Uploaded by sharmila fdo

THESIS DOCTOR OF PHILOSOPHY ABID Aman 2022

advertisement
Political Social Media Marketing: A Relationship Marketing
Perspective
Aman Abid
MPA, MBus, MCom, BS
This thesis is presented for the degree of Doctor of Philosophy of The
University of Western Australia
UWA Business School
Marketing Discipline Group
2021
THESIS DECLARATION
I, Aman Abid, certify that:
This thesis has been substantially accomplished during enrolment in this degree.
This thesis does not contain material which has been submitted for the award of any other
degree or diploma in my name, in any university or other tertiary institution. In the future, no
part of this thesis will be used in a submission in my name, for any other degree or diploma in
any university or other tertiary institution without the prior approval of The University of
Western Australia and where applicable, any partner institution responsible for the joint-award
of this degree.
This thesis does not contain any material previously published or written by another person,
except where due reference has been made in the text and, where relevant, in the Authorship
Declaration that follows. This thesis does not violate or infringe any copyright, trademark,
patent, or other rights whatsoever of any person.
The research involving human data reported in this thesis was assessed and approved by The
University of Western Australia Human Research Ethics Committee (approval #: RA/4/1/9272,
RA/4/20/5056). Written patient consent has been received and archived for the research
involving patient data reported in this thesis.
This thesis contains published work and work prepared for publication. All works were coauthored.
Signature:
Date: 06/08/2021
i
ABSTRACT
This thesis explores social media-enabled online voter relationships. Its central premise is
social media’s ability to cultivate customer relationships. In an era of declining political
engagement, high political volatility, decreasing party identification, and rising minor parties,
developing long-term relationships with voters is more crucial than ever. Social media are ideal
for fostering these relationships. However, studies that attempt to understand online voter
relationships are scant due to the predominantly transactional nature of political marketing
literature. A transactional approach contradicts the ethos of social media, which require a
relational approach to marketing.
This thesis endeavors to further our understanding of online voter relationships and political
relationship marketing on social media. It comprises five papers and utilizes mixed methods,
systematic review, and qualitative research methodologies. The papers utilize various theories,
models, and concepts like the Uses and Gratification Theory, Elaboration-Likelihood Model,
psychological contract, service-dominant orientation, relationship marketing orientation, and
relationship quality. The thesis aims to answer five questions, which allow us to understand
how political brands can strengthen relationships with their followers via social media.
Paper 1
What is the current state of research on political social media marketing?
Paper 2
What is the nature of social media-enabled voter relationships?
Paper 3
Do politicians adopt a relational approach towards social media?
Paper 4
What is the role of marketer-generated content in driving online voter
relationships?
Paper 5
What are the roles of marketer-generated content and behavioral
engagement in driving online voter relationships?
Paper 1 is a systematic literature review. It provides a comprehensive analysis of the research
devoted to political marketing on social media. Political social media marketing is a growing
area of research. Distinct themes are emerging in the literature. Papers 2 and 3 rely on focus
groups with young voters. Paper 2 explores the relationship that transpires between political
brands and voters due to the latter following the former on social media. It reveals that young
voters initiate online voter relationships to satisfy social, informational, and entertainment
gratifications. The primary drivers of this relationship are trust and social exchanges. Further,
ii
relational, ethical, and developmental interactions strengthen online voter relationships. Paper
3 examines the marketing orientation that is adopted by political brands on social media. The
study shows that major political brands, unlike minor political brands, fail to adopt a
relationship marketing orientation on social media. Minor political brands’ success on social
media is due to their abilities to bond with young voters, evoke trust, emphasize shared values,
and educate young voters.
Papers 4 and 5 are online content analyses of the official Facebook pages of the Republican
and Democrat parties. The papers take a content-driven perspective to study online voter
relationships. Specifically, Paper 4 explores the effects of various content cues on followers’
online relationship quality. It shows that content cues like visuals, length, the volume of
comments, and content popularity increase online relationship quality. Furthermore, sharing
interactive or lengthy content decreases online relationship quality, whereas sharing images
foster it. Paper 5 builds on Paper 4 and devises a more comprehensive framework that
demonstrates how content leads to behavioral engagement, which subsequently drives online
relationship quality. The study validates the propositions of the Elaboration-Likelihood Model
and confirms that argument quality is only effective in high-involvement conditions. However,
peripheral cues like source credibility, emotions, visual symbolism, and valence affect
engagement in high and low-involvement conditions.
The thesis makes novel contributions to relationship marketing, social media marketing, and
political marketing. It integrates contemporary concepts and the relationship marketing
paradigm into political marketing. Primarily, it offers guidance about the nature and drivers of
online voter relationships. Besides theoretical contributions, the thesis guides political
marketers regarding the effective utilization of social media and political marketer-generated
content to build relationships with voters.
iii
TABLE OF CONTENTS
THESIS DECLARATION .................................................................................................................................... I
ABSTRACT .......................................................................................................................................................... II
TABLE OF CONTENTS.................................................................................................................................... IV
LIST OF FIGURES ...........................................................................................................................................VII
LIST OF TABLES ........................................................................................................................................... VIII
ACKNOWLEDGEMENTS................................................................................................................................ IX
AUTHORSHIP DECLARATION: CO-AUTHORED PUBLICATIONS.......................................................X
LIST OF PUBLICATIONS............................................................................................................................. XIII
CHAPTER 1: GENERAL INTRODUCTION ....................................................................................................1
1.1 RESEARCH BACKGROUND...............................................................................................................................1
1.1.1 Research problem ..................................................................................................................................1
1.1.2 Relationship marketing ..........................................................................................................................2
1.1.3 Social media marketing .........................................................................................................................3
1.1.4 Political marketing .................................................................................................................................4
1.1.5 Political relationship marketing and social media marketing ................................................................5
1.2 RESEARCH AIMS .............................................................................................................................................6
1.3 RESEARCH METHODOLOGY ............................................................................................................................7
1.3.1 Research paradigm ................................................................................................................................7
1.3.2 Research questions and designs.............................................................................................................7
1.4 THESIS OVERVIEW ........................................................................................................................................10
1.5 SIGNIFICANCE AND ORIGINALITY .................................................................................................................12
1.6 THESIS STRUCTURE ......................................................................................................................................13
CHAPTER 2: POLITICAL SOCIAL MEDIA MARKETING: A THEMATIC REVIEW,
CONCEPTUAL FRAMEWORK, AND RESEARCH AGENDA FOR THE FUTURE ..............................14
2.1 ABSTRACT ....................................................................................................................................................15
2.2 INTRODUCTION .............................................................................................................................................16
2.3 CONCEPTUAL BOUNDARIES ..........................................................................................................................17
2.3.1 Social media ........................................................................................................................................17
2.3.2 Political marketing ...............................................................................................................................17
2.3.3 Political social media marketing .........................................................................................................18
2.4 RESEARCH METHODOLOGY ..........................................................................................................................18
2.5 FINDINGS ......................................................................................................................................................28
2.5.1 A chronological review .......................................................................................................................28
2.5.2 Research settings .................................................................................................................................29
2.5.3 Themes ................................................................................................................................................31
2.6 DISCUSSION..................................................................................................................................................35
2.6.1 Conceptual framework ........................................................................................................................35
2.6.2 Future research agenda ........................................................................................................................42
2.7 CONCLUSION ................................................................................................................................................47
CHAPTER 3: AN EXPLORATION OF SOCIAL MEDIA-ENABLED VOTER RELATIONSHIPS
THROUGH USES AND GRATIFICATION THEORY, PSCYCHOLOGICAL CONTRACT, AND
SERVICE-DOMINANT ORIENTATION ........................................................................................................48
3.1 ABSTRACT ....................................................................................................................................................49
3.2 INTRODUCTION .............................................................................................................................................50
3.3 LITERATURE REVIEW....................................................................................................................................51
3.3.1 Uses and Gratifications Theory ...........................................................................................................51
3.3.2 Psychological contract .........................................................................................................................53
3.3.3 Service-dominant orientation ..............................................................................................................54
iv
3.4 RESEARCH METHODOLOGY ..........................................................................................................................56
3.5 FINDINGS ......................................................................................................................................................59
3.5.1 What are the uses and gratifications that drive voters to follow PEs on social media? ......................59
3.5.2 What are the factors that drive social media-enabled voter relationships? .........................................62
3.5.3 What are the interactions that underpin social media-enabled voter relationships?............................65
3.6 DISCUSSION AND IMPLICATIONS...................................................................................................................67
3.6.1 Discussion............................................................................................................................................67
3.6.2 Research implications and limitations .................................................................................................72
3.6.3 Managerial implications ......................................................................................................................73
3.6.4 Societal implications ...........................................................................................................................74
CHAPTER 4: A RELATIONHIP MARKETING ORIENTATION IN POLITICS: YOUNG VOTERS’
PERCEPTIONS OF POLITICAL BRANDS’ USE OF SOCIAL MEDIA ....................................................75
4.1 ABSTRACT ....................................................................................................................................................76
4.2 INTRODUCTION .............................................................................................................................................77
4.3 LITERATURE REVIEW....................................................................................................................................78
4.3.1 Political relationship marketing and social media ...............................................................................78
4.3.2 Relationship Marketing Orientation ....................................................................................................79
4.4 RESEARCH METHODOLOGY ..........................................................................................................................81
4.5 FINDINGS ......................................................................................................................................................83
4.6 DISCUSSION AND IMPLICATIONS...................................................................................................................86
4.6.1 Dimensions of RMO............................................................................................................................86
