Journal of Air Transport Management 32 (2013) 58e64 Contents lists available at SciVerse ScienceDirect Journal of Air Transport Management journal homepage: www.elsevier.com/locate/jairtraman Online drivers of consumer purchase of website airline tickets Tomás Escobar-Rodríguez a, Elena Carvajal-Trujillo b, * a b Department of Accounting and Information Systems, University of Huelva, Plaza de la Merced, 21002 Huelva, Spain Department of Business Administration and Marketing, University of Huelva, Plaza de la Merced, 21002 Huelva, Spain a b s t r a c t Keywords: Airline marketing Online consumer behavior Behavioral intentions Airline e-commerce Airline tickets This study aims to examine the different drivers of online airline ticket purchasing behavior and to validate a new conceptual framework (Venkatesh et al., 2012) in this context. Based on the Extended Uniﬁed Theory of Acceptance and Use of Technology (UTAUT2), this paper includes seven explanatory variables: performance expectancy, effort expectancy, social inﬂuence, facilitating conditions, hedonic motivation, price saving, and habit. Data from 1360 usable questionnaires, collected in Spain, were tested against the research model. Our ﬁndings indicate that the main predictors of online purchase intention are, in order of relevance, habit, price saving, performance expectancy, and facilitating conditions. However, there is no signiﬁcant impact of effort expectancy on the online purchase intention, social inﬂuence from referents; and hedonic motivation to use the website. On the other hand, the results highlight that the main predictors of use behavior are, in order of importance, online purchase intention, habit, and facilitating conditions. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction The Internet is ideal for the tourism industry due to the characteristics of its products (McCole, 2002) which are: 1) intangible, 2) their production and consumption are inseparable 3) perishable and 4) seasonal. Among the products consumers would rather book online instead of at their local travel agent are airline tickets, accommodation as well as other tourism products and services (Morrison et al., 2001). The advantages that these online purchases bring are convenience, price reductions and product diversity (Forgas et al., 2012). The use of IT permits the tourism and hospitality industry to obtain competitive advantages, offering lower distribution costs, greater tourist satisfaction, marketing efﬁcacy and greater competitiveness (Tsai et al., 2005). In the context of tourism, this study adapts a new model, UTAUT2, to the formulation of online purchase intentions and online purchase (Venkatesh et al., 2012). UTAUT2 presents some advantages over UTAUT in consumer purchase of online airline tickets. In addition, this paper adapts the price value construct because online purchase does not only entail zero cost for the user * Corresponding author. Tel.: þ34 959217821; fax: þ34 959217839. E-mail addresses: [email protected] (T. Escobar-Rodríguez), [email protected] dem.uhu.es (E. Carvajal-Trujillo). 0969-6997/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jairtraman.2013.06.018 but also a series of beneﬁts. In addition, the direct inﬂuence of the adaptation of the price value construct on the use of websites when buying airline tickets is incorporated. This study aims to analyze the different drivers of online airline ticket purchasing behavior and to validate a new conceptual framework, the UTAUT2 (Venkatesh et al., 2012), in this context. 2. Theoretical background The analysis of the intention to use has traditionally been carried out via UTAUT (San Martín and Herrero, 2012), TAM (Bigné et al., 2010; Herrero and San Martín, 2012) and TPB (Ruiz-Mafe et al., 2013), which are models designed for use in an organizational context (i.e., to understand the factors which inﬂuence employee IT acceptance and/or use) but not necessarily in a consumer context (i.e., to speciﬁcally explain IT acceptance and/or use), as is the case with UTAUT2. UTAUT2 is an extension of UTAUT in a “consumer use context” (Venkatesh et al., 2012). What is more, UTAUT2 incorporates additional constructs related to the consumer such as hedonic motivation, price value and habit. As a result, it can be said that its main contribution is its adaptation to consumer IT acceptance. UTAUT2 redeﬁnes UTAUT constructs from the point of view of the consumer instead of from the organization employee (Venkatesh et al., 2012). UTAUT2 also integrates new constructs and new relations to explain consumer IT acceptance and use (Venkatesh et al., 2012). UTAUT2 considers seven constructs; four T. Escobar-Rodríguez, E. Carvajal-Trujillo / Journal of Air Transport Management 32 (2013) 58e64 from