, K & Dong, L 2011, Consumer-citizens of China: the role of foreign brands in the imagined future, Routledge, London. Todman, JB & Dugard, P 2007, Approaching multivariate analysis: an introduction for psychology, Psychology Press, Hove, UK. Torjusen, H, Lieblein, G, Wandel, M & Francis, CA 2001, ‘Food system orientation and quality perception among consumers and producers of organic food in Hedmark County, Norway’, Food Quality and Preference, vol. 12, no. 3, pp. 207-216. Tregear, A, Dent, JB & McGregor, MJ 1994, ‘The demand for organically-grown produce’, British Food Journal, vol. 96, no. 4, pp. 21-25. 275 Tsakiridou, E, Boutsouki, C, Zotos, Y & Mattas, K 2008, ‘Attitudes and behaviour towards organic products: an exploratory study’, International Journal of Retail & Distribution Management, vol. 36, no. 2, pp. 158-175. Tsakiridou, E, Mattas, K & Tzimitra-Kalogianni, I 2006, ‘The influence of consumer characteristics and attitudes on the demand for organic olive oil’, Journal of International Food & Agribusiness Marketing, vol. 18, no. 3/4, pp. 23-31. Uncles, M & Kwok, S 2008, ‘Generalizing patterns of store-type patronage: an analysis across major Chinese cities’, International Review of Retail, Distribution and Consumer Research, vol. 18, no. 5, pp. 473-493. Ureña, F, Bernabéu, R & Olmeda, M 2008, ‘Women, men and organic food: differences in their attitudes and willingness to pay: a Spanish case study’, International Journal of Consumer Studies, vol. 32, no. 1, pp. 18-26. US Department of Agriculture 2008, The People’s Republic of China, organic products: Shanghai organic retail market profile. Vol. CH8821, Gain report. US Department of Agriculture 2012a, Regional coverage: Beijing, viewed 14 January 2012 <http://www.usdachina.org/en_pop.asp?id=66>. US Department of Agriculture 2012b, Organic production and organic food: information access tools, viewed 9 February 2012, <http://www.nal.usda.gov/afsic/pubs/ofp/ofp.shtml>. US Department of Agriculture 2012c, Sustainable agriculture: definitions and terms, viewed 20 March 2012, <http://www.nal.usda.gov/afsic/pubs/terms/srb9902terms.shtml#term23>. Usunier, J-C 2000, Marketing across cultures, 3rd edn, Pearson Education, Harlow, UK. Vandenberg, RJ & Lance, CE 2000, ‘A review and synthesis of the measurement invariance literature: suggestions, ractices, and recommendations for organizational research’, Organizational Research Methods, vol. 3, no. 1, pp. 4-69. Veal, AJ 2005, Business research methods: a managerial approach, 2nd edn, Pearson Addison Wesley, Melbourne. 276 Veeck, A 2000, ‘The revitalisaition of the market place: food market of Nanjing’ in DS Davis (ed.) The consumer revolution in urban China, University of California Press, Berkeley. Veeck, A & Burns, AC 2005, ‘Changing tastes: the adoption of new food choices in postreform China’, Journal of Business Research, vol. 58, no. 5, pp. 644-652. Verbeke, W 1999, Factors influencing the consumer decision-making process toward meat PhD thesis, Department of Agricultural Economics, Ghent University, Belgium. Verbeke, W 2000, ‘Influences on the consumer decision-making process towards fresh meat: insights from Belgium and implications’, British Food Journal, vol. 102, no. 7, pp. 522-538. Verdurme, A, Gellynck, X & Viaene, J 2002, ‘Are organic food consumers opposed to GM food consumers?’, British Food Journal, vol. 104, no. 8/9, pp. 610-623. Verdurme, A & Viaene, J 2003, ‘Consumer beliefs and attitude towards genetically modified food: basis for segmentation and implications for communication’, Agribusiness, vol. 19, no. 1, pp. 91-113. Verhoef, P, C. 2005, ‘Explaining purchases of organic meat by Dutch consumers’, European Review of Agricultural Economics, vol. 32, no. 2, pp. 245-267. Vindigni, G, Janssen, MA & Jager, W 2002, ‘Organic food consumption: a multitheoretical framework of consumer decision-making’, British Food Journal, vol. 104, no. 8/9, pp. 624-642. Walsh, AD, Laczniak, RN & Carlson, L 1998, ‘Mothers’ preferences for regulating children’s television’, Journal of Advertising, vol. 27, no. 3, pp. 23-36. Wang, C-L, Chen, Z-X, Chan, AKK & Zheng, Z-C 2000, ‘The influence of hedonic values on consumer behaviors: an empirical investigation in China’, Journal of Global Marketing, vol. 14, no. 1, pp. 169-186. Wang, CL & Chen, ZX 2004, ‘Consumer ethnocentrism and willingness to buy domestic products in a developing country setting: testing moderating effects’, Journal of Consumer Marketing, vol. 21, no. 6, pp. 391-400. 277 Wang, M 2010, The supervision and international cooperation of China organic product certification, BioFach Conference on International Organic Food Markets and Development, China Green Food Development Center, Beijing. Wang, Y, Liu, D & Tian, X 2008, ‘Analysis of consumer willingness to pay for organic food’ (in Chinese), Journal of Anhui Agricultural Sciences, vol. 36, no. 33. Wang, YJ, Doss, SK, Guo, C & Li, W 2010, ‘An investigation of Chinese consumers’ outshopping motives from a culture perspective: implications for retail and distribution’, International Journal of Retail & Distribution Management, vol. 38, no. 6, pp. 423-442. Wei, R 1997, ‘Emerging lifestyles in China and consequences for perception of advertising, buying behaviour and consumption preferences’, International Journal of Advertising, vol. 16, pp. 261-275. West, S, G, Finch, JF & Curran, PJ 1995, ‘Structural equation models with nonnormal variables’ in RH Hoyle (ed.) Structural equation modeling: concepts, issues, and applications, Sage, Thousand Oaks, Calif. Wier, M & Calverley, C 2002, ‘Market potential for organic foods in Europe’, British Food Journal, vol. 104, no. 1, pp. 45-62. Wild, JJ, Wild, KL, Han, JCY & Rammal, HG 2007, International business: the challenges of globalisation, Pearson Education, Sydney. Willer, H & Yussefi, M (eds) 2006, The world of organic agriculture: statistics & emerging trends 2006, International Federation of Organic Agriculture Movements & Research Institute of Organic Agriculture, Bonn. Williams, CM 2002, ‘Nutritional quality of organic food: shades of grey or shades of green?’, Proceedings of the Nutrition Society, vol. 61, no. 1, pp. 19-24. Williams, PRD & Hammitt, JK 2000, ‘A comparison of organic and conventional fresh produce buyers in the Boston area’, Risk Analysis, vol. 20, no. 5, pp. 735-746. 278 Wirth, F, Stanton, J & Wiley, J 2011, ‘The relative importance of search versus credence product attributes: organic and locally grown’, Agricultural and Resource Economics Review, vol. 40, no. 1, pp. 48-62. Wong, GKM & Yu, L 2003, ‘Consumers’ perception of store image of joint venture shopping centres: first-tier versus second tier cities in China’, Journal of Retailing and Consumer Services, vol. 10, no. 2, pp. 61-70. Wong, J 2004, ‘Organic food and the activist mother’, University of Auckland Business Review, vol. 6, no. 1, pp. 2-11. Worthington, RL & Whittaker, TA 2006, ‘Scale development research: a content analysis and recommendations for best practices’, Counseling Psychologist, vol. 34, no. 6, pp. 806-838. Wu, C & Zhao, R 2011, ‘Shuanghui involved in clenbuterol scandal’ (in Chinese), China Youth Daily, 24 April. Wu, Y 1999, China’s consumer revolution: the emerging patterns of wealth and expenditure, Edward Elgar, Cheltenham, UK. Xie, B, Li, T, & Yi, Q 2011, ‘Organic certification and the market: organic exports from and imports to China’, British Food Journal, vol. 113, no. 10, pp. 1200-1216. Xing, W 2011, ‘Urbanisation to exceed 70% by 2030’, China Daily, 7 April, Xu, G 2008, Olympic dreams: China and sports, 1895-2008, Harvard University Press, Cambridge, Mass. Yaakob, MR & Zakaria, A 2011, ‘Customers’ awareness, perception and future prospects of green products in Pahang, Malaysia’, Journal of Commerce, vol. 3, no. 2, pp. 1-10. Yang, KCC 2004, ‘A comparison of attitudes towards internet advertising among lifestyle segments in Taiwan’, Journal of Marketing Communications, vol. 10, no. 3, pp. 195212. Yeon, EK-K (ed.) 2008, ‘Disparity in China: poverty, inequality and interregional imbalance’ in EK-K Yeon (ed.) Facets of a transforming China: resource, trade and equity, Institute of China Studies, University of Malaya, Kuala Lumpur. 279 Yin, S, Wu, L & Chen, M 2008, ‘The analysis of Chinese consumers’ willing to pay for organic food’ (in Chinese), Journal of Agrotechnical Economics, no. 5, pp. 81-88. Yin, S, Wu, L, Du, L & Chen, M 2010, ‘Consumers’ purchase intention of organic food in China’, Journal of the Science of Food and Agriculture, vol. 90, no. 8, pp. 1361-1367. Yin, T 2010, ‘Chengdu is the best location to develop headquarter economy in west China’ (in Chinese), Chengdu Daily, 16 Aug. Yiridoe, E, Bonti-Ankomah, S & Martin, R 2005, ‘Comparison of consumer perceptions and preference toward organic versus conventionally produced foods: a review and update of the literature’, Renewable Agriculture and Food Systems, vol. 20, no. 4, pp. 193-205. Yoo, B, Donthu, N & Lee, S 2000, ‘An examination of selected marketing mix elements and brand equity’, Journal of the Academy of Marketing Science, vol. 28, no. 2, pp. 195-211. Yuan, X & Liu, T (eds) 2009, 2009 China organic food development trend report 2009 (in Chinese), Organic Service GmbH, Nanjing. Yusuf, F & Brooks, G 2010, ‘Demographics and consumption patterns in urban China’, Population Research and Policy Review, vol. 29, no. 1, pp. 5-17. Zanoli, R & Naspetti, S 2002, ‘Consumer motivations in the purchase of organic food: a means-end approach’, British Food Journal, vol. 104, pp. 643-653. Zepeda, L, Chang, H-S & Leviten-Reid, C 2006, ‘Organic food demand: a focus group study involving Caucasian and African-American shoppers’, Agriculture and Human Values, vol. 23, pp. 385-394. Zepeda, L & Deal, D 2009, ‘Organic and local food consumer behaviour: Alphabet Theory’, International Journal of Consumer Studies, vol. 33, no. 6, pp. 697-705. Zepeda, L & Li, J 2007, ‘Characteristics of organic food shoppers’, Journal of Agricultural and Applied Economics, vol. 39, no. 1, pp. 17-28. 