4.6.2 Research, managerial, and societal implications .................................................................................88
4.6.3 Limitations and directions for future research .....................................................................................90
CHAPTER 5: ONLINE RELATIONSHIP MARKETING THROUGH CONTENT CREATION AND
CURATION ..........................................................................................................................................................91
5.1 ABSTRACT ....................................................................................................................................................92
5.2 INTRODUCTION .............................................................................................................................................93
5.3 LITERATURE REVIEW....................................................................................................................................94
5.3.1 Created and curated content ................................................................................................................94
5.3.2 Social media and relationship quality..................................................................................................95
5.3.3 Relationship quality .............................................................................................................................95
5.3.4 Political relationship marketing and social media ...............................................................................96
5.3.5 Elaboration-Likelihood Model ............................................................................................................96
5.3.6 Content cues and the conceptual framework .......................................................................................97
5.4 RESEARCH DESIGN AND METHODOLOGY ....................................................................................................100
5.4.1 Sample and data collection ................................................................................................................100
5.4.2 Operationalization of variables..........................................................................................................101
5.5 A TALE OF TWO PARTIES: A PRELIMINARY ANALYSIS .................................................................................104
5.6 RESULTS .....................................................................................................................................................104
5.6.1 Descriptive statistics ..........................................................................................................................104
5.6.2 Test of hypotheses .............................................................................................................................105
5.7 DISCUSSION AND IMPLICATIONS.................................................................................................................107
5.7.1 Discussion..........................................................................................................................................107
5.7.2 Theoretical implications ....................................................................................................................108
5.7.3 Managerial implications ....................................................................................................................109
5.7.4 Limitations and future research .........................................................................................................110
CHAPTER 6: SOCIAL MEDIA IN POLITICS: HOW TO DRIVE ENGAGEMENT AND
STRENGTHEN RELATIONSHIPS ................................................................................................................111
6.1 ABSTRACT ..................................................................................................................................................112
6.2 INTRODUCTION ...........................................................................................................................................113
6.3 LITERATURE REVIEW..................................................................................................................................115
6.3.1 Elaboration-Likelihood Model ..........................................................................................................115
6.3.2 Social media behavioral engagement ................................................................................................116
6.3.3 Relationship quality ...........................................................................................................................118
6.4 CONCEPTUAL FRAMEWORK ........................................................................................................................119
6.4.1 Argument quality ...............................................................................................................................119
6.4.2 Source credibility...............................................................................................................................120
v
6.4.3 Emotion .............................................................................................................................................121
6.4.4 Valence ..............................................................................................................................................122
6.4.5 Visual symbolism ..............................................................................................................................123
6.4.6 Social media behavioral engagement and online relationship quality ..............................................124
6.5 RESEARCH METHODOLOGY ........................................................................................................................124
6.5.1 Data....................................................................................................................................................125
6.5.2 Coding of content ..............................................................................................................................125
6.5.3 Coding of comments..........................................................................................................................127
6.6 RESULTS .....................................................................................................................................................128
6.6.1 Descriptive statistics ..........................................................................................................................128
6.6.2 Hypotheses testing .............................................................................................................................129
6.7 DISCUSSION AND IMPLICATIONS.................................................................................................................136
6.7.1 Marketer-generated content and social media behavioral engagement .............................................137
6.7.2 Social media behavioral engagement and online relationship quality ..............................................138
6.7.3 Theoretical implications ....................................................................................................................139
6.7.4 Managerial implications ....................................................................................................................140
6.8 LIMITATIONS AND FUTURE RESEARCH........................................................................................................140
CHAPTER 7: GENERAL DISCUSSION AND CONCLUSION ..................................................................141
7.1 SUMMARY AND KEY FINDINGS ...................................................................................................................141
7.1.1 The current state of research on political social media marketing (Paper 1, RQ1) ...........................141
7.1.2 Deconstructing social media-enabled voter relationships (Paper 2, RQ2) ........................................141
7.1.3 Marketing orientation of Australian political brands on social media (Paper 3, RQ3) .....................142
7.1.4 Political marketer-generated content, behavioral engagement, and online relationships (Papers 4 and
5, RQs 4 and 5) ...........................................................................................................................................142
7.2 AN INTEGRATIVE DISCUSSION ....................................................................................................................143
7.3 OVERALL THEORETICAL IMPLICATIONS .....................................................................................................144
7.4 MANAGERIAL IMPLICATIONS......................................................................................................................146
7.5 SOCIETAL IMPLICATIONS ............................................................................................................................147
7.6 LIMITATIONS AND FURTHER RESEARCH .....................................................................................................147
REFERENCES ...................................................................................................................................................150
APPENDIX A. INTRODUCTION OF PARTICIPANTS..............................................................................186
vi
LIST OF FIGURES
Figure 1.1 Project map .............................................................................................................................................9
Figure 2.1 Political social media marketing (2011-2020) ......................................................................................28
Figure 2.2 Conceptual framework..........................................................................................................................40
Figure 3.1 Chapter 3 (Paper 2) ...............................................................................................................................48
Figure 4.1 Chapter 4 (Paper 3) ...............................................................................................................................75
Figure 5.1 Chapter 5 (Paper 4) ...............................................................................................................................91
Figure 5.2 Conceptual framework........................................................................................................................100
Figure 6.1 Chapter 6 (Paper 5) .............................................................................................................................111
Figure 6.2 Conceptual framework........................................................................................................................119
vii
LIST OF TABLES
Table 1.1 Research questions and designs ...............................................................................................................8
Table 2.1 Identification and acquisition .................................................................................................................20
Table 2.2 Search results .........................................................................................................................................21
Table 2.3 Articles included in the review ..............................................................................................................22
Table 2.4 Publication outlets ..................................................................................................................................29
Table 2.5 Research themes.....................................................................................................................................31
Table 2.6 Details of variables included in the conceptual framework ...................................................................36
Table 3.1 Service-dominant orientation .................................................................................................................55
Table 3.2 Examples of items used as a guide for focus group discussions ............................................................58
Table 3.3 Summary of findings ..............................................................................................................................67
Table 4.1 Focus group findings..............................................................................................................................84
Table 5.1 Operationalization of variables ............................................................................................................101
Table 5.2 Coding items and examples for relationship quality ............................................................................103
Table 5.3 Regression Table and ANOVA Results ...............................................................................................106
Table 6.1 Content coding manual ........................................................................................................................126
Table 6.2 Descriptive analysis .............................................................................................................................129
Table 6.3 MANOVA Results – Testing H1- H5..................................................................................................130
Table 6.4 Regression Results – Testing H6 .........................................................................................................134
Table 6.5 PROCESS Results................................................................................................................................135
viii
ACKNOWLEDGEMENTS
I am grateful and indebted to my supervisor, Dr Sanjit K. Roy, for his support and guidance,
which enabled me to finish this research project. I acknowledge the openness, patience, and
kindness with which Sanjit guided me towards the finish line. I thank him for helping me
beyond academics and offering me valuable lessons about professional life and career and
taking an interest in my personal life and wellbeing.
I offer my sincere gratitude to my co-supervisor, Dr Paul Harrigan. Paul played an instrumental
role in this PhD project from its conception till the final stages. I learned a great deal from Paul
about research and academia. For this, I am highly obliged to him.
I recognize the support of my co-supervisor, Dr Tauel Harper. Tauel was always open to my
ideas. I thank him for opening up a new discipline and thought to me. I am appreciative of his
time and insights.
The author acknowledges that this research is supported by an Australian Government
Research Training Program (RTP) Scholarship.
I thank my co-authors Dr Shasha Wang and Dr Jennifer Lees-Marshment for their contributions
to this project. I acknowledge Dr Momoko Fujita’s influence on this project. Her thesis served
as a guide to my thesis. I also thank my colleague Kim Feddema for her support.
Finally, I recognize the sacrifice and contributions of my family, particularly my mother,
Najma, and my wife, Summra, who have been encouraging, understanding, and patient during
my PhD.
ix
AUTHORSHIP DECLARATION: CO-AUTHORED
PUBLICATIONS
This thesis contains work that has been published, under-review, or yet to be submitted.
Details of the work: Abid, A., Roy, S. K., Lees-Marshment, J., Dey, B., & Muhammad,
S. S. (2021). Political social media marketing: A thematic review, conceptual
framework, and research agenda for the future. The manuscript is being submitted to an
international journal.
Location in thesis: Chapter 2 (Paper 1)
Student contribution to work: Aman conceptualized the study, collected data,
analyzed data, and wrote the manuscript in consultation with Dr Roy. The finished
manuscript was shared with Dr Lees-Marshment, who reviewed the paper, suggested
improvements, and offered valuable advice on the agenda for the future research. A final
review was undertaken by Dr Dey and Dr Muhammad who offered relevant suggestions.