UTAUT (performance expectancy, effort expectancy, social inﬂuence and facilitating conditions) plus three new constructs: hedonic motivation, price value, and habit. UTAUT2 adapts the deﬁnitions of the seven constructs to consumer technology acceptance (Venkatesh et al., 2012; Table 1). The incorporation of the hedonic or intrinsic motivation construct complements UTAUT which only considers the extrinsic motivation or utilitarian value through the performance expectancy construct (Venkatesh et al., 2012). The performance expectancy construct has proved to be the strongest predictor of use intention (Venkatesh et al., 2003). It is precisely in the consumer context that hedonic motivation has been demonstrated to be a relevant predictor of technology acceptance and use (Brown and Venkatesh, 2005; Childers et al., 2001). The second construct incorporated in UTAUT2, price value, endeavors to consider the differences between the monetary cost of technology use in the organizational context where UTAUT was conﬁgured, and the consumer context. In this study; the online airline ticket purchasing does not entail a cost for the consumer but could result in saving money or obtaining the best product for a given price (Babin et al., 1994; Ryan and Rao, 2008). Thus, the proposed model includes price saving and perceived beneﬁts instead of the price value construct. Finally, the habit construct has been used in several papers as a predictor of technology use. The habit has been introduced as predictor of technology actual use in the literature (Davis and Venkatesh, 2004; Kim and Malhotra, 2005; Kim et al., 2005; Limayem et al., 2007; Venkatesh et al., 2012; Wang et al., 2013). Limayem et al. (2007) indicated that future research should continue analyzing the inﬂuence of habit in technology actual usage. UTAUT2 integrates the habit construct to explain consumer technology acceptance and use (Venkatesh et al., 2012). Then, several papers showed habit is a relevant factor inﬂuencing technology use and Venkatesh et al. (2012) incorporated habit into UTAUT2 in order to explain consumer technology use. UTAUT2 operationalizes habit according to Limayem et al. (2007), that is to say, as a self-reported perception. According to Limayem et al. (2007)’s approximation, which operationalize habit following a perception-based approach, it has been demonstrated that habit has a direct effect on technology use and furthermore a more moderate effect on intention, since intention to use technology is of less importance as habit increases (Limayem et al., 2007). Bearing in mind the relations and constructs of UTAUT2 and the reviewed literature, we propose the following hypothesis with respect to online purchase intention and online purchase: 59 H1. The performance expectancy in the use of airline company e-commerce websites positively affects online purchase intention. H2. The effort expectancy in the use of airline company e-commerce websites positively affects online purchase intention. H3. The social inﬂuence regarding the use of airline company e-commerce websites positively affects online purchase intention. H4. The facilitating conditions perceived in the use of airline company e-commerce websites positively affect online purchase intention. H5. The hedonic motivation perceived in the use of airline company e-commerce websites positively affects online purchase intention. H6. The habit regarding the use of airline company e-commerce websites positively affects online purchase intention. H7. The facilitating conditions perceived in the use of airline company e-commerce websites positively affect online purchase use. H8. The habit regarding the use of airline company e-commerce websites positively affects online purchase use. H9. The online purchase intention en airline company e-commerce websites positively affects online purchase use. H10. Price saving and perceived beneﬁts in the use of airline company e-commerce websites positively affect online purchase intention. H11. Price saving and perceived beneﬁts in the use of airline company e-commerce websites positively affect online purchase use. 3. Methodology 3.1. Measurements We adapt a preliminary list of measurement items from ecommerce, IT acceptance and tourism literature: original UTAUT, extended UTAUT (UTAUT2), online shopping, e-commerce, other studies and associated theories (Venkatesh et al., 2003; Wu and Wang, 2005; Jensen, 2012; San Martín and Herrero, 2012; Venkatesh et al., 2012; Wen, 2012). 