280 Zhang, X 2002, ‘The dynamics of Chinese consumers: a case of Shanghai food consumption’, Journal of International Food & Agribusiness Marketing, vol. 14, no. 1, pp. 47-66. Zhang, X 2007, The high quality vegetable industry in China: an economic analysis and policy studies (in Chinese), PhD thesis, College of Economics & Management, Shandong Agricultural University, Shandong. Zhang, X, Dagevos, H, He, Y, van der Lans, I & Zhai, F 2008, ‘Consumption and corpulence in China: a consumer segmentation study based on the food perspective’, Food Policy, vol. 33, no. 1, pp. 37-47. Zhao, X, Chambers, E, Matta, Z, Loughin, T & Carey, E 2007, ‘Consumer sensory analysis of organically and conventionally grown vegetables’, Journal of Food Science, vol. 72, no. 2, pp. 87-91. Zhao, Y, Wells, JH, Xie, J & Xu, S 2000, ‘The features of the Chinese food packaging’, Foodservice Research International, vol. 12, no. 3, pp. 197-212. Zhou, JX, Arnold, MJ, Pereira, A & Yu, J 2010, ‘Chinese consumer decision-making styles: a comparison between the coastal and inland regions’, Journal of Business Research, vol. 63, no. 1, pp. 45-51. Zhou, X 2008a, ‘Eat, drink and sing, and be modern and global: food, karaoke and “middle class” consumers in China’ in C Jaffrelot and P van der Veer (eds) Patterns of middle class consumption in India and China, Sage, New Delhi, pp. 110-126. Zhou, Z 2008b, The development of organic sector in China, ‘Guaranteeing integrity of organic products from China’ workshop, International Federation of Organic Agriculture Movements. Zhou, Z 2011, ‘Shuanghui apologizes over additive scandal’, China Daily, 17 March. Zhou, Z, Zong, L, Yang, Y & Xiao, X 2004, The theory and practice of Chinese ecological agriculture and organic agriculture (in Chinese), China Environmental Science Publishing House, Beijing. 281 Zhu, H, Wang, Q, Yan, L & Wu, G 2009, ‘Are consumers what they consume?-Linking lifestyle segmentation to product attributes: an exploratory study of the Chinese mobile phone market’, Journal of Marketing Management, vol. 25, no. 3-4, pp. 295-314. Zhuang, G, Tsang, ASL, Zhou, N, Li, F & Nicholls, JAF 2006, ‘Impacts of situational factors on buying decisions in shopping malls: an empirical study with multinational data’, European Journal of Marketing, vol. 40, no. 1-2, pp. 17-43. 282 APPENDICES Appendix 1: Questionnaires Appendix 1A: Survey Questionnaire in English Dear Sir / Madam, This is a study conducted by Ms Jue Chen, who is a PhD student at the Swinburne University of Technology, Australia. She is researching the factors associated with the consumption of organic food products in China. The information you provide will help her to develop a better understanding of the factors that influence the Chinese consumers’ buyer behaviour of organic food products. Completion of this questionnaire is taken as your ‘Informed Consent’ to participate in this research. ‘Informed Consent’ means that: all questions about the research have been answered to your satisfaction your participation in the research is voluntary and you may withdraw at any time the survey is strictly anonymous and your responses will be kept confidential and only aggregated results (not individual responses) will be mentioned in the research output. This survey is a purely academic, and it is completely independent of any commercial interests.We highly value your feedback, and would be appreciative if you could take about 15 minutes to complete this survey. We shall be grateful if you could please inform your friends and colleagues who are consumers of organic food about this online questionnaire. If you have any questions or want more information about this survey please contact myself Jue Chen Faculty of Business and Enterprise Swinburne University of Technology PO Box 218 Hawthorn Victoria 3122, Australia Email: juchen@swin.edu.au Ph (W): +61 3 9214 5821 Or: Jue Chen Room1028, Building 8, No 83, Guanglin 1st Road, Shanghai, China Post code: 200083 Tel: 021-65171278 (China), Mobile: 135 1219 3808 (China) Email: juchen@swin.edu.au 283 Or my PhD supervisor Dr. Antonio Lobo, Faculty of Business and Enterprise Swinburne University of Technology PO Box 218 Hawthorn Victoria 3122, Australia Email: alobo@swin.edu.au Ph (W): +61 3 9214 8535 Have you ever heard of the term ‘organic food’? YES (please proceed to answer the questions below) NO (thank you for your time, you need not continue this survey) This questionnaire consists of a total of 5 sections. 284 SECTION 1: This section relates to your general knowledge about organic food 1.1 Please tick the logos you recognise. 1) 1.2 yes 2) yes 3) yes 4) yes 5) yes 6) yes How knowledgeable are you about organic food (please tick only one box)? I have heard of it, but am not sure what it means I know a little about what ‘organic’ means I know a lot about organically produced food 1.3 What is the first thing which comes to your mind when you think about organic food (please tick only one box)? No additives No pesticide Free of GM contents Environmentally produced Something natural Green Other: …………………………… (please specify) 1.4 Have you ever purchased organic food? YES (please proceed to question 1.5) NO (please go straight to section 2) 1.5 When did you last purchase any organic food? More than one year ago Half a year ago A week ago A month ago 285 Not important Neutral Important 2.1 The taste of organic food 1 2 3 4 5 2.2 The smell of organic food 1 2 3 4 5 2.3 The appearance of organic food 1 2 3 4 5 2.4 The overall quality of organic food 1 2 3 4 5 2.5 The price of organic food 1 2 3 4 5 2.6 1 2 3 4 5 2.7 The availability of organic food in convenience stores and supermarkets The promotion and advertising of organic food 1 2 3 4 5 2.8 The value of organic food relative to its price 1 2 3 4 5 2.9 The environmental benefits of organic food 1 2 3 4 5 2.10 1 2 3 4 5 2.11 The idea of face saving (mianzi) when purchasing organic food The social status of people purchasing organic food 1 2 3 4 5 2.12 The knowledge of organic food products 1 2 3 4 5 2.13 The awareness of organic food products 1 2 3 4 5 2.14 The country of origin of the organic food 1 2 3 4 5 2.15 Organic food that is produced in China 1 2 3 4 5 2.16 The brand name of the organic food 1 2 3 4 5 2.17 1 2 3 4 5 2.18 Government regulations and policies relating to the sale of organic food Food safety in relation to organic food 1 2 3 4 5 2.19 Certification relating to the quality of organic food 1 2 3 4 5 2.20 Enforcement relating to the quality of organic food 1 2 3 4 5 2.21 Packaging of organic food 1 2 3 4 5 2.22 Correct labelling of organic food 1 2 3 4 5 2.23 Information about the nutritional value of organic food 1 2 3 4 5 Extremely important Important attributes for the purchase of organic food (Circle the preferred answer) Not important at all SECTION 2: Important attributes of organic food When responding to these statements focus on your knowledge and experience for the purchase of organic foods. Please indicate by circling only one number the level of importance of each statement. Selecting 1 means that it is not important at all and selecting 5 means that it is extremely important. You may select any of the numbers in between 1 and 5 to indicate the level of importance. There are no right or wrong answers – all we are interested in is a number that best shows your experiences related to organic food. 286 SECTION 3: Your perceptions, attitudes and beliefs regarding organic food Please indicate the extent to which you agree with the statements listed below. Selecting 1 means that you strongly disagree with the statement and selecting 5 means that you strongly agree. You may select any of the numbers in between 1 and 5 to indicate the strength of your agreement. There are no right or wrong answers – all we are interested in is a number that best shows your experiences related to organic food. Disagree Neutral Agree 3.1 Organic food has a pleasant texture 1 2 3 4 5 3.2 Organic food looks nice 1 2 3 4 5 3.3 Organic food smells nice 1 2 3 4 5 3.4 Organic food tastes good 1 2 3 4 5 3.5 Consuming organic food is trendy 1 2 3 4 5 3.6 Organic food has high nutritional value 1 2 3 4 5 3.7 Generally speaking, the higher the price of a 1 2 3 4 5 Strongly Agree organic food (Circle your preferred answer) Strongly Disagree Perception, attitudes and beliefs regarding product, the higher the quality 3.8 The price of organic food is too high 1 2 3 4 5 3.9 Organic food is good value for money 1 2 3 4 5 3.10 When I buy a food product, I always read the label 1 2 3 4 5 3.11 I don’t trust organic food certification bodies 1 2 3 4 5 3.12 I trust the outlets which sell certified organic food 1 2 3 4 5 3.13 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 3.16 Chinese government regulate food marketing, and have developed significant policies Lack of adequate government control of media allows advertisers to take advantage of consumers Logos depicting types of organic foods should be controlled More land should be allocated for organic farming 1 2 3 4 5 3.17 The market for ‘organic’ and ‘green’ is chaotic 1 2 3 4 5 3.18 1 2 3 4 5 3.19 Organic food is good for myself and my family’s health Organic food has no harmful effects 1 2 3 4 5 3.20 I like the brands associated with organic food 1 2 3 4 5 3.21 1 2 3 4 5 3.22 I worry about there being harmful chemicals in my food Organic food does not contain pesticides 1 2 3 4 5 3.23 Organic food is good for the environment 1 2 3 4 5 3.14 3.15 287 3.24 I believe that organic food has superior quality 1 2 3 4 5 3.25 I personally think I should always buy organic food 1 2 3 4 5 3.26 To me, it is important that the food I usually eat can be easily found in the food outlets near my house or workplace Organic food labels mean high quality food products When I like something, I will buy it without too much deliberation I always do whatever I feel like and whenever I feel like it The sole purpose of making money is to spend it 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 3.32 Sometimes I feel like spending money on anything I lay my eyes on I often make impulse purchases 1 2 3 4 5 3.33 I always try something new and unique 1 2 3 4 5 3.34 I love fashionable and trendy products 1 2 3 4 5 3.35 It does not hurt to be trendy if I feel like it 1 2 3 4 5 3.36 I am often influenced by advertisements of new products I can easily influence people around me during conversation My friends often consult me when they cannot make up their own mind I have a strong desire to be successful 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Chinese people should buy domestic products rather than imported products China should levy heavy tariff on foreign products to reduce their quantity into China If two organic food products were the same in quality, but one was imported and the other was Chinese, I would pay more for the imported product I would purchase organic food coming from a country I approve of politically Chinese people should not buy foreign products, because it would hurt domestic business, and cause more unemployment 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 3.27 3.28 3.29 3.30 3.31 3.37 3.38 3.39 3.40 3.41 3.42 3.43 3.44 288 4.4 4.5 4.6 4.7 4.8 I am eager to check out organic food products because of advertisements and promotion I intend to try out organic food products 1 2 3 4 5 1 2 3 4 5 I am interested in experiencing the benefits of using organic foods It is likely that I will buy organic food products when they become available I can recall the brand names and labelling of some of the organic food products I will probably use organic food products in the future 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 I will recommend usage of organic foods to my friends and relatives If I had to purchase organic foods again, I would make the same choice 1 2 3 4 5 1 2 3 4 5 Neutral Strongly Agree 4.3 Agree 4.2 Disagree 4.1 Purchase intention regarding organic food (Circle your preferred answer) Strongly Disagree SECTION 4: Your purchase intention regarding organic food Please indicate the extent to which you agree with the statements listed below. Selecting 1 means that you strongly disagree with the statement and selecting 5 means that you strongly agree. You may select any of the numbers in between 1 and 5 to indicate the strength of your agreement. There are no right or wrong answers – all we are interested in is a number that best shows your experiences related to organic food. SECTION 5: May I ask a few questions about you and your family? The following questions are for classification purposes only, and will be kept strictly confidential. 