Co-author signatures and dates:
Sanjit K. Roy
Jennifer Lees-Marshment
Bidit Dey
Syed Sardar
Muhammad
Details of the work: Abid, A., & Harrigan, P. (2020). An exploration of social mediaenabled voter relationships through uses and gratifications theory, psychological
contract, and service-dominant orientation. Australasian Marketing Journal, 28(2), 71–
82. https://doi.org/10.1016/j.ausmj.2020.02.002
Location in thesis: Chapter 3 (Paper 2)
Student contribution to work: Aman conceptualized the study, reviewed the literature,
designed the research, collected data, analyzed data, and wrote the manuscript in
consultation with Dr Harrigan. The article went through two rounds of revision with the
journal, which included consultation with Dr Harrigan.
Co-author signatures and dates:
Paul Harrigan
x
Details of the work: Abid, A., Harrigan, P., & Roy, S. (2021). A relationship marketing
orientation in politics: Young voters’ perceptions of political brands’ use of social
media. Journal of Strategic Marketing, 29(4), 359–374.
https://doi.org/10.1080/0965254X.2020.1777457
Location in thesis: Chapter 4 (Paper 3)
Student contribution to work: Aman conceptualized the study, reviewed the literature,
designed the research, collected data, analyzed data, and wrote the manuscript in
consultation with Dr Harrigan. The finished manuscript was shared with Dr Roy, who
reviewed the article and offered valuable suggestions for improvement. The article went
through one round of revision with the journal, which included consultations with Dr
Harrigan and Dr Roy.
Co-author signatures and dates:
06/08/2021
Details of the work: Abid, A., Harrigan, P., & Roy, S. K. (2020). Online relationship
marketing through content creation and curation. Marketing Intelligence & Planning,
38(6), 699–712. https://doi.org/10.1108/MIP-04-2019-0219
Location in thesis: Chapter 5 (Paper 4)
Student contribution to work: Aman conceptualized the study, reviewed the literature,
designed the research, collected data, and wrote the manuscript in consultation with Dr
Harrigan. Data analysis was conducted under the supervision of Dr Roy. The paper went
through three rounds of revision, which included consultations with Dr Harrigan and Dr
Roy.
Co-author signatures and dates:
06/08/2021
Details of the work: Abid, A., Harrigan, P., Wang, S., Roy, S. K., & Harper, T. (2020).
Journal of Marketing Management. A revise and resubmit decision has been received.
Location in thesis: Chapter 6 (Paper 5)
Student contribution to work: Aman conceptualized the study, reviewed the literature,
designed the research, collected data, and wrote the manuscript in consultation with Dr
Harrigan. Dr Wang performed the ANOVA analysis and guided Aman during data
analysis and writing process. The manuscript was shared with Dr Roy and Dr Harper,
who provided valuable feedback.
Co-author signatures and dates:
06/08/2021
09/082021
xi
Student signature:
Date: 02/082021
I, Dr Sanjit K. Roy, certify that the student’s statements regarding their contribution to
each of the works listed above are correct.
Coordinating supervisor signature:
Date: 29/07/2021
xii
LIST OF PUBLICATIONS
The following publications are directly related to this thesis.
Refereed journal articles
An exploration of social media-enabled voter relationships through uses and gratifications
theory, psychological contract and service-dominant orientation. Australasian Marketing
Journal, 28(2), 71–82. https://doi.org/10.1016/j.ausmj.2020.02.002 (ABDC: A, IF: 3.29)
A relationship marketing orientation in politics: Young voters’ perceptions of political
brands’ use of social media. Journal of Strategic Marketing, 29(4), 359–374.
https://doi.org/10.1080/0965254X.2020.1777457 (ABDC: A, IF: 3.6)
Abid, A., Harrigan, P., & Roy, S. K. (2020). Online relationship marketing through content
creation and curation. Marketing Intelligence & Planning, 38(6), 699–712.
https://doi.org/10.1108/MIP-04-2019-0219 (ABDC: A, IF: 2.99)
Article in the revision stage
Abid, A., Harrigan, P., Wang, S., Roy, S., and Harper, T. (2020). Social media in politics: How
to drive engagement and strengthen relationships. Journal of Marketing Management. (ABDC:
A, IF: 2.74) – (R&R received, revision due in September, 2021)
Finished article – not submitted yet
Abid, A., Roy, S. K., Lees-Marshment, J., Dey, B. L., & Muhammad, S. S. (2021). Political
social media marketing: A thematic review, conceptual framework, and research agenda for
future research.
Conference papers
Abid, A., & Harrigan, P. (2019). Online voter relationships. Presented at the Australian and
New Zealand Marketing Academy Conference, Wellington, New Zealand, 2-4 December.
Abid, A., Harrigan, P., Roy, S. K. (2019). Online relationships with political brands. Presented
at the 8th IBS Conference on Marketing and Business Strategy, Hyderabad, India, 15-16
November.
xiii
CHAPTER 1: GENERAL INTRODUCTION
This thesis explores political marketing on social media using the relationship marketing lens.
The cultivation of long-term relationships with voters is possible in the era of social media.
With current times witnessing low political involvement and high political cynicism, fostering
relationships with voters is needed more than ever. Traditional political marketing activities
are not ideal for this purpose. Similarly, the view that voters are rational beings driven by
economic-utilitarian concerns is losing ground. Indeed, voters are emotional beings who form
‘parasocial’ relationships with politicians and political parties. Driven by this social and
relational viewpoint, the project explores social media-enabled voter relationships and their
drivers.
1.1 Research background
1.1.1 Research problem
Social media are transforming the political landscape. They provide political parties and
politicians with an unprecedented ability to engage voters and foster relationships with them.
Political brands, however, have been slow to adapt their marketing strategies to social media
(Parsons & Rowling, 2018). Social media are ideally suited to relationship marketing. They
are incompatible with the traditional one-way communication associated with political
marketing (Harris & Harrigan, 2015; Lees-Marshment, 2014). Despite marketing’s
paradigmatic shift from marketing mix to relationship marketing (Grönroos, 1994), political
marketing is stagnant and does not reflect the dynamic nature of marketing. A relationship
marketing perspective focuses on society, mutual benefits, value, and the long term, which
makes it an ideal lens to study political marketing (Ormrod et al., 2013). However, this
relational perspective is lacking in political marketing literature.
The decline in political trust, identification, and satisfaction is a worrying trend plaguing
western democracies since the Second World War (Dalton, 2013). This voter disengagement
is prominent among young voters (Pich et al., 2018; Tiffany & Winchester, 2015). Much is
written on the causes and consequences of this voter alienation. In fact, traditional political
marketing is one of the factors driving this alienation (Henneberg & O’Shaughnessy, 2009).
Political marketing scholars believe that political relationship marketing can reverse voter
1
disengagement and apathy, and increase voter trust and participation (e.g. Ormrod et al., 2013;
Henneberg & O’Shaughnessy, 2009; Bannon, 2005; Scammel, 1999).
The marketing capabilities of social media are acknowledged in commercial (Alalwan et al.,
2017) and political marketing literature (Williams, 2017). The research project rests on the
premise that social media help firms create and sustain relationships with customers. However,
there is limited understanding of the roles that marketer-generated content and social media
play in the development of these relationships (Steinhoff et al., 2019; Sheth, 2017). The same
holds for political marketing, which is a young field that seldom assimilates contemporary
marketing concepts (Perannagari & Chakrabarti, 2020). Due to the scant research in the
domain, little is known about the nature and drivers of online voter relationships. Subsequently,
political brands have limited guidance to rely upon, which hampers the effective application of
social media marketing in politics.
The research project is conceptualized considering these broad gaps in the literature. However,
studies also address different academic gaps at the individual level, which are equally pertinent
to marketing and political marketing.
1.1.2 Relationship marketing
The term relationship marketing is attributed to Berry (1983). Relationship marketing is “all
marketing activities directed toward establishing, developing, and maintaining successful
relational exchanges” (Morgan & Hunt, 1994, p. 22). It gained prominence during the twilight
of the last century, eventually replacing the marketing mix as the dominant paradigm of
marketing (Grönroos, 1997; Grönroos, 1994; Gummesson, 1994). In essence, the shift reflected
a move from transactional marketing to a relationship-based approach to marketing, which
placed customers and customer retention at the heart of marketing practice and research.
Factors like market competition, global competitiveness, a move towards service-based
economies, and technological advancements like computerization aided the rise of relationship
marketing (Sheth, 2017; Palmatier, 2008).
Literature shows that relationship trust, commitment, satisfaction, and quality are the key
variables in relationship marketing, driving various outcomes like profit, referrals, loyalty, and
cooperation (Rooney et al., 2021; Verma et al., 2016; Palmatier et al., 2006). At the micro-
2
level, relationship marketing practices support firms through loyalty cards, direct marketing,
smartphone applications, and customer databases. However, a macro view of relationship
marketing conceives it as an orientation or a way of doing business that sees firms develop
long-term and mutually beneficial relationships with customers. Notably, a relational
orientation is not exclusive to customers but encompasses all the stakeholders in an
organization’s ecosystem (Payne & Frow, 2017).