28 items were generated through this procedure. Table 4 provides a detailed summary of items measured through multi-item scales for the measurement of the constructs in which responses from the participants were measured by a seven-point Likert scale. This seven point scale anchored from 1 (¼“strongly disagree”) to 7 (¼“strongly agree”). Table 1 Deﬁnition of constructs in UTAUT and UTAUT2. Core constructs UTAUT deﬁnitions UTAUT2 deﬁnitions Performance expectancy “The degree to which an individual believes that using the system will help him or her to attain gains in job performance” “The degree of ease associated with the use of the system” “The degree to which using a technology will provide beneﬁts to consumers in performing certain activities” “The degree of ease associated with consumers’ use of technology” “The consumers perceive that important others (e.g., family and friends) believe they should use a particular technology” “Consumers’ perceptions of the resources and support available to perform a behavior (e.g., Brown and Venkatesh, 2005; Venkatesh et al., 2003)” “The fun or pleasure derived from using a technology” “Consumers’cognitive tradeoff between the perceived beneﬁts of the applications and the monetary cost for using them” (Dodds et al., 1991) “The extent to which people tend to perform behaviors automatically because of learning” (Limayem et al., 2007) Effort expectancy Social inﬂuence “The degree to which an individual perceives that important others believes he or she should use the new system” Facilitating conditions “The degree to which an individual believes that an organizational and technical infrastructure exists to support use of the system” Hedonic motivation Price value Not considered Not considered Habit Not considered 60 T. Escobar-Rodríguez, E. Carvajal-Trujillo / Journal of Air Transport Management 32 (2013) 58e64 With this scale variables can be measured that are not directly observed or quantiﬁable (Churchill and Iacobucci, 2002). In order to validate the questionnaire, tourism professionals and academics assessed whether these items were likely to be appropriate for the aim of to analyzing the different drivers of online airline ticket purchasing behavior. On the basis of the comments realized, some modiﬁcations were made to this instrument such as the rewording of some items for reason of clarity. Following this, a pre-test was conducted on 110 randomly selected subjects of different genders and ages with experience in purchasing tickets via airline company websites. During the pretest, subjects who had not previously purchased airlines tickets online were eliminated. On the basis of the feedback received modiﬁcations were carried out regards question clarity. A questionnaire was created in English and this questionnaire was administered in Spanish for consumers from Spain who had experienced of using airline company websites to purchase tickets The English questionnaire was translated to Spanish by a professional native English translator and researchers independently translated the English questionnaire to Spanish. After analyzing the different independently translated Spanish questionnaires, the ﬁnal version of the questionnaire was agreed on. This Spanish questionnaire was translated back to English by another professional native English translator to ensure the consistency between the Spanish and the English versions of the questionnaire (Brislin, 1970; Venkatesh et al., 2012). As shown in Table 4, the Performance expectancy in this study was measured by four items (San Martín and Herrero, 2012; Venkatesh et al., 2012). The Effort expectancy was measured by four items (Venkatesh et al., 2012). The Facilitating conditions construct comprised three items (San Martín and Herrero, 2012; Venkatesh et al., 2012). The Hedonic motivation and Social Inﬂuence constructs were measured by three items each (Venkatesh et al., 2012). The Price-saving orientation construct had three items (Jensen, 2012; Wen, 2012). The Habit construct was composed of four items (Venkatesh et al., 2012). The Behavioral intention was measured by three items (San Martín and Herrero, 2012; Venkatesh et al., 2012) and the Use construct had one item (Venkatesh et al., 2012). 3.2. Sample and data collection The sample was chosen by different non-random samplings because the population size was unknown (San Martín and Herrero, 2012). First, the quota sampling method was employed in our study (Kim et al., 2009; Kim et al., 2011; San Martín and Herrero, 2012). This method was used to identify the number of Spanish individuals needed to match this sample with the target population structure in both age and gender. In order to calculate the appropriate number of Spanish respondents in each age and gender category the target population data were obtained from a study produced by the Asociación para la Investigación de Medios de Comunicación on “Encuesta AIMC a usuarios de Internet 2011” (AIMC) (see Table 2). Males and females were 54.8 per cent and 45.2 per cent, respectively. Age