5.1 What types of organic foods have you recently purchased? Dairy Fruit & Vegetable Meat & Seafood Grains Other 5.2 What is your intended use for the most recent purchase of organic food? Own consumption for family for children for elderly people Gift Other 5.3 How much extra are you willing to pay for organic food as compared to conventional food? (only tick one) Between 20%-50% Between 51%-100% Great than 100% No extra 5.4 If you were buying imported organic food, from which country or region would you prefer most? (only tick one) Europe U.S.A Australia / New Zealand Japan / Korea HK / Taiwan Don’t mind 289 5.5 Where would you prefer to buy organic food from? (only tick one) Supermarket On line Local market Mail order catalogues Organic farm Speciality stores Other 5.6 How often do you purchase organic food products? (only tick one) At least once a week Every fortnight Monthly Occasionally 5.7 Please indicate your monthly family income? (tick one) Less than RMB5, 000 Between RMB5, 001-10,000 Between RMB10, 001-20,000 More than RMB20, 000 5.8 Your household includes? (tick one) Yourself (single) Couple without children With young children With parents With adult children 5.9 Please indicate your gender (tick one) Female Male 5.10 Please indicate your age group: (tick one) Between 18-30 5.11 Between 31-45 Bachelor degree 5.14 High school 2 years college or associate’s degree Postgraduate or above Please indicate your occupation: (tick one) White collar 5.13 Greater than 60 Please specify the highest level of education you have completed: (tick one) Below high school 5.12 Between 46-60 Not working Student Blue collar Other Please indicate which city are you located in? (only tick one) Beijing Shanghai Chengdu Other Shenzhen Please feel free to add any additional information you would like to provide with regard to the purchase of organic food that has not previously been addressed ______________________________________________________________________________________ _______________________________________________________________________________________ THANK YOU FOR YOUR TIME. IT IS VERY MUCH APPRECIATED. 290 Appendix 1B: Survey Questionnaire translated into Chinese Ё᳝ᴎ亳ક⍜䌍ᢑḋ䇗ᶹ ᭀਃ㗙ˈ ▇߽Ѯᮃ࿕ᴀᄺ䆮䙔䇋ᙼখϢϔ乍݇Ѣ᳝ᴎ亳કЁ⍜䌍㗙ⱘ㘨ড়䇗ᶹDŽ៥Ӏᳳᳯ䖭乍 ⷨおৃҹࡴ⏅ᄺᴃ⬠ᇍ䆹乚ඳⱘ䖯ϔℹ䅸ⶹˈ᳝ࡽѢདഄњ㾷Ё⍜䌍㗙䌁ф᳝ᴎ亳કⱘ އᅮᗻ㋴DŽ ᙼᦤկⱘ᠔ֵ᳝ᙃᇚҹϡ䆄ৡᮍᓣ䆄ᔩ, ᠔᳝䯂ो㒧ᵰᇚҹ䲚ড়ᮍᓣ㸼⦄ˈ ӏԩҎ䛑᮴⊩䆚 ߿ᙼⱘϾҎ䌘᭭DŽ ᅠ៤ᴀ䇗ⷨⱘ䯂ोᇚ㾚Ў˖ x ᙼৠᛣᇍ䯂ोЁⱘ᠔᳝䯂乬ߎㄨ x ᙼ㞾ᜓখϢᴀ䇗ⷨ ( ᙼгৃҹ䱣ᯊ⊼䫔ᙼⱘ䯂ो) ᴀ䇗ⷨЎ㒃ᄺᴃᗻ⌏ࡼˈϢӏԩଚϮ⌏ࡼ᮴݇DŽ བᵰ㛑ऴ⫼ᙼᅱ䌉ⱘ 15 ߚ䩳฿ݭᴀ䯂ो, ៥Ӏ㸼⼎कߚᛳ䇶DŽ བ᳝ӏԩ⭥䯂ˈ Ꮰᳯ㦋ᕫ᳝݇䇗ⷨⱘ䖯ϔℹ䌘᭭ˈ 䇋㘨㋏ Jue Chen (मⷨお⫳˖䰜⦣) Faculty of Business and Enterprise Swinburne University of Technology PO Box 218 Hawthorn Victoria 3122, Australia Email: juchen@swin.edu.au Ph (W): +61 3 9214 5821˄▇߽Ѯ˅ 13512193808 (Ё) म⫳ᇐᏜ˖ Dr. Antonio Lobo, alobo@swin.edu.au ᙼ䇈䖛᳝ᴎ亳ક৫˛ 䇈䖛, 䇋㒻㓁ಲㄨҹϟ䯂乬 ≵䇈䖛 ˄䇶䇶ᙼˈ ৃϡᖙ㒻㓁ℸ䯂ो˅ ᴀ䯂ो݅ЎѨϾ䚼ߚ˖ 291 ϔ䚼ߚ: ᳝ᴎ亳કⱘϔ㠀ᐌ䆚 1.1 䇋䗝ᢽᙼⶹ䘧ⱘᷛ䆚 (ᙼৃҹ䗝ᢽϾˈгৃҹϡ䗝) 1) ⶹ䘧 2) ⶹ䘧 3) ⶹ䘧 4) ⶹ䘧 5) ⶹ䘧 6) ⶹ䘧 1.2 ᳝ᴎ亳કⶹ䆚㞾៥䆘Ӌ(া㛑䗝ϔϾㄨḜ)? ៥䇈䖛 “᳝ᴎ亳ક” ˈԚϡ⹂ᅮᰃҔМᛣᗱ ៥ᇍ”᳝ᴎ亳ક”ᰃҔМ᳝ϔ⚍њ㾷 ៥ᕜњ㾷ҔМᰃ “᳝ᴎ亳ક” 1.3 ᔧᙼᛇࠄ᳝ᴎ亳કˈᙼⱘϔὖᗉᰃҔМ(া㛑䗝ϔϾㄨḜ)? ᮴⏏ࡴࠖ ᮴࣪ᄺݰ㥃 ᮴䕀 ⦃ֱѻક ✊ⱘ 㓓㡆ⱘ ݊Ҫ: …………………………….(䇋⊼ᯢ) 1.4 ᙼ䌁ф䖛᳝ᴎ亳ક৫˛ ф䖛 ˄䇋㒻㓁ಲㄨ䯂乬 1.5˅ ≵᳝ф䖛 ( 䇋ⳈಲㄨѠ䚼ߚ) 1.5 ᙼϞϔᰃҔМᯊ䌁фⱘ᳝ᴎ亳ક? ϔᑈҹࠡ ञᑈҹࠡ ϔϾ᳜ҹࠡ ϔϾ᯳ᳳҹࠡ 292 Ѡ䚼ߚ˖᳝ᴎ亳કⱘ䞡㽕⡍ᗻ 2 . 1 ᳝ ᴎ 亳 ક ⱘ ষ ᛳ 2 . 2 ᳝ ᴎ 亳 ક ⱘ ⇨ ੇ 2 . 3 ᳝ ᴎ 亳 ક ⱘ 㾖 2 . 4 ᳝ ᴎ 亳 ક ⱘ ᭈ ԧ ક 䋼 2 . 5 ᳝ 2 . 6 ৃҹ߽֓ᑫ䍙Ꮦ䌁фࠄ᳝ᴎ亳ક 2 . 7 ᳝ ᴎ 亳 ક ⱘ ֗ 䫔 ᑓ ਞ 2 . 8 ᳝ᴎ亳કᗻӋ↨催ˈ⠽᳝᠔ؐ 2 . 9 ᳝ ᴎ 亳 ક ᳝ Ⲟ Ѣ ⦃ ֱ 2.10 䌁 ф ᳝ ᴎ 亳 ક ᰃ ᳝ 䴶 ᄤ ⱘ џ 2.11 䌁ф᳝ᴎ亳કᯊⱘҎⱘ⼒Ӯഄԡ 2.12 ᳝ ᴎ 亳 ક ѻ ક ⱘ ⶹ 䆚 2.13 ᳝ ᴎ 亳 ક ѻ ક ⱘ ⶹ ৡ ᑺ 2.14 ᳝ 2.15 ѻ ⱘ ᳝ ᴎ 亳 ક 2.16 ᳝ ᴎ 亳 ક ⱘ ક ⠠ 2.17 ᬓᑰ݇Ѣ䫔ଂ᳝ᴎ亳કⱘ⊩㾘ᮍ䩜 2.18 亳 ક ᅝ ܼ ⍝ ঞ ࠄ ᳝ ᴎ 亳 ક 2.19 ݇ Ѣ ᳝ ᴎ 亳 ક ⱘ ક 䋼 ⱘ 䅸 䆕 2.20 ᳝ ᴎ 亳 ક ⱘ ક 䋼 ⱘ ᠻ 㸠 2.21 ᳝ 2.22 ᳝ ᴎ 亳 ક ⱘ ℷ ⹂ ᷛ 䆚 2.23 ݇Ѣ᳝ᴎ亳કⱘ㧹ݏӋؐⱘֵᙃ ᴎ ᴎ ᴎ 亳 亳 亳 ક ક ⱘ ⱘ ક ॳ ⱘ Ӌ ѻ ࣙ ᕜ䞡㽕 䞡㽕 Ёゟ ϡ䞡㽕 䌁ф᳝ᴎ亳કⱘ䞡㽕ሲᗻ (䇋䗝ᢽⳌᑨ ⱘ᭄ᄫ) ḍᴀϡ 䞡㽕 ѢᙼϾҎᇍ᳝ᴎ亳કⱘԧ偠䅸䆚ˈ 䇋ᇍϟ߫↣ϔϾ㾖⚍㸼ᯢᙼৠᛣϢ৺ⱘᑺDŽᙼৃҹ䗝ᢽ”ḍ ᴀϡ䞡㽕”ˈ”ϡ䞡㽕”ˈ”Ёゟ”ˈ”䞡㽕””ᕜ䞡㽕”ЁⱘӏԩϔϾ㾖⚍DŽ ᇍᴀ䇗ᶹ㗠㿔, ᙼⱘ㾖⚍≵᳝ ᇍ䫭ПߚDŽ䇋䗝ᢽⳌᑨⱘᮍḚDŽ Ḑ 㺙 293 ϝ䚼ߚ: ݇Ѣ᳝ᴎ亳કⱘ 䅸ⶹ, ᗕᑺ ֵӏᑺ 䇋㸼ᯢᙼᇍϟ߫↣ϔϾ㾖⚍ⱘৠᛣᑺDŽᙼৃҹ䗝ᢽ “ᔎ⚜ডᇍ”ˈ”ডᇍ”ˈ”Ёゟ”ˈ”ৠᛣ””䴲ᐌ ৠᛣ”ЁⱘӏԩϔϾDŽᙼⱘ㾖⚍≵᳝䫭ᇍПߚˈ៥ӀাᰃᏠᳯњ㾷ᙼᇍ᳝ᴎ亳કⱘ䅸ⶹ, ᗕᑺֵӏ ᳝ᴎ亳કᕜད䯏 3.4 ᳝ᴎ亳કᕜདৗ 3.5 ᳝ᴎ亳કᕜ亢㸠 3.6 ᳝ᴎ亳ક᳝ᕜ催ⱘ㧹ݏӋؐ 3.7 ϔ㠀㗠㿔ˈѻક䋼䞣䍞དˈӋḐህ䍞䌉 3.8 ᳝ᴎ亳ક䌉њ 3.9 ᳝ᴎ亳ક⠽᳝᠔ؐ 3.10 ᔧ៥䌁ф亳કᯊˈ៥ϔᅮӮⳟଚᷛ 3.11 ៥ϡֵӏ᳝ᴎ亳કⱘ䅸䆕ᴎᵘ 3.12 3.15 䙷ѯߎଂ䅸䆕䖛ⱘ᳝ᴎ亳કⱘ䳊ଂଚ䅽៥ ֵᕫ䖛 ᬓᑰᏆ㒣ߎৄњ䞡㽕ⱘᬓㄪˈ㾘㣗њ亳ક ⱘ䫔ଂ ᆊ㔎Уᇍၦԧᑨ᳝ⱘⲥˈ䅽ᑓਞଚ䗮 䖛䇃ᇐ⍜䌍㗙㗠㦋߽ ⫼ᴹ㸼ᕕ᳝ᴎ亳કⱘଚᷛᑨ䆹ফࠄⲥㅵ 3.16 ⱘೳഄᑨ䆹䕀࣪Ў᳝ᴎ⬄ݰ 3.17 Ꮦ䴶Ϟᇍ”᳝ᴎ””㓓㡆”ⱘὖᗉᕜ⏋х 3.18 ᳝ᴎ亳ક᳝ⲞѢ៥ᆊҎⱘعᒋ 3.19 ᳝ᴎ亳ક᮴᳝ᆇᬜᑨ 3.20 ៥୰Ϣ᳝ᴎ亳ક᳝݇ⱘક⠠ 3.21 ៥ᕜᢙᖗ៥ⱘ亳⠽䞠Ӯ᳝᳝ᆇⱘ࣪ড়⠽ 3.22 ᳝ᴎ亳કϡݰ㥃 3.23 ᳝ᴎ亳ક᳝ⲞѢ⦃๗ 3.24 ៥മֵ᳝ᴎ亳ક䋼䞣ϞЬ 3.13 3.14 䴲ᐌ ৠᛣ 3.3 ৠᛣ ᳝ᴎ亳કⳟϞএϡ䫭 Ёゟ 3.2 ডᇍ 3.1 ᳝ᴎ亳કⱘ䅸ⶹ, ⳟ ⊩ֵӏᑺ( 䗝ᢽ ⱘ᭄ᄫϞ⬏) ᳝ᴎ亳ક᳝ҸҎᛝᙺⱘ䋼ᛳ ᔎ⚜ ডᇍ ᑺ DŽᙼৃҹ䗝ᢽӏԩϔϾᮍḚ㸼ᯢᙼⱘৠᛣϢ৺ⱘᑺDŽ 294 3.25 3.26 3.27 3.28 ህ៥ϾҎ㗠㿔ˈ៥䅸Ў៥ᑨ䆹ᘏᰃф᳝ᴎ 亳ક ᇍ៥㗠㿔ˈ ៥ᆊ៥Ꮉ䰘䖥㛑фࠄ៥ ৗⱘ亳⠽ᕜ䞡㽕 ᳝ᴎ亳કⱘᷛㅒᷛᖫⴔ亳⠽ⱘ催䋼䞣 3.30 ៥୰ҔМˈ ህӮ↿ϡ⢍䈿ഄфϟˈϡএ ⏅ᗱ❳㰥 ៥ᘏᰃᛇфҔМህфҔМˈ ᛇҔМᯊф ህҔМᯊф 䌮䪅ⱘଃϔⳂⱘህᰃ㢅䪅 3.31 ᳝ᯊ៥ⳳᛇ㾕ҔМфҔМ 3.32 ៥ᐌӮ᳝ⱘࡼކ䌁ф 3.33 ៥ᘏᰃᇱ䆩ᮄ༛ⱘϰ㽓 3.34 ៥୰ᮄ╂ⱘϰ㽓 3.35 া㽕៥催݈ˈ䍊ᯊ傺≵ҔМ䯂乬 3.36 ៥ᕜᆍᯧ㹿ᮄѻકⱘᑓਞ᠔ᕅડ 3.37 ៥ᕜᆍᯧѸ䇜ЁऴЏ㾦 3.38 ៥ⱘ᳟টҪӀᣓϡᅮЏᛣᯊ㒣ᐌ䇋ᬭ៥ 3.39 ៥␈ᳯ៤ࡳ 3.40 ЁҎᑨ䆹фѻⱘˈϡᑨ䆹ф䖯ষⱘϰ 㽓 Ёᬓᑰᑨ䆹ᇍ䖯ষѻકࡴ݇ˈ Ң㗠 ޣᇥ䖯ষѻક᭄䞣 བᵰϸ⾡᳝ᴎ亳ક᳝ৠḋⱘ䋼䞣, ϔ⾡ᰃ 䖯ষⱘˈϔ⾡ᰃѻⱘˈ៥ᚙᜓҬ⚍䪅 䌁ф䖯ষⱘ᳝ᴎ亳ક ៥ᜓᛣ䌁ф᳝ᴎ亳કᴹ㞾Ѣ៥ᬓ⊏Ϟ䅸ৃ ⱘᆊ ЁҎϡᑨ䆹䌁фѻકˈ৺߭Ӯᕅડ ⇥ᮣѻϮˈᇐ㟈ⱘЁҎ༅ϮDŽ 3.29 3.41 3.42 3.43 3.44 295 ಯ䚼ߚ˖᳝ᴎ亳કⱘ䌁фᛣ 䇋㸼ᯢᙼᇍϟ߫↣ϔϾ㾖⚍ⱘৠᛣᑺDŽᙼৃҹ䗝ᢽ”ᔎ⚜ডᇍ”ˈ”ডᇍ”ˈ”Ёゟ”ˈ”䌲៤”ˈ”䴲ᐌ䌲 4.2 ⬅Ѣᑓਞ֗䫔ⱘॳˈ៥䖿ߛᏠᳯ䌁ф᳝ᴎ 亳ક ៥ᠧㅫᇱ䆩᳝ᴎ亳ક 4.3 ៥᳝݈䍷ԧ偠ৗ᳝ᴎ亳કⱘⲞ໘ 4.4 བᵰ᳝ᴎ亳ક᳝पˈ៥ᕜৃ㛑Ӯф 4.5 ៥㛑ᛇ䍋᳝ѯ᳝ᴎ亳કⱘક⠠ଚᷛ 4.6 ᇚᴹ៥ৃ㛑Ӯৗ᳝ᴎ亳ક 4.7 ៥Ӯ៥ⱘ᳟҆ট㤤᳝ᴎ亳કⱘ⫼䗨 4.8 བᵰ៥ϡᕫϡݡϔ䌁ф᳝ᴎ亳કˈ៥Ӯخৠ ḋⱘ䗝ᢽ 4.1 Ѩ䚼ߚ˖ৃҹ䯂ᙼϾ᳝݇ᙼϾҎᆊᒁⱘ䯂乬৫˛ ϟ߫䯂乬ҙ⫼Ѣߚ㉏⫼ˈ ᇚӮϹḐֱᆚ 5.1 ᳔䖥ᙼ䌁ф䖛ા⾡᳝ᴎ亳ક? ˄ৃҹ䗝˅ чࠊક 5.2 㒭ᄽᄤ 5.6 ݊Ҫ______ 㒭㗕Ҏ 䗕⼐ ܼᆊৗ ݊Ҫ_______ ᬃҬ 51%-100% Ѣ 100% ϡᜓҬ བᵰᙼ䳔㽕䗝ᢽ䖯ষ᳝ᴎ亳કˈ ᙼᜓᛣ䌁фાϾᆊഄऎⱘѻક˛(া㛑䗝ϔϾㄨḜ) ⌆ 5.5 䈋⠽㊂亳 ⳌᇍѢϔ㠀亳કˈᙼᜓᛣࡴᇥ䞥乱䌁фৠ㉏ⱘ᳝ᴎ亳ક? (া㛑䗝ϔϾㄨḜ˅ ᬃҬ 20%-50% 5.4 㙝㉏⍋剰 ᙼ᳔䖥䌁ф᳝ᴎ亳કᰃЎњ ? ˄ৃҹ䗝˅ 㞾Ꮕৗ 5.3 ∈ᵰ㬀㦰 㕢⌆ ᮹ᴀ/䶽 ৄ/佭␃ ▇߽Ѯ/ᮄ㽓݄ ᮴᠔䇧 ᙼ୰ԩ໘䌁ф᳝ᴎ亳ક˛(া㛑䗝ϔϾㄨḜ˅ 䍙Ꮦ ݰ䌌Ꮦഎ ᳝ᴎݰഎ ϧपᑫ 㔥Ϟ Ⳃᔩ䫔ଂ ݊Ҫ _______ ᙼ䌁ф᳝ᴎ亳કⱘ਼ᳳЎ? (া㛑䗝ϔϾㄨḜ˅ 㟇ᇥ↣᯳ᳳϔ ↣ϸ᯳ᳳ ↣᳜ يᇨ 296 䴲ᐌ 䌲៤ Ёゟ ডᇍ ᔎ⚜ ᳝ᴎ亳કⱘ䌁фᛣ ( 䗝ᢽⱘ᭄ᄫϞ⬏) ডᇍ ៤”ЁⱘӏԩϔϾ㾖⚍ˈ㸼ᯢᙼⱘৠᛣϡৠᛣᑺˈᙼⱘ㾖⚍≵᳝䫭ᇍПߚDŽ 5.7 䇋䗝ᢽᙼⱘᆊᒁ᳜ᬊܹ? (া㛑䗝ϔϾㄨḜ˅ ԢѢҎ⇥Ꮥ 5ˈ000 ܗ Ҏ⇥Ꮥ 5, 001-10,000 П䯈 Ҏ⇥Ꮥ 10, 001-20,000 П䯈 催Ѣ Ҏ⇥Ꮥ 20, 000 5.8 ᙼᆊ䞠ԣⴔ䙷ѯҎ? (া㛑䗝ϔϾㄨḜ) ϔϾҎԣ (ऩ䑿) ཛ˄≵ᄽᄤ˅ Ϣ≵៤ᑈⱘᄽᄤϔ䍋ԣ Ϣ㗕Ҏϔ䍋ԣ Ϣ៤ᑈᄽᄤϔ䍋ԣ 5.9 䇋䗝ᢽᙼⱘᗻ߿: (া㛑䗝ϔϾㄨḜ˅ ⬋ 5.10 ཇ 䇋䗝ᢽᙼⱘᑈ啘↉: (া㛑䗝ϔϾㄨḜ˅ 18 -30 П䯈 5.11 ᴀ⾥ 催Ё/Ёϧ ϧ ⸩ҹϞ˄ࣙᣀ⸩˅ 㪱乚 ϡᎹ ᄺ⫳ ݊Ҫ _______ 䇋䗝ᢽᙼⳂࠡሙԣජᏖ˖(া㛑䗝ϔϾㄨḜ˅ ࣫Ҁ ៤䛑 5.14 60 ቕҹϞ 䇋䗝ᢽᙼⳂࠡ㘠Ϯ: (া㛑䗝ϔϾㄨḜ˅ ⱑ乚 5.13 46-60 П䯈 䇋䗝ᢽᙼⱘ᳔催ᄺग़: (া㛑䗝ϔϾㄨḜ˅ 催Ёҹϟ 5.12 31-45 П䯈 Ϟ⍋ ⏅ഇ ݊Ҫ ________ བᙼ䅸Ў᳝ӏԩᴀ䯂ोЁ⍝ঞⱘ݇Ѣ᳝ᴎ亳ક⍜䌍ⱘֵᙃঞ䯂乬ˈ 䇋ᙼ䇈ᯢ üüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüü üüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüüü 䴲ᐌᛳ䇶ᙼⱘᬃᣕ! 297 Appendix 2: Summary of key authors from selected studies in different continents and countries Country Austria Belgium Croatia Denmark Finland Germany Greece Ireland Italy Netherlands Norway Slovenia Spain Sweden Turkey United Kingdom Authors Europe Gotschi et al. 2010 Aertsens, Mondelaers & Huylenbroeck 2009; Verdurme, Gellynck & Viaene 2002 First & Brozin 2009; Radman 2005 Grunert & Juhl 1995; Millock, Wier & Andersen 2004 Tarkiainen & Sundqvist 2005 Janssen, Heid & Hamm 2009 Arvanitoyannis, Krystallis & Kapirti 2003; Botonaki et al. 2006; Chryssochoidis 2000; Chryssochoidis & Krystallis 2005; Fotopoulos & Chryssochoidis 2000; Fotopoulos & Krystallis 2002a, b; Fotopoulos, Krystallis & Ness 2003; Krystallis, Arvanitoyannis & Chryssohoidis 2006; Krystallis & Chryssohoidis 2005; Krystallis, Fotopoulos & Zotos 2006; Magkos, Arvaniti & Zampelas 2006; Tsakiridou et al. 2008; Tsakiridou, Mattas & Tzimitra-Kalogianni 2006 Davies, Titterington & Cochrane 1995; Moore 2006; O’Donovan & McCarthy 2002; Roddy, Cowan & Hutchinson 1994; Roddy, Cowan & Hutchinson 1996 Canavari et al. 2002; Chinnici, D’Amico & Peorino 2002; Cicia, Giudice & Scarpa 2002; de Magistris & Gracia 2008; Saba & Messina 2003; Vindigni, Janssen & Jager 2002; Zanoli & Naspetti 2002 Schifferstein & Oude Ophuis 1998; Stobbelaar et al. 2007; Verhoef 2005 Honkanen, Verplanken & Olsen 2006; Storstad & Bjørkhaug 2003; Torjusen et al. 2001 Kuhar & Juvancic 2010 Gil, Gracia & Sanchez 2000; Sanjuán et al. 2003; Soler, Gil & Sánchez 2002; Ureña, Bernabéu & Olmeda 2008 Magnusson et al. 2001; Magnusson et al. 2003; Shepherd, Magnusson & Sjödén 2005; Tanner & Kast 2003 Oraman & Unakitan 2010; Özcelik & Uçar 2008 Barnes, Vergunst & Topp 2009; Beharrell & MacFie 1991; Brennan, Gallagher & McEachern 2003; Connor & Douglas 2001; Harper & Makatouni 2002; Hill & Lynchehaun 2002; Lodorfos & Dennis 2008; Makatouni 2002; McEachern & McClean 2002; McEachern & Willock 2004; Padel & Foster 2005; Rimal, Moon & Balasubramanian 2005; Sparks & Shepherd 1992; Tregear, Dent & McGregor 1994 298 Overall European region United States Canada Costa Rica Australia New Zealand China India Japan Malaysia Taiwan Thailand DenmarkNew Zealand Denmark, Great Britain, and Italy France-Italy Germany-UK Aarset et al. 2004; Margetts et al. 1997; Thøgersen 2009; Wier & Calverley 2002 Americas Abrams, Meyers & Irani 2010; Bellows et al. 2008; Dahm, Samonte & Shows 2009; Finch 2005; Govindasamy, DeCongelio & Bhuyan 2005; Howard & Allen 2006; Huang 1996; Kiesel & Villas-Boas 2007; Klonsky & Greene 2005; Li, Zepeda & Gould 2007; Lin, Smith & Huang 2008; Loureiro & Hine 2002; Onozaka 2007; Onyango, Hallman & Bellows 2007; Rimal, Moon & Balasubramanian 2006; Smith, Huang & Lin 2009a, b; Thompson 1998; Thompson & Kidwell 1998; Williams & Hammitt 2000; Wirth, Stanton & Wiley 2011; Zepeda, Chang & Leviten-Reid 2006; Zepeda & Deal 2009; Zepeda & Li 2007; Zhao et al. 2007 Anders & Moeser 2008; Essoussi & Zahaf 2008; Haghiri & McNamara 2007; Hamzaoui-Essoussi & Zahaf 2012; Larue et al. 2004 Aguirre 2007; González 2009 Oceania Bhaskaran et al. 2006; Chang & Zepeda 2005; Chang, Zepeda & Griffith 2005; Kristiansen et al. 2010; Kristiansen & Smithson 2008; Lea & Worsley 2005; Lockie et al. 2004; Lockie et al. 2002; Kristiansen & Smithson 2008; Paull 2008b; Smith & Paladino 2010 Mather, Knight & Holdsworth 2005; Wong 2004 Asia Dai, Zhu & Ying 2006; International Trade Centre 2011; Nees 2011; Thøgersen & Zhou 2012; Wang, Liu & Tian 2008; Xie, Li, & Yi 2011; Yin, Wu & Chen 2008; Yin et al. 2010 Chakrabarti & Baisya 2007 Gendall, Betteridge & Bailey 1999; Hu, Chen & Yoshida 2006 Ahmad & Juhdi 2010; Yaakob & Zakaria 2011Salleh et al. 2010; Shaharudin et al. 2010a, 2010b,. 