The vast growth of relationship marketing has resulted in divergence rather than convergence,
prompting calls for greater focus (Sheth, 2017). The literature is not abreast with recent changes
in technology and social media (Rooney et al., 2021; Gummesson, 2017), along with being
excessively reliant on a firm-dominant perspective (Rooney et al., 2021). Experts believe that
relationship marketing needs to be extended to novel contexts and researchers need to discard
surveys and adopt qualitative methods to understand customer relationships (Gummerus et al.,
2017). Presently, there is a need to understand how relationship marketing can win customers’
hearts and cooperation using social media (Sheth, 2017).
1.1.3 Social media marketing
Social media marketing refers to “the utilization of social media technologies, channels, and
software to create, communicate, deliver, and exchange offerings that have value for an
organization's stakeholders” (Tuten & Solomon, 2017, p. 18). Social media marketing is used
for branding, advertising, relationship-building, stimulating word-of-mouth (WOM) or usergenerated content (UGC), and influencing consumer behavior (Alalwan et al., 2017). It is
linked to higher customer spending and profitability, stronger customer relationships, and
greater customer engagement (Clarke et al., 2016; Kumar et al., 2016; Maecker et al., 2016).
There are various mechanisms through which social media marketing occurs (e.g. official
brand communities, official channels, advertising, social media influencers, etc.). This project
focuses on the official social media channels of political brands. These official channels play
a role in retaining customers while increasing loyalty and advocacy (Sashi et al., 2019).
Researchers are particularly interested in social media’s abilities to create social and emotional
ties with customers and facilitate relationship-building (Achen 2017; Alalwan et al., 2017).
Studies show that engagement with a brand’s social media channel leads to strong customer
relationships (Achen, 2017; Clarke et al., 2016). Social media facilitate dialogue and foster a
3
sense of community, making them ideal for online relationship marketing (Steinhoff et al.,
2019). Moreover, social media are ideal for communication, interaction, and value creation,
the three building blocks of the relationship marketing process (Abeza et al., 2020; Grönroos,
2004). Indeed, social media are reviving interest in relationship marketing, which was
dwindling in recent times (Sheth, 2017).
Despite the exponential growth in social media marketing literature, there are gaps in our
understanding of social media-enabled relationship marketing. Primarily, the role of marketergenerated content in strengthening online relationships remains under-researched in the
literature, which predominantly focuses on the effects of social media engagement on
relationships (e.g. Achen, 2017; Clark et al., 2016). Brands need to offer content that fosters
relationships with customers (Steinhoff et al., 2019). This content-led approach to studying
online relationships is missing in the literature. Likewise, the nature and drivers of online
relationships remain unexplored in the literature. Lastly, there is no holistic framework that
guides political marketing managers in building relationships using social media.
1.1.4 Political marketing
Political marketing is a relatively new discipline. It builds upon two distinct academic domains.
It has attracted substantial attention over the last two decades (Perannagari & Chakrabarti,
2019). The decline in party affiliation, ideological politics, and political involvement have
magnified the use and influence of political marketing. Political marketing seeks to “establish,
maintain, and enhance long-term voter relationships at a profit for society and political parties,
so that the objectives of the individual political actors and organizations involved are met”
(Henneberg, 1996, p. 777).
Marketers assert that political candidates can be “marketed as well as soap” (Marland, 2003;
Kotler & Levy, 1969). The cliché is a criticism of political marketing, which is often viewed
as a distraction from meaningful political discourse (Ormrod et al., 2013). Nevertheless, the
societal significance of political marketing is undeniable. The 2019-2020 election cycle (USA)
saw political actors spend over $8.5 billion on TV, Radio, and digital mediums (Homonoff,
2020). Similarly, Australian political parties spent a record AUD 82 million on political
advertising in the 2019 federal elections (Pollitt, 2019). Notably, political marketing is not
4
limited to elections. An increasing number of political actors benefit from political marketing
during government (Joathan & Lilleker, 2020).
Despite the growth, literature in the field of political marketing is fragmented. The reliance on
contemporary marketing concepts is rare and the majority of studies concentrate on the effects
of marketing communication and techniques on electoral and voter outcomes (Perannagari &
Chakrabarti, 2019). This instrumental view of marketing means that political marketing is
overly reliant on obsolete marketing concepts (Henneberg et al., 2009). Few studies incorporate
contemporary marketing paradigms or concepts like relationship marketing, customer
engagement, service-dominant logic, and co-creation. In brief, political marketing does not
reflect the recent changes in marketing practice and thought (Lees-Marshment, 2019;
Perannagari & Chakrabarti, 2019; Ormrod et al., 2013; O’Cass, 2009; Butler & Harris, 2009;
Brennan & Henneberg, 2008).
The role of social media in political marketing came to prominence during Barack Obama’s
presidential campaign in 2008. His campaign was built around new media (Cogburn &
Valenzela, 2011) and earned various accolades like the Advertising Age's marketer of the year
for 2008. Donald Trump and the US presidential election (2016) reignited interest in the
domain, which is set to grow in prominence (Appel et al., 2020). It is particularly relevant to
young voters who are prolific users of social media (Pich et al., 2018; Tiffany & Winchester,
2015). Chapter 2 provides a comprehensive review of the current state of research exploring
political marketing on social media.
1.1.5 Political relationship marketing and social media marketing
Among the various marketing schools of thought, relationship marketing is compatible with
political marketing due to the benefits it offers to politicians, voters, democracy, and society
(Ormrod et al., 2013; Henneberg et al., 2009; Henneberg & O’Shaughnessy, 2009; Vankov,
2013; Bannon, 2005; Scammel, 1999; Egan, 1999). Traditional political marketing increases
voter apathy and detachment, whereas personal and social connections formed through
relationship marketing can lead to higher involvement among voters. Political parties and
politicians need to develop genuine relationships with their constituents to gain votes,
particularly during non-election periods. At the macro level, such an approach benefits society
and democracy, whereas its micro application helps politicians build lasting allegiances with
5
voters rather than merely renting their support during electoral periods (Henneberg &
O’Shaughnessy, 2009). Notably, a relational approach is advised towards all political
stakeholders and not just voters (Hughes & Dann, 2009).
Despite the positive perception of political relationship marketing, empirical research in the
domain is scarce. Interestingly, all articles are recent and embedded in the social media context,
with Dean et al. (2015) being a rare exception. Limited literature reveals that politicians’
relationship-building activities on social media help democracy by increasing political
participation and efficacy (Anim et al., 2019). Social media marketing assists candidates in
building a positive image, resulting in higher voter-candidate relationship equity (Hultman et
al., 2019). Political brands, however, have failed to market in a manner that reflects the
philosophy of relationship marketing (Parsons & Rowling, 2018; Harris & Harrigan, 2015).
Beyond this, our understanding of political relationship marketing on social media is deficient.
For instance, social media-enabled voter relationships remain unexplored beyond Hultman et
al. (2019). Additionally, research needs to ascertain what a relationship-oriented political brand
looks like (Henneberg & O’Shaughnessy, 2009). Consequently, politicians and political parties
have a limited understanding of what they can do to connect with voters and what constitutes
a relational approach to social media.
1.2 Research aims
The primary aim of this project is to understand the nature and drivers of social media-enabled
voter relationships. The secondary aim of the thesis is to extend contemporary marketing
concepts to political marketing. Lastly, the project aims to provide actionable insights that help
political brands develop and sustain relationships with voters. To do so, the project relies on
relationship marketing as the metatheoretical lens across the studies, which is due to its
compatibility with both political marketing and social media marketing. This uniform
theoretical underpinning offers consistency to the project and helps to integrate and cognize
the distinct studies at an aggregate level. Concepts utilized in this project are either indigenous
to relationship marketing (e.g. relationship marketing orientation – Paper 3; Callaghan et al.,
1995), have been studied in the relationship marketing literature (e.g. Elaboration-Likelihood
Model – Papers 4 and 5; Chen & Ku, 2013), or proposed as being pertinent to relationship
marketing (Uses and Gratification Theory - Paper 2; Thaichon et al., 2020).
6
1.3 Research methodology
1.3.1 Research paradigm
The five studies included in the thesis (Papers 1-5, Chapters 2-6) vary in their methodological
underpinnings and approaches. Although the project engages the interpretivist and positivist
paradigms of research, it is primary driven by pragmatism (Morgan, 2007). As per the
pragmatic paradigm of research, studies should focus on outcomes, i.e., research questions, and
produce practically significant knowledge that solves problems. Further, due to the paradigm’s
tolerance of distinct epistemologies, it is ideal for conducting research that relies on a mixedmethods approach, i.e., both quantitative and qualitative methodologies (Fujita, 2018; Doyle
et al., 2009; Cameron, 2009; Morgan, 2007). Therefore, the project can be classified as being
driven by a pragmatic approach. However, this is not true for individual studies like Papers 2
and 3 (Chapters 3 and 4), which rely on the interpretivist paradigm. Unlike sequential
exploratory (Qualitative followed by Quantitative) and explanatory studies (Quantitative
followed by Qualitative), the present thesis may be seen as concurrently timed studies
(Qualitative + Quantitative) that answer various related but distinct research questions
concerning the same phenomenon by using different methodological approaches (Cameron,
2009; Mertens, 2005). The subsequent section elaborates on the various research questions and
the specific research designs used to answer them.