groups were represented as follows: 20.9 per cent for ages 18e24, 25.6 per cent for ages 25e34, 24 per cent for ages 35e44, and 29.6 per cent for ages 45 and over. Second, the convenience sampling method was utilized to collect the quotas (San Martín and Herrero, 2012). In different sessions the interviewers were instructed to administer the questionnaire to subjects from their geographical areas and who during the last six months had used, airline company e-commerce websites either to search for speciﬁc information or to purchase airline tickets. From November 2 to December 21, 2012 the main survey was administered to 1600 subjects from Table 2 Quota sampling method in terms of gender and age. Sampling (n ¼ 1360) Gender Age a Male Female Total 18e24 25e34 35e44 >45 Total Populationa (%) Sample (%) 54.8 45.2 100 20.9 25.6 24 29.6 100 53.6 46.4 100 22.65 31.10 23.46 22.57 100 AIMC (2012). different geographical areas and who were chosen to match the proportion of age and gender of target population who had visited different airline tickets websites during the last six months. 1504 (94%) questionnaires were completed by the subjects, 1360 (85%) of which were collected for data analysis, after eliminating some questionnaires that had been deemed unusable. Table 3 shows the gender and the age of the respondent subjects. There were more males (53.6%) than females (46.4%). The largest group of respondents (31.1%) was the 25e34 age group, followed by ages between 35 and 44 (23.46%), and ages over 45 (22.57%). 4. Results In this study, to analyze the structural equation modeling (SEM), AMOS 20.0 was used. SEM simultaneously tests the multiple hypotheses by estimating the relationships between a set of multiple independent and dependent variables in a structural model (Gefen et al., 2000). Whereas partial-least-squares (PLS) maximize variance using a principle component estimation. Both, SEM and PLS have been used in information system papers (Nusair et al., 2013). We have used SEM due to the large sample size, the constructs are reﬂective and not formative, and the research is conﬁrmatory and not exploratory. The analysis of data took place through a two-stage methodology, in which the measurement model was developed ﬁrst and evaluated separately from the structural model (Fornell and Laker, 1981). 4.1. Measurement model evaluation The measurement model was assessed utilizing conﬁrmatory factor analysis (CFA) (Anderson and Gerbing, 1992). Given the maximum likelihood method of estimation requires the data set to be multivariate normal, this assumption was evaluated and the results reported that it was not violated. Thus, the maximum likelihood was used. This estimation method in SEM is one of the most commonly utilized (Ryu, 2011). The goodness-of-ﬁt measures Table 3 Gender and age characteristics of respondents. Gender Age Characteristics Frequency % Male Female Total 18e24 25e34 35e44 >45 Total 729 631 1360 308 423 319 307 1360 53.6 46.4 100 22.65 31.10 23.46 22.57 100 T. Escobar-Rodríguez, E. Carvajal-Trujillo / Journal of Air Transport Management 32 (2013) 58e64 61 Table 4 Measurement model assessment results. Scale items Performance expectancy (PE) (a ¼ 0.919) PE1. I ﬁnd airline company e-commerce websites very useful in the purchasing process. PE2. Using airline company e-commerce websites increases my chances of achieving things that are important to me in the purchasing process. PE3. Using airline company e-commerce websites helps me accomplish things more quickly in the purchasing process. PE4. I can save time when I use airline company e-commerce websites in the purchasing process. Effort expectancy (EE) (a ¼ 0.933) EE1. Learning how to use airline company e-commerce websites is easy for me. EE2. My interaction with airline company e-commerce websites is clear and understandable. EE3. I ﬁnd airline company e-commerce websites easy to use. EE4. It is easy for me to become skillful at using airline company e-commerce websites. Social Inﬂuence (SI) (a ¼ 0.918) SI1. People who are important to me think that I should use airline company e-commerce websites. SI2. People who inﬂuence my behavior think that I should use airline company e-commerce websites. SI3. People whose opinions that I value prefer that I use airline company e-commerce websites. Facilitating conditions (FC) (a ¼ 0.831) FC1. I have the resources necessary to use airline company e-commerce websites. FC2. I have the knowledge necessary to use airline company e-commerce websites. FC3. I feel comfortable using airline company e-commerce websites. Hedonic motivation (HM) (a ¼ 0.916) HM1. Using airline company e-commerce websites is fun. HM2. Using airline company e-commerce websites is enjoyable. HM3. Using airline company e-commerce websites