2010c Chen 2007, 2009 Roitner-Schobesberger et al. 2008 Cross-cultural studies Squires, Juric & Cornwell 2001 et al. 2012 Guido et al. 2010 Baker, Thompson & Engelken 2004 Source: author’s research 2011 299 Appendix 3: Ethics approval From: “Kaye Goldenberg” Wednesday-November 11, 2009 12:11 <KGOLDENBERG@groupwise.swin. PM edu.au> To: <alobo@swin.edu.au>, <juchen@swin.edu.au> CC: “Resethics” <Resethics@groupwise.swin.edu.au> Subject: SUHREC Project 2009/245 Ethics Clearance [View] [Save As] Attachments: Mime.822 (5 KB) To: Dr Antonio Lobo, FBE/Ms Jue Chen Dear Dr Lobo, SUHREC Project 2009/245 A study investigating the determinants of consumer buyer behaviour relating to purchase of organic food products in China Dr Antonio Lobo, FBE/Ms Jue Chen Approved Duration: 11/11/2009 To 11/11/2010 [Adjusted] I refer to the ethical review of the above project protocol undertaken on behalf of Swinburne’s Human Research Ethics Committee (SUHREC) by SUHREC Subcommittee (SHESC2) at a meeting held on 15 October 2009. Your responses to the review, and request for amendment to the protocol (by revising the questionnaire), as e-mailed on 4 and 5 November respectively, were put to a nominated SHESC2 delegate for consideration. I am pleased to advise that, as submitted to date, the project has approval to proceed in line with standard on-going ethics clearance conditions here outlined. - All human research activity undertaken under Swinburne auspices must conform to Swinburne and external regulatory standards, including the National Statement on Ethical Conduct in Human Research and with respect to secure data use, retention and disposal. - The named Swinburne Chief Investigator/Supervisor remains responsible for any 300 personnel appointed to or associated with the project being made aware of ethics clearance conditions, including research and consent procedures or instruments approved. Any change in chief investigator/supervisor requires timely notification and SUHREC endorsement. - The above project has been approved as submitted for ethical review by or on behalf of SUHREC. Amendments to approved procedures or instruments ordinarily require prior ethical appraisal/ clearance. SUHREC must be notified immediately or as soon as possible thereafter of (a) any serious or unexpected adverse effects on participants and any redress measures; (b) proposed changes in protocols; and (c) unforeseen events which might affect continued ethical acceptability of the project. - At a minimum, an annual report on the progress of the project is required as well as at the conclusion (or abandonment) of the project. - A duly authorised external or internal audit of the project may be undertaken at any time. Please contact me if you have any queries about on-going ethics clearance. The SUHREC project number should be quoted in communication. Chief Investigators/Supervisors and Student Researchers should retain a copy of this e-mail as part of project record-keeping. Best wishes for the project. Kaye Goldenberg Secretary, SHESC2 ******************************************* Kaye Goldenberg Administrative Officer (Research Ethics) Swinburne Research (H68) Swinburne University of Technology P O Box 218 HAWTHORN VIC 3122 Tel +61 3 9214 8468 Fax +61 3 9214 5267 301 Appendix 4: Detailed analysis of the Pilot study(study 1) Table A: Descriptive information 3.1 Organic food has a pleasant texture Mean 5% Trimmed Mean 3.2 Organic food looks nice Mean 5% Trimmed Mean 3.3 Organic food smells nice Mean 5% Trimmed Mean 3.4 Organic food tastes good Mean 5% Trimmed Mean 3.5 Organic products are in fashion Mean 5% Trimmed Mean 3.6 Organic food has high nutritional value Mean 5% Trimmed Mean 3.7 Generally speaking, the higher the price of a product, the Mean higher the quality 5% Trimmed Mean 3.8 The price of organic food is too high Mean 5% Trimmed Mean 3.9 Organic food is good value for money Mean 5% Trimmed Mean 3.10 When I buy a food product, I always read the label Mean 5% Trimmed Mean 3.11 I don’t trust organic food certification bodies Mean 5% Trimmed Mean 3.12 I trust the outlets which sell certified organic food Mean 5% Trimmed Mean 3.13 Chinese government regulate food marketing, and have Mean developed significant policies 5% Trimmed Mean 3.14 Lack of adequate government control of media allows Mean advertisers to take advantage of consumers 5% Trimmed Mean 3.15 Logos depicting types of organic foods should be controlled Mean 5% Trimmed Mean 3.16 More land should be allocated for organic farming Mean 5% Trimmed Mean 3.17 The market for organic and green is chaotic Mean 5% Trimmed Mean 3.54 3.57 3.63 3.66 3.29 3.31 3.49 3.49 3.11 3.13 3.64 3.66 3.56 3.62 3.99 4.02 3.28 3.28 3.81 3.84 3.17 3.16 3.27 3.29 3.00 2.99 3.87 3.94 4.31 4.38 3.96 4.02 4.32 4.38 302 3.18 Organic food is good for myself and my family’s health Mean 5% Trimmed Mean 3.19 Organic food has no harmful effects Mean 5% Trimmed Mean 3.20 I like the brands associated with organic food Mean 5% Trimmed Mean 3.21 I worry about harmful chemicals in my food Mean 5% Trimmed Mean 3.22 Organic food does not contain pesticides Mean 5% Trimmed Mean 3.23 Organic food is good for the environment Mean 5% Trimmed Mean 3.24 I believe organic food has superior quality Mean 5% Trimmed Mean 3.25 I personally think I should always buy organic food Mean 5% Trimmed Mean 3.26 To me, it is important that the food I usually eat can be Mean easily found in the food outlets near my house or work place 5% Trimmed Mean 3.27 Organic food labels mean high quality food products Mean 5% Trimmed Mean 3.28 When I like something, I will buy it without too much Mean deliberation 5% Trimmed Mean 3.29 I always do whatever I feel like and whenever I feel like it Mean 5% Trimmed Mean 3.30 The sole purpose of making money is to spend it Mean 5% Trimmed Mean 3.31 Sometimes I feel like spending money on anything I lay my Mean eyes on 5% Trimmed Mean 3.32 I often make impulse purchases Mean 5% Trimmed Mean 3.33 I always try something new and unique Mean 5% Trimmed Mean 3.34 I love fashionable and trendy products Mean 5% Trimmed Mean 3.35 It does not hurt to be trendy if I feel like it Mean 5% Trimmed Mean 3.36 I am often influenced by advertisements of new products Mean 4.03 4.07 3.56 3.58 3.44 3.46 4.17 4.24 3.84 3.90 3.97 4.01 3.55 3.57 3.18 3.16 3.93 3.98 3.48 3.50 3.02 3.05 2.91 2.93 2.71 2.70 3.11 3.12 3.20 3.22 3.54 3.57 3.21 3.23 3.03 3.06 3.02 303 5% Trimmed Mean 3.37 I can easily influence people around me during Mean conversation 5% Trimmed Mean 3.38 My friends often consult me when they cannot make up Mean their own mind 5% Trimmed Mean 3.39 I have a strong desire to be successful Mean 5% Trimmed Mean 3.40 Chinese people should buy domestic products rather than Mean imported products 5% Trimmed Mean 3.41 China should levy heavy tariff on foreign products to Mean reduce their quantity into China 5% Trimmed Mean 3.42 If two organic food products were the same in quality, but Mean one was imported and the other was Chinese, I would pay 5% Trimmed Mean 3.43 I would purchase organic food come from a country I Mean approve of politically 5% Trimmed Mean 3.44 Chinese people should not buy foreign products, because it Mean would hurt domestic business, and causes more unemployment 5% Trimmed Mean 4.1 I am eager to check out organic food products because of Mean advertisements and promotion 5% Trimmed Mean 4.2 I intend to try out organic food products Mean 5% Trimmed Mean 4.3 I am interested in experiencing the benefits of using organic Mean foods 5% Trimmed Mean 4.4 It is likely that I will buy organic food products when they Mean become available 5% Trimmed Mean 4.5 I can recall the brand names and labelling of some of the Mean organic food products 5% Trimmed Mean 4.6 I will probably use organic food products in the future Mean 5% Trimmed Mean 4.7 I will recommend usage of organic foods to my friends and Mean relatives 5% Trimmed Mean 4.8 If I had to purchase organic foods again, I would make the Mean same choice 5% Trimmed Mean 3.04 3.23 3.26 3.47 3.50 4.01 4.06 3.02 3.03 2.87 2.88 2.87 2.85 2.99 2.99 2.67 2.69 2.67 2.69 3.84 3.86 3.98 4.01 3.93 3.96 3.23 3.25 3.97 3.99 3.75 3.79 3.69 3.72 304 Table B: Results of the EFA for the product-related construct KMO and Bartlett’s Test Kaiser-Meyer-Olkin measure of sampling adequacy. Bartlett’s Test of Sphericity .696 Approx. chi-square 298.716 df 21 Sig. .000 Total variance explained Initial Eigenvalues Extraction sums of squared loadings Rotation sums of squared loadingsa Comp -onent Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 2.628 37.545 37.545 2.628 37.545 37.545 2.300 2 1.226 17.508 55.053 1.226 17.508 55.053 1.958 3 .942 13.456 68.508 4 .790 11.286 79.794 5 .620 8.852 88.647 6 .480 6.862 95.509 7 .314 4.491 100.000 Extraction Method: Principal Component Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance. 305 Pattern matrixa Component 1 2 3.2 Organic food looks nice .882 -.115 3.1 Organic food has a pleasant texture .865 .043 3.3 Organic food smells nice .694 .119 3.9 Organic food is good value for money -.119 .779 3.5 Consuming organic food is trendy .149 .648 3.6 Organic food has high nutritional value .179 .643 3.7 Generally speaking, the higher the price of a product, the higher the quality -.051 .485 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. a. Rotation converged in 5 iterations. Structure matrix Component 1 2 3.1 Organic food has a pleasant texture .878 .302 3.2 Organic food looks nice .848 .148 3.3 Organic food smells nice .730 .326 3.9 Organic food is good value for money .113 .743 3.6 Organic food has high nutritional value .371 .697 3.5 Consuming organic food is trendy .342 .693 3.7 Generally speaking, the higher the price of a product, the higher the quality .094 .470 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. Component correlation matrix Component 1 2 1 1.000 .298 2 .298 1.000 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. 306 Table C: EFA results for the regulatory construct KMO and Bartlett’s test Kaiser-Meyer-Olkin measure of sampling adequacy. Bartlett’s Test of Sphericity .771 Approx. chi-square 225.354 df 15 Sig. .000 Total variance explained Initial Eigenvalues Extraction sums of squared loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % 1 2.558 42.627 42.627 2.558 42.627 42.627 2 .963 16.056 58.683 3 .798 13.308 71.990 4 .726 12.094 84.084 5 .505 8.416 92.499 6 .450 7.501 100.000 Extraction Method: Principal Component Analysis. 307 Component matrixa Component 1 3.10 When I buy a food product, I always read the label .567 3.11 I don’t trust organic food certification bodies .401 3.14 Lack of adequate government control of media allows advertisers to take advantage of consumers .637 3.15 Logos depicting types of organic foods should be controlled .783 3.16 More land should be allocated for organic farming .702 3.17 The market for organic and green is chaotic .750 Extraction Method: Principal Component Analysis. a. 1 components extracted. 