1.3.2 Research questions and designs
How can political brands strengthen relationships with voters on social media?
To answer this broad question, the project relies on a mixed-methods approach, selecting
methods based on their appropriateness and ability to answer the research questions. Therefore,
the project builds upon the strengths of both qualitative and quantitative methodologies. Table
1.1 presents the central questions and research methodologies associated with the five studies
that comprise this thesis. By having a different research question in each study, the project was
able to investigate various pertinent topics and offer diverse perspectives. Primarily, a theorydriven, deductive approach dominated this project. For instance, deductive codes were used to
analyze data in the qualitative studies (Papers 2-3, Chapters 3-4). The approach was feasible
since the project extends marketing concepts to a novel context (Elo & Kyngäs, 2008; Hsieh
& Shannon, 2005). Similarly, the coding of content in Papers 4 and 5 (Chapters 5 and 6) also
adopted a deductive approach. However, an inductive approach was used to code data that did
7
not fit deductive codes in the qualitative studies (Papers 2-3, Chapters 3-4), allowing for the
emergence of novel findings that had conceptual or contextual significance.
Table 1.1 Research questions and designs
Paper 1/Chapter 2
Paper 2/Chapter 3
Paper 3/Chapter 4
Paper 4/Chapter 5
Paper 5/Chapter 6
Research question
What is the current state of research in
political social media marketing?
What is the nature of social media-enabled
voter relationships?
What is the marketing orientation that
political brands adopt towards social
media marketing?
What is the role of marketer-generated
content in driving online voter
relationships?
What are the roles of marketer-generated
content and behavioral engagement in
driving online voter relationships?
Research design
Thematic systematic literature review of
journal articles on political social media
marketing (2011-2020).
Focus groups with young Australian voters
who follow political brands.
Focus groups with young Australian voters
who follow political brands.
An online content analysis (mixed-methods)
of the official Facebook pages of the two
major US political parties.
An online content analysis (mixed-methods)
of the official Facebook pages of the two
major US political parties.
Figure 1.1 presents a map of the research questions and indicates their overall contribution to
the central research question. Paper 1 (Chapter 2), a systematic literature review, is not included
in this map since reviews do not constitute primary research. Moreover, the review is not
specific to political relationship marketing. Papers 2 and 3 (Chapters 3 and 4) rely on focus
groups that were analyzed using a predominantly deductive approach. A qualitative approach
was considered ideal due to the limited research in political relationship marketing (Ormrod et
al., 2013; Sofaer, 1999). Notably, a qualitative approach is encouraged towards both social
media marketing and relationship marketing research (Dwivedi et al., 2020; Gummerus et al.,
2017). The rationale behind the research questions in the two qualitative studies is simple. To
strengthen voter relationships via social media, political brands need to understand the nature
of social media-enabled voter relationships. Similarly, political brands need to understand what
a relationship marketing orientation on social media entails.
8
How can political brands
strengthen their relationships
with voters on social media?
CH3. What is the nature
of social media enabled
voter relationships?
RQ1. What are the
gratifications that drive
voters to follow political
brands?
RQ2. What are the drivers
of social media-enabled
voter relationships?
RQ3. What are the
interactions that underpin
social media-enabled
voter relationships?
CH4. What is the
marketing orientation
that political brands
adopt towards social
media marketing?
CH5. What is the role of
marketer-generated
content in driving online
voter relationships?
RQ1. Do political
brands adopt a
relationship marketing
orientation towards
social media
marketing?
RQ2. Are there any
dimensions that are
not represented in the
RMO framework?
RQ1. What is the effect
of various content cues
on online relationship
quality?
RQ2. Does content
curation negatively
moderate the effects of
content cues on online
relationship quality?
CH6. What are the roles of
marketer-generated
content and behavioral
engagement in driving
online voter relationships?
RQ1. What are the
effects of various
content cues on
engagement and online
relationship quality?
RQ2. Does behavioral
engagement mediate the
effect of content cues
on online relationship
quality?
Figure 1.1 Project map
Papers 4 and 5 (Chapters 5 and 6), which are online content analyses, involved manual coding
of content cues and comments, followed by quantitative analysis using statistical techniques
like multiple regression (Paper 4) and multivariate analysis of variance (Paper 5). It is a
commonly used method in online content analyses (e.g. Tellis et al., 2019; Tafesse & Wien,
2018; Ashley & Tuten, 2015). A quantitative approach was deemed suitable since the
characteristics and effects of political marketer-generated content are frequently researched
topics (e.g. Walker et al., 2017; Colliander et al., 2017). The research questions raised in the
two quantitative studies relate to the central research question, i.e., to build relationships with
voters via social media, political brands need to generate voter engagement and offer content
that cultivates relationships (Steinhoff et al., 2019).
9
1.4 Thesis overview
This section provides brief overviews of the five papers that comprise the thesis (Chapters 26).
Paper 1 (Chapter 2): Political social media marketing: A thematic review, conceptual
framework, and research agenda for the future
The first paper is a systematic literature review of political social media marketing. The review
brings together 66 papers from six databases (EBSCO, Web of Science, ProQuest, Science
Direct, Emerald, and Scopus) and the Journal of Political Marketing. The span of the review
is 2011-2020. The review reveals a growing interest in the domain since 2016. The thematic
analysis shows an overt focus on campaign and electioneering. However, contemporary topics
like branding, relationship marketing, and user-generated content are beginning to permeate
the field. A conceptual framework is derived from the empirical studies included in the review.
Finally, the paper attempts to stimulate research in the field by summing the current state of
the literature, logically organizing the literature, and advising avenues for further research.
Paper 2 (Chapter 3): An exploration of social media-enabled voter relationships through uses
and gratifications theory, psychological contract, and service-dominant orientation
Paper 2 is a qualitative study that relies on focus groups with young voters. It employs the Uses
and Gratifications Theory, psychological contract, and service-dominant orientation to identify
the motivations, drivers, and interactional dimensions underpinning social media-enabled voter
relationships. As per the findings, young voters initiate relationships with political brands to
satisfy social, informational, and entertainment gratifications. Trust and mutual benefit drive
this relationship, but the primary driver is a social and personal connection. Young voters prefer
relational, ethical, and individuated interactions. The study contributes to the literature by
delineating a common social media activity (following political brands) and the subsequent
voter relationship. It suggests that the voter-political brand relationship is a function of the
political brand’s characteristics.
10
Paper 3 (Chapter 4): A relationship marketing orientation in politics: Young voters’
perceptions of political brands’ use of social media
Paper 3 seeks to understand young voters’ perceptions of the marketing orientation adopted by
political brands on social media. The findings reveal that major political brands like national
politicians and political parties do not adopt an approach that can qualify as a relationship
marketing orientation. Rather, it is the smaller political parties and local politicians that adhere
to a relational orientation. These political entities elicit greater trust and educate young voters
on shared values, which helps them bond with young voters. The study finds that young voters
desire education from political brands. It contributes to the literature by inducting customer
education as the eighth dimension of a relationship marketing orientation. The study extends
the concept of relationship marketing orientation to political and social media contexts.
Paper 4 (Chapter 5): Online relationship marketing through content creation and curation
Paper 4 is an online content analysis of the official Facebook pages of the Republican and
Democrat parties of the United States of America. The marketer-generated content posted by
these parties is coded for various cues and characteristics based on the Elaboration-Likelihood
Model. The effects of these cues on online relationship quality, which was manually derived
by coding the comments, are confirmed using multiple regression. The data comprises 100
posts and 29,000 comments. The analysis shows that social cues (the volume of comments and
post popularity) and content cues (visuals and length) have a positive effect on online
relationship quality. Content curation moderated the effects of length, visuals, and interactivity
on online relationship quality. The study contributes to relationship marketing literature by
adopting a granular, content-driven perspective in understanding the drivers of online
relationships. It is perhaps the first study that explores the effects of content curation and
creation.
Paper 5 (Chapter 6): Social media in politics: How to drive engagement and develop
relationships
Paper 5 builds upon the context and approach adopted in Paper 4 and develops a comprehensive
framework that understands the effects of various cues and characteristics of political marketergenerated content on behavioral engagement (likes and shares) and the impact of this
11
engagement on voter relationships. The findings demonstrate that content cues like argument
quality, visual symbolism, emotions, valence, and source credibility drive behavioral
engagement, which has a positive association with online relationship quality. In line with the
Elaboration-Likelihood Model, the study argues that shares reflect a higher degree of
involvement with the marketer-generated content. The study contributes to the literature by
providing a holistic view of relationship development on social media. It offers insights to
political marketers that can help them create engaging content.
1.5 Significance and originality
Billions of dollars are spent on marketing political parties and candidates on social media
(Homonoff, 2020). Likewise, a large segment of society engages with the content that is
generated by political brands on social media (Newman et al., 2017). An understanding of the
domain has implications for political marketers and democracy. As a whole, the thesis makes
several original contributions to political, relationship, and social media marketing. It integrates
the relationship marketing perspective into political marketing, which is dominated by a
transactional perspective (Ormrod et al., 2013). By adopting a relational perspective across the
four latter studies, the research project provides insights into how social media can be used to
build long-term relationships with voters. Second, it attempts to update political marketing by
integrating contemporary marketing concepts. For instance, concepts like the psychological
contract, service-dominant orientation, relationship marketing orientation, and relationship
quality have no occurrence in political marketing, whereas other theories and models like the
Elaboration-Likelihood Model are rarely utilized (e.g. Iyer et al., 2017).