is very entertaining. Price-saving orientation (PO) (a ¼ 0.827) PO1. I can save money by examining the prices of different airline company e-commerce websites. PO2. I like to search for cheap travel deals in different airline company e-commerce websites. PO3. Airline company e-commerce websites offer better value for my money. Habit (HT) (a ¼ 0.910) HT1. The use of airline company e-commerce websites has become a habit for me. HT2. I am addicted to using airline company e-commerce websites. HT3. I must use airline company e-commerce websites. HT4. Using airline company e-commerce websites has become natural to me. Behavioral intention (BI) (a ¼ 0.941) BI1. I intend to continue using airline company e-commerce websites to purchase tickets in the future. BI2. I will always try to use airline company e-commerce websites to purchase tickets. BI3. I plan to continue to use airline company e-commerce websites frequently to purchase tickets. Use behavior (AU) AU1. How often do you use airline company e-commerce websites to purchase tickets? Standardized loadings t-values SMCs 0.837 0.813 36.339 34.625 0.7 0.661 0.877 39.007 0.768 0.849 36.643 0.721 0.894 0.915 0.905 0.818 42.007 43.682 42.877 36.390 0.8 0.837 0.819 0.669 0.871 0.915 0.88 39.564 42.870 40.216 0.758 0.837 0.774 0.662 0.923 0.983 25.963 38.665 37.121 0.439 0.853 0.967 0.849 0.943 0.829 37.123 43.513 35.523 0.721 0.888 0.687 0.749 0.833 0.739 30.344 35.671 29.825 0.561 0.694 0.546 0.869 0.634 0.881 0.938 38.574 25.152 40.364 44.353 0.755 0.402 0.776 0.88 0.927 0.877 0.902 44.337 40.201 42.216 0.86 0.768 0.813 NA NA NA AVE CCR 0.713 0.908 0.781 0.934 0.790 0.919 0.752 0.899 0.766 0.907 0.600 0.818 0.703 0.903 0.814 0.929 NA NA Model ﬁt statistics: c2 ¼ 1301.87, df ¼ 303, normed-c2 ¼ 4.297, GFI ¼ 0.934, AGFI ¼ 0.912, NFI ¼ 0.964, RMSEA ¼ 0.049, CFI ¼ 0.972. NA ¼ not available. were utilized to examine the measurement model ﬁt. The ﬁt indices, as suggested by the results (Table 4), were obtained within the recommended values, chi-square was 1.301.87 with 303 degrees of freedom, and normed chi-square was 4.297. The goodnessof-ﬁt index (GFI) was 0.934, the adjusted goodness-of-ﬁt index (AGFI) was 0.912, the normed ﬁt index (NFI) was 0.964, the root mean square error of approximation (RMSEA) was 0.049, and the comparative ﬁt index (CFI) was 0.972. Thus, the measurement model indicated an acceptable ﬁt (Hair et al., 1998). A conﬁrmatory factor analysis (CFA) was evaluated to examine reliability, convergent validity, and discriminant validity for proposed constructs. This is to test the indicators represent the constructs and then to evaluate the structural model evaluation (Hair et al., 2011). The reliability evaluates the measurement scale for each construct. Then, the coefﬁcients of Cronbach’s alpha were calculated (Table 4) for constructs and they exceed the 0.7 cut-off value as suggested Hair et al. (1998). Thus, constructs have acceptable reliability. Convergent validity represents the common variance between the indicators and their construct. It is measured by the standardized loadings, the squared multiple correlations (SMC), the average variance extracted (AVE) and construct composite reliability (CCR). All factor loadings (Table 4) in the indicators for the constructs exceeded the 0.5 cut-off value as suggested Steenkamp and Van Trijp (1991). Factor loadings were all signiﬁcant, the t-value (Table 4) exceeded the 1.96 cut-off value. All SMC were greater than 0.4 (Bollen, 1989). Additionally, the AVE and CCR were calculated. The AVE (Table 4) exceeded the 0.5 cut-off value (Fornell and Laker, 1981), and the CCR (Table 4) exceeded the 0.7 cut-off value (Fornell and Laker, 1981). Thus, these results support the convergent validity of each construct. The discriminant validity was examined by following the process suggested by Anderson and Gerbing (1988). Thus, conﬁdence intervals for the correlation between constructs were calculated and compared with the unit. None of the intervals includes value 1.0, thus these results support the discriminant validity of the model (Table 5). 