308 Table D: EFA results for the lifestyle construct KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett’s Test of Sphericity .748 Approx. Chi-Square 660.854 df 55 Sig. .000 Total Variance Explained Extraction Sums of Squared Loadings Initial Eigenvalues Rotation Sums of Squared Loadingsa Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total 1 3.624 32.944 32.944 3.624 32.944 32.944 3.090 2 1.553 14.118 47.062 1.553 14.118 47.062 2.562 3 1.330 12.087 59.149 1.330 12.087 59.149 1.929 4 .924 8.399 67.548 5 .869 7.900 75.448 6 .599 5.441 80.890 7 .580 5.276 86.165 8 .520 4.730 90.895 9 .458 4.168 95.063 10 .314 2.851 97.914 11 .229 2.086 100.000 Extraction Method: Principal Component Analysis. a. When components are correlated, sums of squared loadings cannot be added to obtain a total variance. 309 Pattern Matrixa Component 1 2 3 3.34 I love fashionable and trendy products .812 .071 .078 3.32 I often make impulse purchases .736 .106 .102 Lifestyle_9 3.36 I am often influenced by advertisements of new products .682 .013 .006 3.33 I always try something new and unique .679 .101 .149 3.35 It does not hurt to be trendy if I feel like it .673 .108 .063 3.28 When I like something, I will buy it without too much deliberation .032 .916 .004 3.29 I always do whatever I feel like and whenever I feel like it .068 .898 .099 3.30 The sole purpose of making money is to spend it .187 .672 .025 3.38 My friends often consult me when they cannot make up their own mind .010 .066 .829 3.37 I can easily influence people around me during conversation .077 -.027 .734 3.39 I have a strong desire to be successful .036 .088 .590 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. a. Rotation converged in 5 iterations. Structure Matrix Component 1 2 3 3.34 I love fashionable and trendy products .809 .217 .285 3.32 I often make impulse purchases .745 .340 .112 3.35 It does not hurt to be trendy if I feel like it .693 .328 .134 3.36 I am often influenced by advertisements of new products .685 .245 .179 3.33 I always try something new and unique .684 .153 .315 3.28 When I like something, I will buy it without too much deliberation .281 .906 .137 3.29 I always do whatever I feel like and whenever I feel like it .264 .890 .219 3.30 The sole purpose of making money is to spend it .409 .732 .129 3.38 My friends often consult me when they cannot make up their own mind .210 .065 .821 3.37 I can easily influence people around me during conversation .282 .166 .759 3.39 I have a strong desire to be successful .152 .167 .594 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. 310 Component Correlation Matrix Component 1 2 3 1 1.000 .340 .268 2 .340 1.000 .155 3 .268 .155 1.000 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. Table E: EFA results for the Ethnocentrism construct KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .689 Bartlett’s Test of Sphericity Approx. Chi-Square 167.247 df 3 Sig. .000 Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % 1 2.073 69.115 69.115 2.073 69.115 69.115 2 .531 17.708 86.822 3 .395 13.178 100.000 Component Extraction Method: Principal Component Analysis. 311 Component Matrixa Component 1 3.40 Chinese people should buy domestic products rather than imported products .864 3.41 China should levy heavy tariff on foreign products to reduce their quantity into China .821 3.44 Chinese people should not buy foreign products, because it would hurt domestic business, and causes more unemployment .808 Extraction Method: Principal Component Analysis. a. 1 components extracted. Table F: EFA results for the beliefs and attitudes construct KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett’s Test of Sphericity Approx. Chi-Square .900 830.123 df 45 Sig. .000 Total Variance Explained Initial Eigenvalues Component Total Extraction Sums of Squared Loadings % of Variance Cumulative % 1 4.898 48.985 48.985 2 .947 9.473 58.458 3 .856 8.560 67.018 4 .668 6.677 73.695 5 .622 6.225 79.920 6 .521 5.213 85.133 7 .469 4.694 89.828 8 .390 3.900 93.728 9 .342 3.416 97.143 10 .286 2.857 100.000 Total 4.898 % of Variance Cumulative % 48.985 48.985 Extraction Method: Principal Component Analysis. 312 Component Matrixa Component 1 3.24 I believe organic food has superior quality .801 3.23 Organic food is good for the environment .778 3.18 Organic food is good for myself and my family’s health .736 3.27 Organic food labels mean high quality food products .729 3.25 I personally think I should always buy organic food .711 3.20 I like the brands associated with organic food .703 3.19 Organic food has no harmful effects .699 3.22 Organic food does not contain pesticides .686 3.21 I worry about harmful chemicals in my food .627 3.26 To me, it is important that the food I usually eat can be easily found in the food outlets near my house or workplace .476 Extraction Method: Principal Component Analysis. a. 1 components extracted. Table G: EFA Results for the Pre-purchase evaluation construct KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett’s Test of Sphericity Approx. Chi-Square df Sig. .711 287.968 10 .000 313 Total Variance Explained Compo nent Initial Eigenvalues Total % of Variance Extraction Sums of Squared Loadings Cumulative % 1 2.512 50.235 50.235 2 .964 19.283 69.518 3 .792 15.844 85.362 4 .466 9.315 94.677 5 .266 5.323 100.000 Total % of Variance 2.512 Cumulative % 50.235 50.235 Extraction Method: Principal Component Analysis. Component Matrixa Component 1 4.3 I am interested in experiencing the benefits of using organic foods .851 4.2 I intend to try out organic food products .839 4.4 It is likely that I will buy organic food products when they become available .803 4.5 I can recall the brand names and labelling of some of the organic food products .485 4.1 I am eager to check out organic food products because of advertisements and promotion .452 Extraction Method: Principal Component Analysis. a. 1 components extracted. 314 Table H: EFA results for the Behavioural/purchase intention construct KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett’s Test of Sphericity .704 Approx. Chi-Square 234.229 df 3 Sig. .000 Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Component Total % of Variance Cumulative % 1 2.233 74.431 74.431 2 .469 15.635 90.066 3 .298 9.934 100.000 Total % of Variance Cumulative % 2.233 74.431 74.431 Extraction Method: Principal Component Analysis. 315 Component Matrixa Component 1 4.7 I will recommend usage of organic foods to my friends and relatives .890 4.8 If I had to purchase organic foods again, I would make the same choice .876 4.6 I will probably use organic food products in the future .821 Extraction Method: Principal Component Analysis. a. 1 components extracted. 316 Appendix 5: Detailed analysis of the main study (study 2) Table A: CFA for the product-related construct Discriminant validity for product _related construct F2 F1 Product_5 Product_6 Product_7 Product_9 Product_1 Product_2 F2 1.000 F1 .833 1.000 Product_5 .579 .482 1.000 Product_6 .528 .439 .305 1.000 Product_7 .235 .196 .136 .124 1.000 Product_9 .457 .380 .264 .241 .108 1.000 Product_1 .524 .629 .303 .276 .123 .239 1.000 Product_2 .535 .643 .310 .282 .126 .244 .405 1.000 Product_3 .555 .666 .321 .293 .131 .253 .419 .428 Product_3 1.000 It suggested that there is lack of discriminant validity. Although product _7 shows significant difference, one item can’t form a construct. CFA for product-related measure (after deleted statement 3.7/3.9)-one factor Sample Covariances (Group number 1) Product_6 Product_5 Product_3 Product_2 Product_1 Product_6 .688 .179 .189 .159 .172 Product_5 Product_3 Product_2 Product_1 .656 .230 .193 .193 .538 .228 .215 .549 .255 .581 Condition number = 4.730 Eigenvalues 1.409 .521 .454 .331 .298 317 Standardised regression weights (Group number 1-Default model) Product_1 Product_2 Product_3 Product_5 Product_6 <--<--<--<--<--- PRODUCT_-RELATED PRODUCT_-RELATED PRODUCT_-RELATED PRODUCT_-RELATED PRODUCT_-RELATED Estimate .625 .650 .658 .537 .445 Implied correlations (Group number 1-Default model) Product_6 Product_5 Product_3 Product_2 Product_1 Product_6 1.000 .239 .293 .290 .278 Product_5 Product_3 Product_2 Product_1 1.000 .353 .349 .335 1.000 .428 .411 1.000 .406 1.000 Standardised residual covariances (Group number 1-Default model) Product_6 Product_5 Product_3 Product_2 Product_1 Product_6 .000 .801 .538 -.928 -.173 Product_5 Product_3 Product_2 Product_1 .000 .994 -.810 -.661 .000 -.208 -.768 .000 1.306 .000 CMIN Model Default model Saturated model Independence model NPAR 10 15 5 CMIN 16.690 .000 803.505 DF 5 0 10 P .005 CMIN/DF 3.338 .000 80.350 RMR, GFI Model Default model Saturated model Independence model RMR .013 .000 .166 GFI .993 1.000 .678 AGFI .978 PGFI .331 .517 .452 318 Baseline Comparisons Model Default model Saturated model Independence model NFI Delta1 .979 1.000 .000 RFI rho1 .958 .000 IFI Delta2 .985 1.000 .000 TLI rho2 .971 .000 CFI .985 1.000 .000 RMSEA Model Default model Independence model RMSEA .050 .291 LO 90 .025 .274 HI 90 .077 .308 PCLOSE .450 .000 Standardised RMR = .0224 Table B: CFA for the regulatory construct Sample Correlations (Group number 1) Regulatory_1 Regulatory_5 Regulatory_6 Regulatory_7 Regulatory_1 1.000 Regulatory_5 .116 1.000 Regulatory_6 .244 .338 1.000 Regulatory_7 .106 .213 .265 1.000 Regulatory_8 .171 .264 .353 .264 Regulatory_8 1.000 Condition number = 3.225 Eigenvalues 1.964 .921 .796 .711 .609 319 Standardised Regression Weights: (Group number 1-Default model) Regulatory_8 Regulatory_7 Regulatory_6 Regulatory_5 Regulatory_1 <--<--<--<--<--- REGULATION REGULATION REGULATION REGULATION REGULATION Estimate .545 .421 .669 .490 .316 CMIN Model Default model Saturated model Independence model NPAR 10 15 5 CMIN 8.065 .000 442.258 DF 5 0 10 P .153 CMIN/DF 1.613 .000 44.226 RMR, GFI Model Default model Saturated model Independence model RMR .013 .000 .144 GFI .997 1.000 .805 AGFI .990 PGFI .332 .707 .536 NFI Delta1 .982 1.000 .000 RFI rho1 .964 IFI Delta2 .993 1.000 .000 TLI rho2 .986 Baseline Comparisons Model Default model Saturated model Independence model .000 .000 CFI .993 1.000 .000 RMSEA Model Default model Independence model RMSEA .026 .215 LO 90 .000 .198 HI 90 .057 .232 PCLOSE .891 .000 Standardised RMR = .0183 320 Table C: CFA for the Lifestyle measures Standardised Regression Weights: (Group number 1-Default model) Estimate Lifestyle_9 <--- VARIETY-_SEEKING .537 Lifestyle_7 <--- VARIETY-_SEEKING .792 Lifestyle_6 <--- VARIETY-_SEEKING .791 Lifestyle_5 <--- VARIETY-_SEEKING .669 Lifestyle_3 <--- SELF-_INDULGENCE .607 Lifestyle_2 <--- SELF-_INDULGENCE .825 Lifestyle_1 <--- SELF-_INDULGENCE .786 Lifestyle_11 <--- OPINION-_LEADERSHIP .665 Lifestyle_10 <--- OPINION-_LEADERSHIP .728 Lifestyle_12 <--- OPINION-_LEADERSHIP .454 Standardised Residual Covariances (Group number 1-Default model) Lifestyle _12 Lifestyle _12 Lifestyle _10 Lifestyle _11 Lifestyle _1 Lifestyle _2 Lifestyle _3 