Third, the thesis extends the relationship marketing perspective into social media marketing,
where much exploration is needed regarding the development of online relationships (Steinhoff
et al., 2019). Fourth, the research provides political marketers with an in-depth understanding
of the strategic (orientation) and tactical elements (content) of political relationship marketing
on social media. Fifth, the study demonstrates the two-way transferability of knowledge in
studying political marketing, where marketing contributes to our understanding of politics and
vice versa. This is evinced in the emergence of novel findings that have implications for
commercial and political marketers alike.
12
Besides the aforesaid, Papers 4 and 5 are among the first studies that adopt a content-driven
perspective in understanding relationships on social media. Additionally, the two studies offer
political marketing managers an understanding of marketer-generated content that resonates
with followers. Similarly, Papers 2 and 3 offer a profound understanding of young voters, an
important group of voters that is swayed by social media (Tiffany & Winchester, 2015).
Importantly, the two qualitative studies offer an Australian perspective, which was missing in
the literature. Finally, each paper contributes to the literature by tackling a unique research
problem. For instance, Paper 4 contributes to social media marketing literature by explicating
the distinct roles of content curation and creation, a topic unexplored in prior studies. Similarly,
Paper 1 is the first systematic literature review of articles devoted to the study of political
marketing on social media.
1.6 Thesis structure
The thesis comprises seven chapters. These chapters include a general introduction (Chapter
1), Papers 1-5 (Chapters 2-6), and an integrative discussion of the findings attained from the
five papers (Chapter 7). It is worth mentioning that the individual papers appeared in or are
targeted towards different journals. Since journals adopt different language, formatting, and
structuring styles, minor inconsistencies are present between the individual chapters. Further,
it is worth mentioning that the thesis is being presented as a series of papers, which
inadvertently leads to repetition. For instance, a similar introduction or literature review of
political relationship marketing. However, this is a recognized weakness of adopting this
approach (Graduate Research School, 2016).
13
CHAPTER 2: POLITICAL SOCIAL MEDIA MARKETING: A
THEMATIC REVIEW, CONCEPTUAL FRAMEWORK, AND
RESEARCH AGENDA FOR THE FUTURE
This chapter provides a comprehensive and systematic review of political social media
marketing literature (Paper 1). It is yet to be submitted to a journal. The paper is the most recent
manuscript resulting from this research project. Therefore, the systematic literature review
features publications resulting from this thesis. The paper uses the term political social media
marketing (PSMM) and political marketing on social media interchangeably.
14
2.1 Abstract
Social media have provided an impetus to research across various academic disciplines. We
focus on political marketing and conduct a systematic literature review of journal articles
exploring political marketing on social media. Despite the domain’s growth and practical
significance, a systematic literature review is missing. We adopt the SPAR-4-SLR protocol to
conduct a structured thematic review. Our review spans six databases (EBSCO, ProQuest,
Science Direct, Web of Science, Emerald, and Scopus) and comprises sixty-six journal articles
that were published between 2011-2020. We highlight the key publication outlets, findings,
theories, methodologies, data sources, and contexts. We then describe the various themes that
exist in the literature. We categorize the variables explored in the domain and derive a
conceptual framework from these categories. Finally, we recommend an agenda for future
research that takes into consideration the research gaps highlighted in our review. Through
logical synthesis and presentation of relevant literature, we aim to stimulate research that
expands political marketing.
15
2.2 Introduction
Social media have ushered in an era of significant changes in the political, social, and
commercial spheres of life. Consequently, social media attract academics from numerous
disciplines. This is especially true for politics and marketing. Social media marketing’s role in
politics will continue to rise (Appel et al., 2020). The recent US election cycle (2020) witnessed
political candidates and their political action committees spend $1.6 billion on digital
marketing (Homonoff, 2020). This figure was $22 million in 2008 when Barack Obama won
the presidential election. Allocation of vast resources necessitates a deeper understanding of
politics from the marketing perspective (Lees-Marshment, 2019). Like the disruptive
technologies preceding them, such as the printing press, radio, TV, and email, social media
have changed political marketing (Cacciotto, 2017). Unlike prior communication technologies,
social media allow voters to interact with political brands and other citizens. This makes them
distinct from earlier media and more consequential.
Political marketing is complex because of its diverse origins in marketing and political science,
which are distinct academic domains with their own sets of theories, principles, and visions.
As a result, the existing literature is fragmented and spans various academic domains.
Systematic literature reviews are pertinent in this scenario as they synthesize literature, offer a
holistic understanding, and steer a discipline towards theory building (Dwivedi et al., 2020;
Paul & Criado, 2020), which is very valuable to a dynamic and relatively young field like
political marketing (Ormrod et al., 2013; Patrick & Butler, 2009). Political social media
marketing (PSMM), like other fields of research, can only advance if prior studies are
structured and presented logically (Kumar et al., 2020). Unfortunately, there are no review
articles that synthesize political marketing literature that is devoted to social media. Therefore,
our article has multiple goals. First, we conduct a thematic analysis to identify the themes that
prevail in the literature. Second, we develop a conceptual framework that is derived from the
variables highlighted in our review. Third, in light of the research gaps highlighted in our
review, we propose an agenda for future research.
The article is organized as follows. First, we offer a brief description of social media, political
marketing, and political social media marketing. Second, we explain the systematic review
process, which is built upon the SPAR-4-SLR protocol and best practices highlighted in the
literature (e.g. Paul et al., 2021; Rowley & Keegan, 2020; Paul & Criado, 2020). The
16
subsequent section comprises chronological, contextual, and thematic analyses of the literature.
The discussion section includes the conceptual framework and the proposed research agenda.
We acknowledge the various limitations of our review in the conclusion.
2.3 Conceptual boundaries
2.3.1 Social media
Social media hold a prominent place in politics (Boulianne, 2020; Jungherr, 2016) and
marketing (Dwivedi et al., 2020; Alalwan et al., 2017). Social media are “a group of internet‐
based applications that build on the ideological and technical foundations of Web 2.0, and that
allow the creation and exchange of user-generated content” (Kaplan & Haenlein, 2010, p. 61).
In our study, we focus on five popular social media platforms: YouTube, Facebook, Instagram,
Twitter, and Snapchat. However, the latter does not feature in our review despite its utility as
a political marketing tool (Bossetta, 2018).
2.3.2 Political marketing
Political marketing is an old tradition. “Codifying political marketing could take the discussion
back to Aristotle’s writings on Politics and Rhetoric” (Butler & Harris, 2009, p. 152). The
modern discipline has origins in Kotler and Levy’s (1969) expansionary view of marketing.
Political marketing is “a set of activities, processes, or political institutions used by political
organizations, candidates and individuals to create, communicate, deliver, and exchange
promises of value with voter consumers, political party stakeholders, and society at large”
(Hughes & Dann, 2009, p. 244). A sub-discipline that draws on politics and marketing, political
marketing needs to continually evolve to reflect changes in its parent disciplines (Hughes &
Dann, 2009). Researchers note that this is not the case and a second wave of research is needed
since political marketing does not reflect the dynamism and evolution of marketing (Ormrod
et al., 2013; O’Cass, 2009; Henneberg & O’Shaughnessey, 2009). A recent bibliometric review
of political marketing articles (1996-2018) reveals that political marketing literature is in its
nascence and is fragmented (Perannagari & Chakrabarti, 2020). Finding common grounds
between two distinct academic traditions is not an easy task (Lees-Marshment, 2014; Ormrod
et al., 2013). The diversity of political marketing requires a systematic approach to future
research to prevent duplication of research, build theory, assist young researchers, and identify
research gaps (Paul & Criado, 2020).
17
2.3.3 Political social media marketing
Political social media marketing, or political marketing on social media, refers to the use of
social media to create, communicate, and deliver value for stakeholders (adapted from: Tuten
& Solomon, 2013). The research domain gained traction following Barack Obama’s online
presidential campaign (2008) (Newman, 2016; Miller, 2013). Social media marketing covers
many areas like advertising, branding, eWOM, user-generated content, relationship marketing,
and customer behavior (Alalwan et al., 2017). Unlike prior media, social media provide
politicians with an unfiltered and direct communication channel. Social media are interactive
and voters are influential on social media. Consequently, social media require a different
approach to political marketing than the one utilized on traditional media. Specifically, political
marketing on social media demands a more interactive and relational approach (Cacciotto,
2017; Harris & Harrigan, 2015; Lees-Marshment, 2014).