4.2. Structural model Given successful results in the measurement model evaluation, the researchers evaluated the structural model to test H1 through H11. The goodness-of-ﬁt measures were utilized to examine the structural model ﬁt. The ﬁt indices were obtained within the recommended values (Table 6), chi-square was 1306.156 with 307 degrees of freedom, and normed chi-square was 4.255. The goodness-of-ﬁt index (GFI) was 0.934, the adjusted goodness-of-ﬁt index (AGFI) was 0.913, the normed ﬁt index (NFI) was 0.963, the root mean square error of approximation (RMSEA) was 0.049, and the comparative ﬁt index (CFI) was 0.972. These ﬁt indices indicated an acceptable structural model. The results presented in Fig. 1 0.733 (0.651; 0.815) 0.821 (0.793; 0.849) 0.615 (0.525; 0.705) 0.707 (0.669; 0.745) 0.705 (0.671; 0.739) 0.578 (0.492; 0.664) Table 6 Summary of test results for the structural model. H1 H2 H3 H4 H5 H10 H6 H9 H7 H11 H8 P-value Supported? Construct R-squared PE / BI EE / BI SI / BI FC / BI HM / BI PO / BI HT / BI BI / AU FC / AU PO / AU HT / AU 0.141 0.017 0.019 0.097 0.021 0.310 0.462 0.392 0.102 0.002 0.328 <0.01 0.649 0.318 <0.001 0.406 <0.001 <0.001 <0.001 <0.001 0.958 <0.001 Yes No No Yes No Yes Yes Yes Yes No Yes Behavioral intention 0.82 Use behavior 0.579 5. Conclusions 0.628 0.672 0.66 0.501 0.641 0.82 0.734 0.727 0.592 (0.363; 0.463 (0.449; 0.541) (0.346; 0.442) (0.39; 0.494) (0.435; 0.527) (0. 277; 0.461) 0.413 0.495 0.394 0.442 0.481 0.369 0.505) 0.815) 0.662) 0,74) 0.858) 0.773) 0.652) b a Correlation between variables. Conﬁdence interval for the correlation. (0.405; (0.763; (0.582; (0.668; (0.802; (0.713; (0.472; 0.455 0.789 0.622 0.704 0.83 0.743 0.562 (0.539; (0.427; (0.626; (0.573; (0.576; (0.644; (0.804; (0.632; Path and Table 6 summarize the relationship between the different constructs. (0.605; 0.677) (0.79; 0.85) (0.7; 0.768) (0.697; 0.757) (0.506; 0.678) HM EE SI PE 0.581a 0.473 0.66 0.611 0.612 0.682 0.826 0.714 0.623)b 0.519) 0.694) 0.649) 0.648) 0.72) 0.848) 0.796) HT PE SI EE HM FC PO BI AU Table 5 Conﬁdence interval for the correlations between pairs of latent variables. Hypothesized paths Goodness ﬁt statistics: c2 ¼ 1306.156, df ¼ 307, normed-c2 ¼ 4.254, GFI ¼ 0.934, AGFI ¼ 0.913, NFI ¼ 0.963, RMSEA ¼ 0.049, CFI ¼ 0.972. (0.59; 0.666) (0.632; 0.712) (0.624; 0.696) (0.411; 0.591) BI PO T. Escobar-Rodríguez, E. Carvajal-Trujillo / Journal of Air Transport Management 32 (2013) 58e64 FC 62 This research contributes to explaining how consumers use airline websites to purchase air tickets. Our ﬁndings indicate that the main predictors of online purchase intention are, in order of relevance: habit; price saving; performance expectancy; and facilitating conditions. Thus, online purchase intention depends on the individual habit of using the website; the price saving obtained in direct purchase of airline tickets directly from airline company websites; the levels of performance expected by the consumer in completing the online transaction; and the facilitating conditions available. However, there is no signiﬁcant impact of effort expectancy on the online purchase intention, the social inﬂuence from referents; and the hedonic motivation to use the website. On the other hand, the results highlight that the main predictors of use behavior are, in order of importance: online purchase intention; habit and facilitating conditions. Hence, use behavior depends on: online purchase intention, the individual habit in using the website; and the facilitating conditions available to individuals. However, the hypothesized causal relationship introduced in the UTAUT2 of the inﬂuence of price saving on use behavior was not supported. Given the results obtained then, the greater the habit of individuals the more likely they are to have a greater purchase intention, and a greater probability of actual use of airline company websites for online ticket purchase. It is suggested therefore that airline companies in Spain should formulate marketing communication strategies that create habit of online airline ticket purchase intention and thus achieve greater online purchase intention in individuals, and therefore generates an online purchase behavior. This implies that in order to increase online sales via direct purchase from their website, airline companies should advertise on the Internet and/or in other more traditional media about the different usage contexts and occasions on which the website can be utilized to buy a ticket online. Examples of usage contexts and occasions (weekend trip, summer holyday, business trip, present for friends and family) would be the online purchase of an airline ticket. Also, airline company websites should offer, as well as other promotions, a discount on the next online ticket purchase (loyalty programs). The existence of these promotions can be communicated to individuals via email so that they visit the airline company website. With regard to price saving, it can be said that this variable plays a relevant role as a direct driver of online purchase intention. This means that greater money savings, a greater the chance of T. Escobar-Rodríguez, E. Carvajal-Trujillo / Journal of Air Transport Management 32 (2013) 58e64 63 Fig. 1. . Results of testing model. obtaining the best product for a given price and greater perceived beneﬁts of the airline ticket online purchasing, will result in greater intention to use the websites to