Lifestyle _5 Lifestyle _6 Lifestyle _7 Lifestyle _9 Lifestyle _10 Lifestyle _11 Lifestyl e_1 Lifestyl e_2 Lifestyl e_3 Lifestyl e_5 Lifestyl e_6 Lifestyl e_7 Lifestyl e_9 .000 -.503 .000 .366 .102 .000 -2.006 .409 -.254 .000 -1.263 .021 -.247 .255 .000 .100 2.836 -.063 -.281 -.360 .000 .221 -.186 -1.617 .061 1.341 4.382 .000 .974 -1.299 -1.012 -1.555 -1.142 .505 .058 .000 1.071 .516 -.376 -.731 -.771 2.563 -.764 .722 .000 1.343 2.978 2.430 .750 .956 2.959 1.567 -.894 -.996 .000 CMIN Model Default model Saturated model Independence model NPAR 23 55 10 CMIN 125.621 .000 2768.959 DF 32 0 45 P .000 CMIN/DF 3.926 .000 61.532 321 RMR, GFI Model Default model Saturated model Independence model RMR .037 .000 .245 GFI .973 1.000 .519 AGFI .954 PGFI .566 .412 .424 Default model Saturated model Independence model NFI Delta1 .955 1.000 .000 RFI rho1 .936 IFI Delta2 .966 1.000 .000 TLI rho2 .952 RMSEA Model Default model Independence model RMSEA .056 .255 Baseline Comparisons Model .000 LO 90 .046 .247 HI 90 .067 .263 .000 CFI .966 1.000 .000 PCLOSE .157 .000 Standardised RMR = .0424 Table D: CFA for Ethnocentrism measures Sample Correlations (Group number 1) Ethnocentric_1 Ethnocentric_2 Ethnocentric_5 Ethnocentric_1 1.000 Ethnocentric_2 .618 1.000 Ethnocentric_5 .577 .562 1.000 Condition number = 5.702 Eigenvalues 2.172 .447 .381 Sample moments: Eigenvalues 2.172, .447, .381, it suggests one factor solution Standardised Regression Weights: (Group number 1-Default model) Ethnocentric_5 <--- ETHNOCENTRISM Ethnocentric_2 <--- ETHNOCENTRISM Ethnocentric_1 <--- ETHNOCENTRISM Estimate .724 .771 .801 322 CMIN Model Default model Saturated model Independence model NPAR 5 6 3 CMIN .132 .000 925.508 DF 1 0 3 P .717 CMIN/DF .132 .000 308.503 RMR, GFI Model Default model Saturated model Independence model RMR .004 .000 .389 GFI 1.000 1.000 .593 AGFI .999 PGFI .167 .185 .296 Baseline Comparisons Model Default model Saturated model Independence model NFI Delta1 1.000 1.000 .000 RFI rho1 1.000 .000 IFI Delta2 1.001 1.000 .000 TLI rho2 1.000 .000 CFI 1.000 1.000 .000 RMSEA Model Default model Independence model RMSEA .000 .575 LO 90 .000 .544 HI 90 .062 .606 PCLOSE .907 .000 Standardised RMR = .0020 Table E: CFA for the beliefs and attitudes measures Sample Correlations (Group number 1) Personal_10 Personal_8 Personal_7 Personal_10 1.000 Personal_8 .390 1.000 Personal_7 .422 .482 1.000 Personal_6 .286 .366 .474 Personal_3 .275 .394 .361 Condition number = 5.268 Personal_6 Personal_3 1.000 .243 1.000 323 Eigenvalues 2.493 .777 .707 .550 .473 Sample moments: Eigenvalues, it suggests one factor solution Standardised Regression Weights: (Group number 1-Default model) Estimate Personal_3 <--- PERSONAL .502 Personal_6 <--- PERSONAL .572 Personal_7 <--- PERSONAL .753 Personal_8 <--- PERSONAL .670 Personal_10 <--- PERSONAL .555 CMIN Model Default model Saturated model Independence model NPAR 10 15 5 RMR, GFI Model Default model Saturated model Independence model RMR .016 .000 .201 GFI .990 1.000 .637 AGFI .970 PGFI .330 .456 .425 Default model Saturated model Independence model NFI Delta1 .976 1.000 .000 RFI rho1 .953 IFI Delta2 .981 1.000 .000 TLI rho2 .963 RMSEA Model Default model Independence model RMSEA .062 .320 CMIN 22.781 .000 960.071 DF 5 0 10 P .000 CMIN/DF 4.556 .000 96.007 Baseline Comparisons Model .000 LO 90 .038 .303 HI 90 .089 .337 .000 CFI .981 1.000 .000 PCLOSE .193 .000 Standardised RMR = .0253 324 Table F: CFA for Pre-purchase evaluation measures Sample Correlations (Group number 1) Prepurchase_4 Prepurchase_3 Prepurchase_2 Prepurchase_1 Prepurchase_4 1.000 Prepurchase_3 .523 1.000 Prepurchase_2 .482 .496 1.000 Prepurchase_1 .202 .231 .306 1.000 Condition number = 4.540 Eigenvalues 2.158 .860 .506 .475 Sample moments: Eigenvalues 2.158, 0.860, 0.506, 0.475, it suggests one factor solution Standardised Regression Weights: (Group number 1-Default model) Prepurchase_1 Prepurchase_2 Prepurchase_3 Prepurchase_4 <--<--<--<--- PRE-PURCHASE_EVALUATION PRE-PURCHASE_EVALUATION PRE-PURCHASE_EVALUATION PRE-PURCHASE_EVALUATION Model Default model Saturated model Independence model Estimate .346 .699 .725 .699 CMIN NPAR CMIN 8 13.953 10 .000 4 749.317 DF 2 0 6 RMR, GFI Model RMR GFI Default model .014 .992 Saturated model .000 1.000 Independence model .145 .680 P .001 CMIN/DF 6.977 .000 124.886 AGFI .962 PGFI .198 .467 .408 Baseline Comparisons NFI RFI IFI Model Delta1 rho1 Delta2 Default model .981 .944 .984 Saturated model 1.000 1.000 Independence model .000 .000 .000 TLI rho2 .952 .000 CFI .984 1.000 .000 325 Model Default model Independence model RMSEA RMSEA LO 90 .044 .080 .365 .343 HI 90 .122 .387 PCLOSE .081 .000 Standardised RMR = .0255 Table G: CFA for Behavioural/purchase intention measures Sample Correlations (Group number 1) Be_intention_1 1.000 .437 .427 Be_intention_2 Be_intention_3 Be_intention_1 Be_intention_2 1.000 Be_intention_3 .416 1.000 Condition number = 3.304 Eigenvalues 1.854 .585 .561 Sample moments: Eigenvalues 1.854, 0.585, 0.561, it suggests one factor solution Standardised Regression Weights: (Group number 1-Default model) Estimate Be_intention_3 <--- PURCHASE_INTENTION .650 Be_intention_2 <--- PURCHASE_INTENTION .640 Be_intention_1 <--- PURCHASE_INTENTION .670 Model Default model Saturated model Independence model NPAR 5 6 3 CMIN CMIN .343 .000 461.587 DF 1 0 3 RMR, GFI Model RMR GFI Default model .004 1.000 Saturated model .000 1.000 Independence model .161 .733 P .558 CMIN/DF .343 .000 153.862 AGFI .999 PGFI .167 .466 .366 326 RMSEA Model Default model Independence model RMSEA LO 90 HI 90 PCLOSE .000 .405 .000 .375 .072 .437 .844 .000 Baseline Comparisons NFI RFI IFI Model Delta1 rho1 Delta2 Default model .999 .998 1.001 Saturated model 1.000 1.000 Independence model .000 .000 .000 TLI rho2 1.004 .000 CFI 1.000 1.000 .000 Standardised RMR = .0049 327 Appendix 6: Discriminant validity Table A: Discriminant validity for section one ETHNOCENTRISM SELF_INDULGENCE VARIETY_SEEKING ETHNOCENTRISM 1.000 SELF_INDULGENCE .248 1.000 .177 .463 1.000 VARIETY_SEEKING REGULATORY REGULATORY PRODUCT_RELATED .046 .044 .127 1.000 PRODUCT_RELATED .126 .176 .231 .272 1.000 Ethnocentric_1 Ethnocentric_2 Ethnocentric_5 Lifestyle_1 Lifestyle_2 Lifestyle_3 Lifestyle_6 Lifestyle_7 Lifestyle_9 Regulatory_5 Regulatory_6 Regulatory_7 Regulatory_8 Product_1 Product_2 Product_3 Product_5 Product_6 .792 .782 .725 .195 .205 .149 .140 .146 .088 .024 .029 .020 .024 .078 .083 .081 .067 .056 .196 .194 .180 .788 .826 .602 .365 .381 .230 .023 .028 .020 .023 .108 .116 .113 .094 .078 .140 .139 .128 .365 .383 .279 .788 .823 .497 .066 .080 .057 .068 .142 .152 .148 .123 .103 .036 .036 .033 .034 .036 .026 .100 .104 .063 .522 .635 .449 .534 .168 .179 .174 .145 .121 .100 .099 .092 .139 .145 .106 .182 .190 .115 .142 .173 .122 .146 .616 .659 .641 .532 .444 Table B: Discriminant validity for section two PRE-PURCHASE_EVALUATION PURCHASE_INTENTION PERSONAL_ATTITUDES Prepurchase_2 Prepurchase_3 Prepurchase_4 Be_intention_1 Be_intention_2 Be_intention_3 Personal_6 Personal_7 Personal_8 Personal_10 PRE-PURCHASE _EVALUATION 1.000 .771 .484 .652 .696 .767 .515 .503 .493 .293 .368 .307 .268 PURCHASE _INTENTION PERSONAL _ATTITUDES 1.000 .573 .503 .537 .591 .668 .652 .640 .347 .435 .363 .317 1.000 .316 .337 .371 .383 .374 .366 .606 .760 .634 .554 328 .000 .288 .000 .239 .062 .127 .321 .000 .000 .038 .037 .016 LIFESTYLE -.069 .197 .383 .000 .000 .282 PRODUCT .067 .379 .274 .000 .000 .000 REGULATORY LIFESTYLE PRODUCT REGULATORY BELIEFS/ATTITUDES PRE_-PURCHASE BEHAVIOUR_INTENTION .000 .000 .081 .133 .161 .210 .145 LIFESTYLE .000 .093 .026 .119 .079 ETHNOCENTRISM .371 .000 .000 .000 .077 .197 PRODUCT .332 .000 .000 .000 .000 .054 REGULATORY .122 BELIEFS/ ATTITUDES .000 .000 .000 .000 .000 .000 PRE_PURCHASE .000 .000 .000 .000 .000 .000 BEHAVIOUR_INTENTION .000 .000 .000 .000 .000 .000 .000 .232 .621 .000 .000 .000 .196 .000 SELF_INDULGENCE .000 .000 .000 .000 .000 .000 .000 .000 SELF_INDULGENCE .000 .000 .000 .000 BEHAVIOUR_INTENTION .000 .000 .000 PRE_PURCHASE .000 .000 .000 BELIEFS/ ATTITUDES .000 .000 .000 Table B: Standardised Indirect Effects (Group number 1-Default model) LIFESTYLE PRODUCT REGULATORY BELIEFS/ ATTITUDES PRE_PURCHASE BEHAVIOUR_INTENTION ETHNOCENTRISM Table A: Standardised Direct Effects (Group number 1-Default model) Appendix 7: The output of the final model .000 329 VARIETY_SEEKING .000 .000 .000 .000 .000 .000 .000 .000 VARIETY_SEEKING .000 .000 .000 LIFESTYLE PRODUCT REGULATORY ATT/BEL PRE_-PURCHASE BEHAVIOUR_INTENTION SELF_INDULGENCE VARIETY_SEEKING .000 .288 .081 .372 .222 .337 .766 .601 .161 .246 .193 LIFESTYLE .321 .093 .026 .158 .116 ETHNOCENTRISM .000 .000 .301 .000 .000 .282 .461 .395 PRODUCT .000 .000 .399 .000 .000 .000 .274 .432 REGULATORY .000 .000 .354 .000 .000 .000 .000 .196 ATT/BEL Table C: Standardised Total Effects (Group number 1-Default model) .000 .000 .000 .000 .621 .000 BEHAVIOUR_INTENTION .000 .000 .000 .000 .000 PRE_PURCHASE .000 .000 .000 .000 .000 .000 .000 .000 SELF_INDULGENCE .000 .000 .000 .000 .000 .000 .000 .000 330 VARIETY_SEEKING .000 .000 .000 .000 .000 Table D: Regression Weights: (Group number 1-Default model) Estimate S.E. C.R. P LIFESTYLE <--- ETHNOCENTRISM .121 .022 5.386 *** PRODUCT <--- LIFESTYLE .415 .087 4.788 *** REGULATORY <--- PRODUCT .308 .058 5.354 *** ATTITUDES <--- PRODUCT .450 .066 6.820 *** ATTITUDES <--- REGULATORY .294 .053 5.567 *** ATTITUDES <--- LIFESTYLE .404 .098 4.111 *** ATTITUDES <--- ETHNOCENTRISM .024 .026 .926 .355 PRE_-PURCHASE <--- ATTITUDES .223 .068 3.257 .001 PRE_-PURCHASE <--- REGULATORY .464 .066 7.066 *** PRE_-PURCHASE <--- LIFESTYLE .119 .105 1.131 .258 PRE_-PURCHASE <--- PRODUCT .264 .071 3.695 *** PRE_-PURCHASE <--- ETHNOCENTRISM .027 .029 .922 .356 VARIETY_-SEEKING <--- LIFESTYLE 1.000 SELF_-INDULGENCE <--- LIFESTYLE 1.692 .262 6.448 *** BEHAVIOUR_-INTENTION <--- PRE_-PURCHASE .566 .058 9.729 *** BEHAVIOUR_-INTENTION <--- ATTITUDES .241 .061 3.947 *** BEHAVIOUR_-INTENTION <--- LIFESTYLE .222 .095 2.347 .019 BEHAVIOUR_-INTENTION <--- PRODUCT -.085 .062 -1.363 .173 BEHAVIOUR_-INTENTION <--- REGULATORY .075 .061 1.234 .217 BEHAVIOUR_-INTENTION <--- ETHNOCENTRISM .011 .026 .420 .674 331 Table E: Factor loadings in the final best-fit model Standardised Regression Weights: (Group number 1-Default model) Estimate Product_6 <--- PRODUCT .481 Product_5 <--- PRODUCT .542 Product_3 <--- PRODUCT .645 Product_2 <--- PRODUCT .622 Product_1 <--- PRODUCT .613 Regulatory_8 <--- REGULATORY .529 Regulatory_7 <--- REGULATORY .494 Regulatory_6 <--- REGULATORY .632 Regulatory_5 <--- REGULATORY .485 Lifestyle_9 <--- VARIETY_-SEEKING .497 Lifestyle_7 <--- VARIETY_-SEEKING .823 Lifestyle_6 <--- VARIETY_-SEEKING .789 Lifestyle_3 <--- SELF_-INDULGENCE .601 Lifestyle_2 <--- SELF_-INDULGENCE .825 Lifestyle_1 <--- SELF_-INDULGENCE .790 Ethnocentric_5 <--- ETHNOCENTRISM .724 Ethnocentric_2 <--- ETHNOCENTRISM .784 Ethnocentric_1 <--- ETHNOCENTRISM .790 Personal_6 <--- ATTITUDES .594 Personal_7 <--- ATTITUDES .742 Personal_8 <--- ATTITUDES .651 Personal_10 <--- ATTITUDES .571 Prepurchase_2 <--- PRE_-PURCHASE .652 Prepurchase_3 <--- PRE_-PURCHASE .700 Prepurchase_4 <--- PRE_-PURCHASE .763 Be_intention_3 <--- BEHAVIOUR_-INTENTION .643 Be_intention_2 <--- BEHAVIOUR_-INTENTION .650 Be_intention_1 <--- BEHAVIOUR_-INTENTION .666 332 Table F: Squared Multiple Correlations: (Group number 1-Default model) ETHNOCENTRISM LIFESTYLE PRODUCT REGULATORY BELIEFS/ATTITUDES PRE_-PURCHASE BEHAVIOUR_-INTENTION SELF_-INDULGENCE VARIETY_-SEEKING Prepurchase_4 Be_intention_1 Be_intention_2 Be_intention_3 Prepurchase_3 Prepurchase_2 Personal_10 Personal_8 Personal_7 Personal_6 Ethnocentric_1 Ethnocentric_2 Ethnocentric_5 Lifestyle_1 Lifestyle_2 Lifestyle_3 Lifestyle_6 Lifestyle_7 Lifestyle_9 Regulatory_5 Regulatory_6 Regulatory_7 Regulatory_8 Product_1 Product_2 Product_3 Product_5 Product_6 Estimate .000 .103 .083 .080 .413 .395 .664 .587 .361 .583 .443 .423 .413 .490 .425 .326 .423 .550 .352 .625 .614 .525 .624 .680 .361 .622 .677 .247 .235 .400 .244 .279 .376 .387 .416 .293 .231 333 dimension3 dimension2 dimension3 dimension3 dimension3 dimension3 Total_purchase_intention Tukey HSD (I) 5.11 Please specify the (J) 5.11 Please specify the highest highest level of education you level of education you have have completed: (tick one) completed: (tick one) 1 2 3 4 5 2 1 3 4 5 3 1 2 4 5 4 1 2 3 5 5 1 2 3 4 *. The mean difference is significant at the 0.05 level. Hypothesis H7C Table A: Multiple Comparisons Appendix 8: Hypotheses seven Mean Difference (I-J) -.208* -.371* -.258* -.292* .208* -.164 -.050 -.084 .371* .164 .113 .079 .258* .050 -.113 -.034 .292* .084 -.079 .034 Std. Error .074 .071 .066 .083 .074 .060 .054 .074 .071 .060 .050 .071 .066 .054 .050 .066 .083 .074 .071 .066 Sig. .040 .000 .001 .004 .040 .051 .881 .785 .000 .051 .153 .799 .001 .881 .153 .986 .004 .785 .799 .986 Lower Bound -.41 -.56 -.44 -.52 .01 -.33 -.20 -.29 .18 .00 -.02 -.12 .08 -.10 -.25 -.21 .06 -.12 -.27 -.15 Upper Bound -.01 -.18 -.08 -.06 .41 .00 .10 .12 .56 .33 .25 .27 .44 .20 .02 .15 .52 .29 .12 .21 95% Confidence Interval 334 N 94 161 202 368 93 918 Mean 3.47 3.68 3.84 3.73 3.76 3.72 dimension2 dimension3 dimension3 dimension3 Total_purchase_intention Tukey HSD (I) 5.7 Please indicate your (J) 5.7 Please indicate your monthly family income? monthly family income? (tick one) (tick one) 1 2 3 4 2 1 3 4 3 1 2 4 4 1 2 3 *. The mean difference is significant at the 0.05 level. dimension3 Std. Deviation .575 .533 .591 .568 .564 .575 Table C: Multiple Comparisons Hypothesis H7d 1 2 3 4 5 Total Total_purchase_intention Table B: Descriptives Mean Difference (I-J) -.182* -.104 -.206* .182* .078 -.025 .104 -.078 -.103 .206* .025 .103 Std. Error .059 .042 .042 .030 .058 .019 Std. Error .043 .055 .076 .043 .057 .077 .055 .057 .085 .076 .077 .085 Sig. .000 .238 .034 .000 .521 .989 .238 .521 .618 .034 .989 .618 2 2 2 1 2 1 Maximum 5 5 5 5 5 5 Upper Bound -.07 .04 -.01 .29 .23 .17 .25 .07 .11 .40 .22 .32 95% Confidence Interval Minimum Lower Bound -.29 -.25 -.40 .07 -.07 -.22 -.04 -.23 -.32 .01 -.17 -.11 95% Confidence Interval for Mean Lower Bound Upper Bound 3.35 3.59 3.60 3.76 3.76 3.92 3.67 3.79 3.65 3.88 3.69 3.76 335 1 2 3 4 Total N 394 310 147 66 917 Total_purchase_intention Mean 3.63 3.81 3.74 3.84 3.73 Table D: Descriptives Std. Deviation .545 .572 .643 .540 .576 Std. Error .027 .033 .053 .066 .019 95% Confidence Interval for Mean Lower Bound Upper Bound 3.58 3.69 3.75 3.88 3.63 3.84 3.71 3.97 3.69 3.76 Minimum 2 2 1 2 1 Maximum 5 5 5 5 5 336 Table E: Descriptive Statistics of important attributes (23 statements) N Minimum Maximum Std. Mean Deviation 2.19 Certification relating to the quality of organic food 948 1 5 4.24 .842 2.4 The overall quality of organic food 950 1 5 4.24 .829 2.20 Enforcement relating to the quality of organic food 951 1 5 4.22 .868 2.23 Information about the nutritional value of organic 958 food 1 5 4.15 .851 2.18 Food safety in relation to organic food 949 1 5 4.14 .890 2.22 Correct labelling of organic food 950 1 5 4.07 .832 2.9 The environmental benefits of organic food 950 1 5 3.93 .929 2.5 The price of organic food 947 1 5 3.90 .875 2.8 The value of organic foods relative to its price 946 1 5 3.88 .923 2.17 Government regulations and policies relating to the 947 sale of organic food 1 5 3.87 .961 2.1 The taste of organic food 960 1 5 3.79 .845 2.6 The availability of organic food in convenience stores 946 and supermarkets 1 5 3.76 .867 2.2 The smell of organic food 951 1 5 3.73 .847 2.12 The knowledge of organic food products 947 1 5 3.69 .924 2.13 The awareness of organic food products 951 1 5 3.57 .913 2.16 The brand name of the organic food 956 1 5 3.52 .930 2.21 Packaging of organic food 952 1 5 3.41 .973 2.15 Organic food that is produced in China 950 1 5 3.39 .978 2.3 The appearance of organic food 947 1 5 3.34 .947 2.14 The country of origin of the organic food 949 1 5 3.30 1.043 2.7 The promotion and advertising of organic food 940 1 5 3.27 .977 2.11 The social status of people purchasing organic food 950 1 5 2.35 1.103 2.10 The idea of face saving (mianzi) when purchasing 948 organic food 1 5 2.30 1.084 Valid N (listwise) 859 337 Own Consumption For Children For Elderly Gift For Family Total 254 90 27 11 328 710 35.8 12.7 3.8 1.5 46.2 100.0 Percent 35.8 12.7 3.8 1.5 46.2 100.0 Valid Percent 35.8 48.5 52.3 53.8 100.0 Cumulative Percent 2.1 The taste of organic Own Consumption food For Children For Elderly Gift For Family Total 2.2 The smell of organic Own Consumption food For Children For Elderly Gift For Family Total 2.3 The appearance of Own Consumption organic food For Children For Elderly N 254 89 27 11 327 708 253 90 27 11 322 703 254 90 27 Mean 3.83 3.82 3.63 3.45 3.83 3.82 3.86 3.58 3.67 3.64 3.71 3.75 3.25 3.40 3.48 Std. Deviation .888 .716 1.115 .820 .847 .858 .846 .793 .961 .674 .836 .840 .974 .909 1.051 Std. Error .056 .076 .214 .247 .047 .032 .053 .084 .185 .203 .047 .032 .061 .096 .202 95% Confidence Interval for Mean Lower Bound Upper Bound 3.72 3.94 3.67 3.97 3.19 4.07 2.90 4.01 3.74 3.92 3.75 3.88 3.76 3.97 3.41 3.74 3.29 4.05 3.18 4.09 3.62 3.80 3.68 3.81 3.13 3.37 3.21 3.59 3.07 3.90 Minimum 1 2 1 2 1 1 1 1 2 3 1 1 1 1 2 Table G: Descriptives results of single intended use and important attributes Valid Frequency Table F: Single usage group Maximum 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 338 2.8 The value of organic foods relative to its price 2.7 The promotion and advertising of organic food 2.6 The availability of organic food in convenience stores and supermarkets 2.5 The price of organic food 2.4 The overall quality of organic food Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total 11 323 705 250 89 27 11 326 703 250 90 27 11 324 702 250 89 27 10 324 700 252 89 27 11 320 699 249 90 27 11 325 702 3.36 3.41 3.36 4.32 4.09 4.37 4.27 4.24 4.25 3.88 4.04 3.81 3.82 3.93 3.92 3.73 3.81 4.00 3.80 3.79 3.78 3.20 3.27 3.37 3.64 3.27 3.25 3.94 3.84 3.59 3.73 3.88 3.88 .809 .966 .964 .773 .821 .629 .786 .821 .798 .872 .806 .962 .603 .877 .866 .904 .864 .784 .632 .870 .875 1.022 .939 .792 .809 .965 .974 .944 .898 .971 .905 .879 .909 .244 .054 .036 .049 .087 .121 .237 .045 .030 .055 .085 .185 .182 .049 .033 .057 .092 .151 .200 .048 .033 .064 .099 .152 .244 .054 .037 .060 .095 .187 .273 .049 .034 2.82 3.31 3.28 4.23 3.92 4.12 3.74 4.15 4.20 3.77 3.88 3.43 3.41 3.83 3.85 3.62 3.63 3.69 3.35 3.69 3.71 3.07 3.07 3.06 3.09 3.16 3.18 3.83 3.66 3.21 3.12 3.78 3.82 3.91 3.52 3.43 4.42 4.26 4.62 4.80 4.33 4.31 3.98 4.21 4.20 4.22 4.02 3.98 3.84 3.99 4.31 4.25 3.88 3.84 3.33 3.47 3.68 4.18 3.37 3.33 4.06 4.03 3.98 4.33 3.98 3.95 2 1 1 1 1 3 3 1 1 2 2 2 3 1 1 1 1 3 3 1 1 1 1 2 3 1 1 1 1 2 2 1 1 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 339 2.9 The environmental Own Consumption benefits of organic food For Children For Elderly Gift For Family Total 2.10 The idea of face saving Own Consumption (mianzi) when purchasing For Children organic food For Elderly Gift For Family Total 2.11 The social status of Own Consumption people purchasing organic For Children food For Elderly Gift For Family Total 2.12 The knowledge of Own Consumption organic food products For Children For Elderly Gift For Family Total 2.13 The awareness of Own Consumption organic food products For Children For Elderly Gift For Family Total 2.14 The country of origin Own Consumption of the organic food For Children For Elderly 254 88 27 11 326 706 253 90 27 11 324 705 253 89 27 11 327 707 252 89 27 11 326 705 253 90 27 11 326 707 253 90 27 4.01 3.69 3.96 3.91 3.94 3.93 2.16 2.44 2.19 2.64 2.30 2.27 2.14 2.58 2.48 3.18 2.30 2.30 3.66 3.71 3.70 3.73 3.74 3.71 3.38 3.81 3.85 4.09 3.63 3.58 3.17 3.36 3.30 .937 .951 .898 .944 .901 .923 1.094 1.040 1.111 1.120 1.082 1.085 1.078 1.085 1.189 1.079 1.080 1.096 .920 .829 1.171 1.009 .916 .917 .975 .748 .770 .701 .867 .902 1.042 .952 1.171 .059 .101 .173 .285 .050 .035 .069 .110 .214 .338 .060 .041 .068 .115 .229 .325 .060 .041 .058 .088 .225 .304 .051 .035 .061 .079 .148 .211 .048 .034 .066 .100 .225 3.89 3.49 3.61 3.27 3.84 3.86 2.02 2.23 1.75 1.88 2.18 2.19 2.01 2.36 2.01 2.46 2.18 2.22 3.55 3.53 3.24 3.05 3.64 3.64 3.26 3.65 3.55 3.62 3.53 3.51 3.04 3.16 2.83 4.12 3.89 4.32 4.54 4.03 4.00 2.29 2.66 2.62 3.39 2.42 2.35 2.28 2.81 2.95 3.91 2.41 2.38 3.78 3.88 4.17 4.41 3.84 3.77 3.50 3.97 4.16 4.56 3.72 3.64 3.30 3.55 3.76 1 1 3 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 1 1 1 2 2 3 1 1 1 1 1 5 5 5 5 5 5 5 5 5 4 5 5 5 5 5 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 340 2.19 Certification relating to the quality of organic food 2.18 Food safety in relation to organic food 2.17 Government regulations and policies relating to the sale of organic food 2.16 The brand name of the organic food 2.15 Organic food that is produced in China Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total Own Consumption For Children For Elderly Gift For Family Total 11 325 706 254 90 27 11 325 707 254 90 27 11 327 709 253 90 27 11 326 707 252 89 27 11 326 705 253 90 27 11 323 704 3.36 3.37 3.29 3.33 3.38 3.26 3.36 3.40 3.37 3.35 3.51 3.52 3.55 3.56 3.48 3.87 3.72 3.93 4.18 3.88 3.86 4.27 3.87 3.93 4.27 4.10 4.12 4.30 4.08 4.22 4.27 4.27 4.25 .924 1.048 1.038 .995 1.023 1.095 .924 .946 .977 .940 .838 .975 1.036 .944 .935 1.012 .936 .958 .982 .933 .964 .859 .919 .997 .786 .903 .899 .906 .738 .801 .905 .799 .834 .279 .058 .039 .062 .108 .211 .279 .052 .037 .059 .088 .188 .312 .052 .035 .064 .099 .184 .296 .052 .036 .054 .097 .192 .237 .050 .034 .057 .078 .154 .273 .044 .031 2.74 3.25 3.22 3.21 3.16 2.83 2.74 3.30 3.30 3.23 3.34 3.13 2.85 3.46 3.41 3.75 3.53 3.55 3.52 3.78 3.79 4.16 3.67 3.53 3.74 4.00 4.06 4.18 3.92 3.91 3.67 4.18 4.19 3.98 3.48 3.37 3.46 3.59 3.69 3.98 3.50 3.44 3.46 3.69 3.90 4.24 3.67 3.55 4.00 3.92 4.30 4.84 3.98 3.93 4.37 