2.4 Research methodology
Systematic literature reviews (SLRs) are appropriate to synthesize or provide an overview of
an academic domain, develop themes, create conceptual frameworks or models, and propose a
research agenda for the future (Paul & Criado, 2020; Tian et al., 2018). The methodology is
frequently utilized in marketing and management research (Kumar et al., 2020; Rana & Paul,
2017; Tranfield et al., 2003). SLRs are valuable to social media marketing. The
interdisciplinary perspectives, wide-ranging research questions, variety of theories, diversity
of research methods, and the rapidly changing social media landscape mandate regular SLRs
(Dwivedi et al., 2020). Similar concerns hold for political marketing. However, SLRs are
infrequent in political marketing, excluding rare exceptions (e.g. Perannagari & Chakrabarti,
2020; Winther-Nielsen, 2017). Specifically, our review is a domain-based review that can be
further classified as a structured theme-based review (Paul et al., 2011; Paul & Criado, 2020).
These reviews document the various theories, constructs, methods, contexts, and research
themes that exist in the literature, along with offering conceptual frameworks and future
research agendas (e.g. Jiang et al., 2020; Rosado-Serrano et al., 2018; Rana & Paul, 2017).
Our review process follows the instructions laid out in the Scientific Procedures and Rationales
for Systematic Literature Reviews (SPAR-4-SLR) protocol (Paul et al., 2021) and the best
practices that are highlighted in the literature (e.g. Paul & Criado, 2020; Rowley & Keegan,
2020; Snyder et al., 2016). First, the need for such a review was ascertained (Tranfield et al.,
18
2003). There are no prior studies that review the domain, which meets the minimum threshold
of forty articles (Paul et al., 2021). Furthermore, the review is valuable since political social
media marketing will continue to increase in significance and consequence (Appel et al., 2020).
In the assembling stage, the first stage of the SPAR-4-SLR protocol, the research questions and
the review criteria were identified, i.e., decisions pertaining to source type and quality. We
limited the review to articles from academic journals having minimum IF and H-index values
of 0.5 and 15 respectively. We set low threshold values since this is the first review of the
domain and because we chose to exclude conference papers and book chapters. The acquisition
strategy involves decisions regarding databases, search periods, and keywords (Paul et al.,
2021). The selected period reflects the evolution of social media in politics. The time frame is
appropriate since social media were not a significant medium fifteen years ago. Facebook was
not open to the general public, and Twitter and Instagram did not exist. Further, ten years is an
acceptable time frame for SLRs (Paul & Criado, 2020). One generic and five platform-specific
search terms were used to extract relevant articles. Using predefined keywords to extract
literature is a common practice in SLRs (Rowley & Keegan, 2020). The selected platforms are
widely used and relevant to politics. We excluded LinkedIn due to its professional nature.
The selection of databases was driven by prior literature in the field of marketing and
management (e.g. Kumar et al., 2020; Tian et al., 2018). The six databases (ProQuest, EBSCO,
Web of Science, Emerald, Science Direct, Scopus) represent a significant fraction of the
marketing literature. A supplementary search was conducted on the Journal of Political
Marketing’s website. The articles retrieved from non-marketing journals are underpinned in
the political marketing perspective, which was ensured by scanning the keywords, abstract,
and text for marketing terminology (Effing & Spil, 2016). Table 2.1 summarizes the
information pertaining to the identification and acquisition of the articles.
19
Table 2.1 Identification and acquisition
Research objectives
1.
2.
Search strings
Criteria
Scope of research
Type of source
Type of study
Language
Time period
Search parameters
Field of research
Journal quality
3.
1.
2.
3.
4.
5.
6.
Synthesize and systematically present the literature on political social
media marketing
Devise a conceptual framework based on the variables studied in the
literature
Propose an agenda for further research
“political marketing” AND “social media”
"political marketing" AND Facebook
"political marketing" AND Twitter
"political marketing" AND YouTube
"political marketing" AND Instagram
"political marketing" AND Snapchat
Inclusion
Databases (6): EBSCO, Web of Science, Science
Direct, ProQuest, Emerald, Scopus, (also Journal of
Political Marketing)
Peer reviewed journal articles
All
English
1st January 2011 – 31st December, 2020
Search terms appears anywhere in the text (where
available)
• The research is embedded in the political
context
• A complete or partial focus on social
media (including comparative studies)
• The study utilizes a marketing perspective
(such as, appearing in a marketing journal
or the use of marketing or political
marketing in the keywords or abstract)
• The journal impact factor is 0.5 or above
• H-index is 15 or above
• Journal does not require a payment
Exclusion
All other databases
All other sources (e.g. book
chapters, conferences)
No exclusion
All other languages
All other dates
No exclusion
All other studies
All other journals
The arranging stage of the SPAR-4-SLR involves the organization and purification of studies
(Paul et al., 2021). In the organization phase, studies were coded for the country, theory (or
literature review), context, focus, methodology, and data. After removing the duplicates,
weeding out the low quality and predatory journals, and excluding articles that did not meet
the inclusion criteria, a list of fifty-seven articles was finalized. The supplementary search
resulted in nine additional articles. The final tally of articles is sixty-six. The search results are
presented in Table 2.2. They provide an elaborate trail. Notably, we excluded guest editorials
(e.g. Williams, 2017).
20
Table 2.2 Search results
Search string
“political marketing” AND
“social media”
"political marketing" AND
Facebook
"political marketing" AND
Twitter
"political marketing" AND
YouTube
"political marketing" AND
Instagram
"political marketing" AND
Snapchat
Articles selected
Article count (after removing
duplicates)
Supplementary search (JPM)
Final number of articles
Emerald
88
Science
Direct
49
Web of
Science
52
ProQuest
EBSCO
Scopus
236
56
55
58
34
19
206
19
19
49
33
27
192
23
23
34
15
2
107
6
6
23
6
3
61
2
1
6
1
0
11
0
0
6
4
26
8
39
31
57
9
66
The final stage of the SPAR-4-SLR involves theme development and research presentation.
Theme development was undertaken by the first and second authors independently and
compared for consistency. We did not use a pre-determined conceptual framework (Rowley &
Keegan, 2020). However, the themes are based on broad marketing concepts like branding,
marketer-generated content, voter behavior, user-generated content, and relationship
marketing. Similar themes are reported in systematic reviews focusing on social media
marketing (e.g. Alalawan et al., 2017). Several articles feature more than one theme. For
instance, Abid et al. (2019) is included in both political marketer-generated content (MGC)
and political relationship marketing since the article explores the impact of political MGC on
online relationship quality. In addition to the thematic analysis, literature is dissected from
various angles, a standard practice in systematic reviews (Paul & Criado, 2020). Tables and
figures are used to assist the presentation of results (Paul et al., 2021). The articles included in
the review are presented in Table 2.3.
21
Table 2.3 Articles included in the review
Article
Year
Topic
Country
Embedded in
election
Focus
Level of
politics
Social media
Theory or literature utilized
Only social
media (1=yes,
0=no)
1
Methodology
Data collected from
Relationship
tested (1=yes,
0=no)
0
Quantitative
(1=yes, 0=no)
Findings
Abid & Harrigan,
2020
2020
An exploration of the nature and drivers of voter-politician
relationships enabled by social media
Australia
N/A
Candidate and
party
National and
local politics
General
Uses and Gratifications Theory,
psychological contract, servicedominant orientation
Focus groups
Young voters
0
Political parties' Facebook
pages
1
1
Focus groups
Young voters
0
0
1
Interviews and
netnography
Facebook groups and their
moderators
0
0
Social Capital Theory
1
Survey
Young voters
1
1
General
Resource-Based View
1
Survey
Politicians
1
1
Presidential
politics
General
Resource-Based View
1
Survey
Politicians
1
1
Candidate
Local and state
General
Communication tools (public
relations, advertising, media,
social media)
0
Survey
Politicians
1
1
2016
Candidate
Presidential
politics
Twitter
Political humor, inoculation
effect, two-sided messages
0
Experiment
Voters
1
1
USA and EU
N/A
Party
National
politics
General
Data driven marketing, microtargeting
0
Commentary/viewpoint
Secondary data
0
0
Political tweeting and retweeting during and after Presidential
debates
USA
2016
Candidate
Presidential
politics
Twitter
Political tweeting, virality
1
Content analysis
Twitter data and debates
1
1
2020
Evolution of branding into a co-created, technology-driven
phenomenon that allows symbolic expression and relationship
building
USA
N/A
Candidate
National
politics
Twitter and
general
Branding, connective power
1
Conceptual, case-study
Secondary data
0
0
Bode & Dalrymple,
2016
2016
Comparison between social media followers of political
candidates and the general population
USA
2010
Candidate
US Congress
(Congress,
Senate,
Gubernatorial)
Twitter
Opinion leadership, Two-Step
Flow of Information Theory
1
Survey
Followers of political
candidates
1
1
Boerman &
Kruikemeier, 2016
2016
Social media users’ responses to promoted tweets from
political and commercial brands
Netherlands
N/A
Party
National
politics
Twitter
Persuasion Knowledge Model
1
Experiment
College students
1
1
Young voters follow politicians to satisfy informational,
social, and entertainment needs. Social exchanges, trust, and
ethical and developmental interactions drive the online
relationship.
Social cues like the volume of comments and content
popularity and content cues like visuals and length affect
online relationship quality. Curating lengthy and interactive
content has a negative effect on voters’ responses.
Young voters perceive major parties and politicians to be
adopting a sales orientation. The orientation adopted by
local politicians and smaller parties was perceived as
relational.