purchase air tickets. However, in contrast with our expectations, the relation of inﬂuence on online purchase use is positive but not signiﬁcant. Thus, individuals have airline ticket online purchase intention due to the price saving they can obtain, however, this perceived price saving does not inﬂuence actual online purchase, but other factors such as habit or facilitating conditions. For example, airline company websites should publicize via their website that with online direct purchase individuals can obtain lower prices or the best product for a given price such as a ﬂight with fewer o no stopovers or free food during the ﬂight. Performance expectancy (perceived usefulness) which represents utilitarian features is the third driver of the online purchase intention. For this reason airline companies should offer utilitarian features. Included in utilitarian features we highlight free customer care telephone, Twitter or Facebook client communication channels or other types of new media, and the option of presenting boarding pass not only on paper, but by SMS or by WhatsApp. It is demonstrated that effort expectancy does not inﬂuence online purchase intention. The results obtained might be due to the fact that nearly all the individuals included are habituated to using websites, social media platforms (Facebook, Twitter .), which are similar or superior to ecommerce websites. The fact that hedonic motivation does not inﬂuence the online purchase intention can be explained because the individuals frequently use websites and social media platforms, which are more entertaining and fun that the airline company e-commerce websites Additionally, the absence of affect of social inﬂuence on the online purchase intention may be explained by the generalized internet use as an information source about tourism products and services reduces the inﬂuence, positive or negative, of the referents with consumer purchase of online airline tickets. Regarding the inﬂuence of facilitating conditions, this variable inﬂuences online purchase intention and the online purchase use. This implies that individuals’s perceptions about the resources and support available for online purchase inﬂuence the online purchase intention and online purchase use. This shows that airline companies should facilitate the option to purchase via their websites offering diverse media so that consumers are informed if they have any problems formulating their purchase. These could include free or low-cost telephone numbers, email, an active presence in Facebook, Twitter. Finally, regarding the inﬂuence of perceived behavioral intention on perceived use behavior the greater the perceived online purchase intention, the greater the chance of online purchase use. Therefore, managers should try to increase online purchase intention in order to increase online purchase use. Factors such as habit, price saving, hedonic motivation, facilitating conditions, social inﬂuence and performance expectancy can all be acted upon in order to increase online purchase intention. These ﬁndings conﬁrm the validity of UTAUT2 in explaining the online purchase intention and the online purchase use in the context of the direct purchase of airline tickets. References AIMC-Asociación para la investigación de medios de comunicación, 2012. Encuesta AIMC a usuarios de Internet 2011. http://www.aimc.es/. Anderson, J.C., Gerbing, D.W., 1988. Structural modeling in practice: a review and recommended two-step approach. Psychological Bulletin 103 (3), 411e423. Anderson, J.C., Gerbing, D.W., 1992. Assumptions and comparative strengths of the two-step approach. Sociological Methods and Research 20 (3), 321e333. Babin, J.B., Darden, W.R., Grifﬁn, M., 1994. Work and/or fun: measuring hedonic and utilitarian shopping value. Journal of Consumer Research 20 (4), 644e656. Bigné, E., Sanz, S., Ruiz, C., Aldás, J., 2010. Why some internet users don’t buy air tickets online. Information and Communication Technologies in Tourism 6, 209e221. Bollen, K.A., 1989. Structural Equations with Latent Variables. Wiley, New York. Brislin, R.W., 1970. Back translation for cross-cultural research. Journal of CrossCultural Psychology 1 (3), 185e216. Brown, S.A., Venkatesh, V., 2005. Model of adoption of technology in the household: a baseline model test and extension incorporating household life cycle. MIS Quarterly 29 (4), 399e426. Childers, T.L., Carr, C.L., Peck, J., Carson, S., 2001. Hedonic and utilitarian motivations for online retail shopping behavior. Journal of Retailing 77 (4), 511e535. 