4.06 4.32 4.80 4.20 4.19 4.41 4.23 4.54 4.88 4.36 4.31 2 1 1 1 1 1 2 1 1 1 1 1 2 1 1 1 1 2 2 1 1 1 1 2 3 1 1 1 2 2 3 1 1 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 341 2.20 Enforcement relating Own Consumption to the quality of organic For Children food For Elderly Gift For Family Total 2.21 Packaging of organic Own Consumption food For Children For Elderly Gift For Family Total 2.22 Correct labelling of Own Consumption organic food For Children For Elderly Gift For Family Total 2.23 Information about the Own Consumption nutritional value of organic For Children food For Elderly Gift For Family Total 252 89 27 11 326 705 253 90 27 11 326 707 252 90 26 11 326 705 253 90 27 11 327 708 4.28 4.16 4.04 4.36 4.22 4.23 3.36 3.50 3.56 3.36 3.52 3.46 3.99 4.14 3.96 4.18 4.10 4.06 4.20 4.08 3.81 4.18 4.15 4.15 .877 .838 .940 .809 .861 .865 .977 1.030 1.121 1.027 .914 .962 .904 .801 .871 .751 .822 .851 .852 .851 1.388 .874 .816 .863 .055 .089 .181 .244 .048 .033 .061 .109 .216 .310 .051 .036 .057 .084 .171 .226 .046 .032 .054 .090 .267 .263 .045 .032 4.17 3.98 3.67 3.82 4.13 4.17 3.24 3.28 3.11 2.67 3.42 3.39 3.88 3.98 3.61 3.68 4.01 4.00 4.10 3.90 3.27 3.59 4.06 4.08 4.39 4.33 4.41 4.91 4.32 4.30 3.48 3.72 4.00 4.05 3.62 3.53 4.10 4.31 4.31 4.69 4.19 4.13 4.31 4.26 4.36 4.77 4.24 4.21 1 2 2 3 1 1 1 1 1 2 1 1 1 2 2 3 1 1 1 2 1 3 1 1 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 342 2.1 The taste of organic Between Groups food Within Groups Total 2.2 The smell of organic Between Groups food Within Groups Total 2.3 The appearance of Between Groups organic food Within Groups Total 2.4 The overall quality of Between Groups organic food Within Groups Total 2.5 The price of organic Between Groups food Within Groups Total 2.6 The availability of Between Groups organic food in convenience Within Groups stores and supermarkets Total 2.7 The promotion and Between Groups advertising of organic food Within Groups Total 2.8 The value of organic Between Groups foods relative to its price Within Groups Total 2.9 The environmental Between Groups benefits of organic food Within Groups Sum of Squares 2.513 517.617 520.130 6.624 488.799 495.422 4.468 649.169 653.637 4.095 443.328 447.422 2.297 522.911 525.208 2.072 533.162 535.234 2.844 659.337 662.180 3.598 576.055 579.654 6.517 594.219 Table H: ANOVA df 4 703 707 4 698 702 4 700 704 4 698 702 4 697 701 4 695 699 4 694 698 4 697 701 4 701 F .853 2.365 1.204 1.612 .765 .675 .748 1.088 1.922 Mean Square .628 .736 1.656 .700 1.117 .927 1.024 .635 .574 .750 .518 .767 .711 .950 .900 .826 1.629 .848 .105 .361 .559 .609 .548 .169 .308 .052 Sig. .492 343 Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total 2.17 Government regulations and policies relating to the sale of organic food 2.18 Food safety in Between Groups relation to organic food Within Groups Total 2.19 Certification relating to Between Groups the quality of organic food Within Groups Total 2.20 Enforcement relating Between Groups 2.16 The brand name of the organic food 2.15 Organic food that is produced in China 2.14 The country of origin of the organic food 2.13 The awareness of organic food products 2.10 The idea of face saving (mianzi) when purchasing organic food 2.11 The social status of people purchasing organic food 2.12 The knowledge of organic food products 705 4 700 704 4 702 706 4 700 704 4 702 706 4 701 705 4 702 706 4 704 708 4 702 706 4 700 704 4 699 703 4 600.737 7.854 820.478 828.332 22.929 825.099 848.028 .839 591.383 592.221 20.527 554.022 574.549 6.126 754.181 760.307 .943 673.441 674.385 6.932 611.934 618.866 3.103 652.588 655.692 12.555 556.461 569.016 3.357 485.637 488.994 2.358 .248 6.502 .210 .845 5.132 .789 3.948 1.208 .786 .839 .695 .589 .835 1.994 .246 3.139 .795 .776 .930 1.733 .869 .236 .959 1.424 4.877 5.732 1.175 1.532 1.076 1.675 1.964 1.172 .534 .306 .004 .503 .094 .912 .224 .000 .911 .001 .154 344 to the quality of organic Within Groups food Total 2.21 Packaging of organic Between Groups food Within Groups Total 2.22 Correct labelling of Between Groups organic food Within Groups Total 2.23 Information about the Between Groups nutritional value of organic Within Groups food Total 524.956 527.313 3.954 649.647 653.601 2.910 506.344 509.254 4.176 521.839 526.016 700 704 4 702 706 4 700 704 4 703 707 1.068 1.006 1.406 .988 .925 .727 .723 1.044 .742 .750 .230 .404 .371 345 For Family For Gift For Elderly For Children For Own Consumption (I) 5.2.Single usage group For Children 2.13 The awareness of organic For Own food products Consumption 2.11 The social status of people purchasing organic food Tukey HSD Dependent Variable dimension3 dimension3 dimension3 dimension3 dimension3 dimension3 Own Consumption For Elderly Gift For Family Own Consumption For Children Gift For Family Own Consumption For Children For Elderly For Family Own Consumption For Children For Elderly Gift Gift For Children For Elderly Gift For Family Own Consumption For Elderly Gift For Children For Elderly Gift For Children For Elderly Gift For Family (J) 5.2.Single usage group Std. Error .129 .217 .332 .134 .219 .334 .091 .134 .238 .346 .130 .219 .238 .388 .217 .334 .346 .388 .332 .091 .130 .217 .332 .282 .109 .180 .274 .074 .109 .195 .284 Mean Difference (I-J) -.145 .114 -.337 -.442* -.339 -1.040* -.154 .442* .103 -.598 .288 .339 -.103 -.700 .185 1.040* .598 .700 .885 .154 -.288 -.185 -.885 .012 -.432* -.472 -.711 -.246* .432* -.041 -.280 Table I: Five intended usages: Multiple Comparisons .009 .993 .419 .174 .533 .993 .371 .914 .016 .419 .371 .061 .434 .174 .914 .061 1.000 .001 .067 .071 .009 .001 1.000 .862 Sig. .794 .985 .848 .009 .533 .016 .434 .08 -.55 -1.55 -.07 -.26 -.75 -1.76 -.41 .13 -.35 -.36 -.02 -.09 -.64 -.78 -1.79 -.76 -.73 -.96 -1.46 -.45 .13 -.57 -1.06 .81 .75 .35 .64 .94 .55 .36 .78 1.95 1.55 1.76 1.79 .40 .07 .41 .02 .78 -.13 .02 .04 -.04 .73 .49 .50 346 95% Confidence Interval Lower Bound Upper Bound -.50 .21 -.48 .71 -1.24 .57 -.81 -.08 -.94 .26 -1.95 -.13 -.40 .09 For Children 2.18 Food safety in relation to For organic food Own Consumption For Family For Gift For Elderly For Children For Family FOR Gift For Elderly dimension3 dimension3 dimension3 dimension3 dimension3 dimension3 dimension3 dimension3 dimension3 For Family Own Consumption For Children Gift For Family Own Consumption For Children For Elderly For Family Own Consumption For Children For Elderly Gift Own Consumption For Elderly Gift For Family Own Consumption For Children Gift For Family Own Consumption For Children For Elderly For Family Own Consumption For Children For Elderly Gift For Children For Elderly Gift For Family Own Consumption For Elderly Gift .185 .472 .041 -.239 .226 .711 .280 .239 .465 .246* -.185 -.226 -.465 .165 -.007 -.034 -.052 .172 .007 -.027 -.044 .199 .034 .027 -.017 .216* .052 .044 .017 .401* .340 -.007 .168 -.401* -.061 -.408 .106 .180 .195 .318 .178 .274 .284 .318 .272 .074 .106 .178 .272 .114 .205 .298 .111 .189 .205 .333 .187 .287 .298 .333 .286 .078 .111 .187 .286 .110 .181 .275 .075 .110 .196 .285 .403 .067 1.000 .944 .709 .071 .862 .944 .430 .009 .403 .709 .430 .602 1.000 1.000 .990 .892 1.000 1.000 .999 .958 1.000 1.000 1.000 .045 .990 .999 1.000 .003 .327 1.000 .165 .003 .998 .608 -.10 -.02 -.49 -1.11 -.26 -.04 -.50 -.63 -.28 .04 -.47 -.71 -1.21 -.15 -.57 -.85 -.36 -.34 -.55 -.94 -.55 -.59 -.78 -.89 -.80 .00 -.25 -.47 -.76 .10 -.15 -.76 -.04 -.70 -.60 -1.19 .47 .96 .57 .63 .71 1.46 1.06 1.11 1.21 .45 .10 .26 .28 .48 .55 .78 .25 .69 .57 .89 .47 .98 .85 .94 .76 .43 .36 .55 .80 .70 .83 .74 .37 -.10 .47 .37 347 *. The mean difference is significant at the 0.05 level. For Family For Gift For Elderly dimension3 dimension3 dimension3 For Family Own Consumption For Children Gift For Family Own Consumption For Children For Elderly For Family Own Consumption For Children For Elderly Gift -.233 -.340 .061 -.347 -.172 .007 .408 .347 .175 -.168 .233 .172 -.175 .107 .181 .196 .319 .179 .275 .285 .319 .273 .075 .107 .179 .273 .187 .327 .998 .813 .871 1.000 .608 .813 .969 .165 .187 .871 .969 -.52 -.83 -.47 -1.22 -.66 -.74 -.37 -.53 -.57 -.37 -.06 -.32 -.92 .06 .15 .60 .53 .32 .76 1.19 1.22 .92 .04 .52 .66 .57 348 Appendix 9: Referred Journal, Book and Conference Publications associated with this thesis Chen, J 2009, ‘A study investigating the determinants of consumer buyer behaviour relating to purchase of organic food products in China: a project proposal’, ANZMAC doctoral colloquium, Melbourne, 27-29 Nov. Chen, J, Lobo, A & Mascitelli, B 2009, ‘Consumer buyer behaviour associated with the purchase of organic food products in China: development of a conceptual model’, Global Business and Technology Association 11th annual international conference, Prague, 8-11 July, pp. 206-212. Chen, J, Lobo, A & Mascitelli, B 2010, ‘Buyer behaviour of organic food in China: consumers have their say’, Global Business and Technology Association 12th annual international conference, Mpumalanga, South Africa, 5-9 July, pp. 87-94. Chen, J 2010, ‘Factors determining organic food consumption: a first national study in China’, BioFach China conference on international organic food markets and development, Shanghai, 27-28 May. Chen, J, 2010, ‘China agri-business feature: Australia’s share of China’s organic pie’, Australia China Connections, Nov./Dec, pp.32-33 Chen, J & Lobo, A 2010, ‘An exploratory study investigating the dimensions influencing consumers’ purchase intention relating to organic food in Urban China’, Australia & New Zealand Marketing Academy, University of Canterbury, Christchurch, 29 Nov.-1 Dec. Chen, J, Lobo, A & Mascitelli, B 2011, ‘Investigating the buyer behaviour of organic food in urban China’, Global Business and Technology Association 13th annual international conference, Istanbul, 12-16 July. 349 Chen, J 2011 ‘Important attributes to pre-purchase evaluation relating to organic food in urban China’, 3rd Scientific Conference of the International Society of Organic Agriculture Research (ISOFAR)-the 17th Organic World Congress, Namyangju, South Korea, 28 Sept 1 Oct, pp.84-87. Wijesinghe, JMRC & Chen, J 2011, ‘Environmental marketing: a source of competitive advantage’, 2nd International Conference on Business & Information, Kelaniya, Sri Lanka, 20 Oct. Chen, J & Lobo, A 2012, ‘Organic food products in China: determinants of consumers’ purchase intentions’, International Review of Retail, Distribution and Consumers Research, vol. 22, no. 3, pp. 293-314. Chen, J & Lobo, A 2012, ’Consumption of organic food in urban China: investigating determinants important to buyers and a segmentation analysis of their usage pattern’, Business growth in emerging markets: a debate on critical perspectives, Nova, Hauppauge, NY. Lobo, A & Chen, J 2012, ‘The influence of consumers’ lifestyle segments on the purchase intention of organic food in urban China’, Academy of World Business, Marketing & Management Development conference, Budapest, 16-19 July. 350