64 T. Escobar-Rodríguez, E. Carvajal-Trujillo / Journal of Air Transport Management 32 (2013) 58e64 Churchill, G.A., Iacobucci, D., 2002. Marketing Research Methodological Foundations, eighth ed. Mason: Thomson. Davis, F.D., Venkatesh, V., 2004. Toward preprototype user acceptance testing of new information systems: implications for software project management. IEEE Transactions on Engineering Management 51 (1), 31e46. Forgas, S., Palau, R., Sánchez, J., Huertas-García, R., 2012. Online drivers and ofﬂine inﬂuences related to loyalty to airline websites. Journal of Air Transport Management 18, 43e46. Fornell, C., Laker, D.F., 1981. Evaluating structural equation models with unobservable variables and measurement error: algebra and statistics. Journal of Marketing Research 18 (1), 39e50. Gefen, D., Straub, D.W., Boudreau, M.C., 2000. Structural equation modeling and regression: guidelines for research practice. Communications of the Association for Information Systems 4 (7), 1e70. Hair, J., Anderson, R., Tatham, R., Black, W., 1998. Multivariate Data Analysis, ﬁfth ed. Prentice Hall, Upper Saddle River, New Jersey. Hair, J.F., Ringle, C.M., Sarstedt, M., 2011. PLS-SEM: indeed a silver bullet. Journal of Marketing Theory and Practice 19 (2), 139e151. Herrero, Á., San Martín, H., 2012. Developing and testing a global model to explain the adoption of websites by users in rural tourism accommodations. International Journal of Hospitality Management 31 (4), 1178e1186. Jensen, J.M., 2012. Shopping orientation and online travel shopping: the role of travel experience. International Journal of Tourism Research 14 (1), 56e70. Kim, S.S., Malhotra, N.K., 2005. A longitudinal model of continued IS use: an integrative view of four mechanisms underlying post-adoption phenomena. Management Science 51 (5), 741e755. Kim, S.S., Malhotra, N.K., Narasimhan, S., 2005. Two competing perspectives on automatic use: a theoretical and empirical comparison. Information Systems Research 16 (4), 418e432. Kim, H.B., Kim, T., Shin, S.W., 2009. Modeling roles of subjective norms and etrust in customers’ acceptance of airline B2C eCommerce websites. Tourism Management 30 (2), 266e277. Kim, M.J., Chung, N., Lee, C.K., 2011. The effect of perceived trust on electronic commerce: shopping online for tourism products and services in South Korea. Tourism Management 32 (2), 256e265. Limayem, M., Hirt, S.G., Cheung, C.M.K., 2007. How habit limits the predictive power of intentions: the case of IS continuance. MIS Quarterly 31 (4), 705e737. McCole, P., 2002. The role of trust for electronic commerce in services. International Journal of Contemporary Hospitality Management 14 (2), 81e87. Morrison, A.M., Jing, S., O’Leary, J.T., Lipping, A.C., 2001. Predicting usage of the internet for travel bookings: an exploratory study. Information Technology and Tourism 4 (1), 15e30. Nusair, K., Bilgihan, A., Okumus, F., Cobanoglu, C., 2013. Generation Y travelers’ commitment to online social network websites. Tourism Management 35 (April), 13e22. Ruiz-Mafe, C., Sanz-Blas, S., Hernandez-Ortega, B., Brethouwer, M., 2013. Key drivers of consumer purchase of airline tickets: a cross-cultural analysis. Journal of Air Transport Management 27, 11e14. Ryan, C., Rao, U., 2008. Holiday users of the internet: ease of use, functionality and novelty. International Journal of Tourism Research 10 (4), 329e339. Ryu, E., 2011. Effects of skewness and kurtosis on normal-theory based maximum likelihood test statistics in multilevel structural equation modeling. Behavioral Research 43, 1066e1074. San Martín, H., Herrero, Á., 2012. Inﬂuence of the user’s psychological factor son the online purchase intention in rural tourism: integrating innovativeness to the UTAUT framework. Tourism Management 33 (2), 341e350. Steenkamp, J.B., Van Trijp, H.C.M., 1991. The use of LISREL in validating marketing constructs. International Journal of Research in Marketing 8, 283e299. Tsai, H.T., Huang, L., Lin, C.G., 2005. Emerging e-commerce development model for Taiwanese travel agencies. Information and Management 26 (5), 787e796. Venkatesh, V., Morris, M.G., Davis, G.B., Davis, F.D., 2003. User acceptance of information technology: toward a uniﬁed view. MIS Quarterly 27 (3), 425e478. Venkatesh, V., Thong, J.I.L., Xu, X., 2012. Consumer acceptance and use of information technology: extending the uniﬁed theory of acceptance and use of technology. MIS Quarterly 36 (1), 157e178. Wang, C., Harris, J., Patterson, P., 2013. The roles of habit, self-efﬁcacy, and satisfaction in driving continued use of self-service technologies: a longitudinal study. Journal of Service Research. http://dx.doi.org/10.1177/1094670512473200 (in press). Wen, I., 2012. An empirical study of an online travel purchase. Journal of Travel and Tourism Marketing 29 (1), 18e39. Wu, J.H., Wang, S.C., 2005. What drives mobile commerce? an empirical evaluation of the revised technology acceptance model. Information & Management 42 (5), 719e729.