Facebook groups act as ‘ministry of truth’, keeping an eye
on mainstream media, which is seen as manipulative. These
groups try to expose them. The groups co-create value for
citizens. The BJP (India’s ruling party) adopted an
entrepreneurial approach to digital marketing.
Political parties’ customer relationship-building activities
and visibility on social media result in an increase in
political participation among young voters. The relationship
is mediated by political efficacy (internal and external).
Adaptation and relationship-building capabilities predict a
politician’s use of social media, which in turn affects the
politician’s reputation.
A presidential candidate’s ability to lead and innovate has a
positive impact on social media. The use of social media has
a positive effect on a presidential candidate’s popularity and
citizen loyalty.
The use of public relations and social media have a positive
relationship with political performance and voter
orientation. Traditional media use only affects political
performance. Political advertising has no impact.
The viewers of Saturday Night Live skit mocking Trump
having prior knowledge of Trump’s response to SNL and his
history with SNL viewed Clinton and Pence less favorably.
Knowing Trump’s response to SNL created an inoculation
effect.
American political marketing is driven by the vast collection
and assimilation of data from private sources, party
databases, etc. This is partly due to a constitutional
framework that allows it (First Amendment). Canada and
UK are following suit. However, EU laws and attitudes are
very strict on the kind of voter information that can be
stored. An emphasis on data-driven politics is due to party
de-alignment.
The drivers of political tweets (user-generated content) and
retweets vary during and after a political event (presidential
debates and Obama’s final state of the union address).
During debates, Twitter acts as a medium for narration.
Short tweets are more likely to be shared. After debates,
Twitter acts as an interpretive medium. Longer and detailed
tweets, which include success and achievement-oriented
words along with emotion, are retweeted. Tweets surged
when contentious issues were discussed (e.g. immigration
and abortion). Issues and topics became less important after
the debate.
Authors posit a theory of networked branding. Networked
branding takes into account the growing power of
consumers to transform, develop, and sustain brand value
and meaning. Brands act as spaces where communication
and relationship building occurs. Networked branding takes
into account the multiple actors including non-human actors.
NPS (National Park Service, USA) and Donald Trump serve
as examples of this networked brand culture.
Political Twitter users are not reflective of the US
population (race or gender) and are more politically
involved than average Americans. Active engagement on
Twitter positively impacts online political participation.
Exposure to disagreement on Twitter is negatively related to
online political participation. Twitter disagreement
decreases online and offline participation but increases
political tweeting. Political Twitter reading is negatively
linked to political participation. Perceived social influence
and political motivation are linked to retweeting.
Informational motivation has a negative effect though.
Social media users are less skeptical of political tweets
compared to tweets from brands. However, when they see
the promoted banner, their attitude towards political tweets
Abid et al., 2019
2019
Content that leads to online relationship quality
USA
N/A
Party
National
politics
Facebook
Elaboration-Likelihood Model,
relationship marketing
1
Content analysis
Abid et al., 2020
2020
Relationship marketing orientation of Australian political
brands on social media
Australia
N/A
Candidate and
party
National and
local politics
General
Relationship marketing
orientation
1
Amoncar, 2020
2020
How citizen-led Facebook forums generate/co-create value in
Indian politics
India
N/A
Party
National
politics
Facebook
Entrepreneurial marketing
Anim et al., 2019
2019
The effect of political parties’ social media usage on political
participation of young voters
Ghana
N/A
Party
National
politics
General
Antoniades & Mohr,
2020
2020
The effect of politician’s capabilities on social media use and
consequently political reputation
USA
N/A
Candidate
Local and state
Antoniades & Mohr,
2019
2019
The effect of candidate capabilities on popularity and citizens
loyalty
Cyprus
Presidential
election (year not
specified)
Candidate
Antoniades, 2020
2020
The effect of various communication tools on a politician’s
political performance and voter orientation
USA
N/A
Becker, 2020
2020
Politician’s response to critical humor (Trump’s response to
Alec Baldwin’s SNL skit) and its effect on voters
USA
Bennett, 2016
2016
Will data-driven political marketing practices, which are
prevalent in the USA and Canada, be adopted in the EU?
Berman et al., 2019
2019
Billard & Moran,
2020
22
Buccoliero et al.,
2020
2020
An analysis of Clinton’s and Trump’s tweets to analyze their
strategies
USA
2016
Candidate
Presidential
politics
Twitter
Twitter and politics, 2016
elections
1
Content analysis and
social network analysis
Twitter accounts of
candidates
0
1
Cacciotto, 2017
2017
The evolution of political consultancy in the USA
USA
N/A
N/A
Presidential
politics
General
Political consulting in the USA
0
Commentary
Secondary data
0
0
Calderón-Monge,
2017
2017
The expression of emotions by Spanish voters and political
parties’ management of these emotions
Spain
N/A
Party
National
politics
Twitter
Emotions, social movement,
Twitter
1
Content analysis
Twitter pages of political
parties
0
1
Cameron et al., 2016
2016
Social media's ability to predict election outcomes
New Zealand
2011
Candidate
National
politics
Facebook and
Twitter
Social media and election
predictability
1
Content analysis
Candidates' Facebook and
Twitter accounts
1
1
Cogburn &
Espinoza-Vasquez,
2011
2011
The Obama campaign's use of Web 2.0 technologies in 2008
USA
2008
Candidate
Presidential
politics
General
Web2.0, social movements,
campaigning
0
Case-study
Secondary data
0
0
Colliander et al.,
2017
2017
The effect of Swedish candidates’ self-presentation strategy on
interest in party and intention to vote for the party
Sweden
N/A
Candidate and
party
Parliamentary
politics
Twitter
Self-presentation
1
Experiment
Voters
1
1
Cornfield, 2017
2017
Donald Trump’s social media campaign during invisible
primaries and the first GOP debate
USA
2016
Candidate
Presidential
politics
(preprimaries)
Twitter
Debates and 2016
1
Content analysis
(mixed-method)
Candidates' Twitter and
debates
0
1
Dimitrova &
Bystrom, 2017
2017
The relationship between online/social media political
activities and attendance in Iowa Caucuses
USA
2016
N/A
Presidential
politics
(primaries)
General
Communication channels,
political engagement
0
Survey
Voters
1
1
Elder & Phillips,
2017
2017
The ability of targeted Facebook videos to gain new HispanicAmerican followers
USA
2016
Candidate
Presidential
politics
Facebook
Targeted political communication
1
Content analysis
Facebook videos on Hillary
Clinton's page
1
1
Garcia-Castañon et
al., 2011
2011
Age and ethnicity’s impact on online political activities, and
impact of online political participation on offline political
participation
USA
2008
N/A
Presidential
politics
General
Young voters, minorities, political
mobilization, internet in politics
0
Survey
Registered voters (existing
data)
1
1
Grusell & Nord,
2020
2020
Swedish political party leaders’ self-presentation on Instagram
Sweden
2018
Candidate
National
politics
Instagram
Images in politics, Instagram in
politics
1
Content analysis
Instagram posts of leaders
0
1
Grusell & Nord,
2020
2020
The professionalization and digitalization of campaigns in
Sweden (2010, 2014)
Sweden
2010/2014
Party
National
politics
General
Professionalization, digitalization,
party campaign types
0
Mixed-method
Interviews and surveys with
party officials
0
1
Harmer & Wring,
2013
2013
Targeting of women voters in UK general election 2010
UK
2010
Candidate and
party
National
politics
General
Women voters in the UK politics
0
Case-study
Secondary data
0
0
Harris & Harrigan,
2015
2015
The role social media played in the 2010 UK elections
(focusing on Liberal Democrats campaign in two
constituencies)
UK
2010
Party and
candidate
National
politics
Twitter,
Facebook,
YouTube
Technology and political
relationship marketing
1
Case study (mixedmethod)
Interview with candidates
(and team)
0
1
23
becomes negative. Importantly, users suffer from banner
blindness and hardly notice the promoted label.
The analysis breaks down the two campaigns for posting
time, tweet types, word cloud, themes, etc. The daily tweets
were higher for Hillary Clinton. Clinton’s approach was
issue-dominant, whereas Trump attacked more. Attacks
were engaging for Trump and self-focus was effective for
Clinton. Clinton had a professional social media campaign,
whereas Trump’s social media use was spontaneous
(amateurism).
First stage of political consultancy -1920s to 1950s (radio
and television, reliance on PR and advertising). Second
stage - 1960s to 1980s (television, establishment of political
consulting). Third stage - 1980s to 2000s (television and
internet). Fourth stage - 2012 onwards (fast era of
communication - social media and digital). Political
consulting is becoming highly tech-driven and specialized.
The need for specialist consultants will increase.
The researchers observed expressions of negative emotions
(anger, irony, shame) towards the incumbent party and
mixed emotions towards the opposition party. However,
mostly positive emotions, mainly hope (joy, optimism),
were seen on the pages of new parties. The opposition
managed emotions by getting citizens to participate in future
political proposals (manifesto), whereas governing party
adopted a rational approach.
The number of Facebook friends on election day and the
change in the number of friends in the preceding months can
predict an election’s outcome. However, the effect is
minuscule, which means that social media presence may be
impactful in very close races. Facebook predicted elections
better than Twitter.
Obama’s 2008 campaign relied on new media, with his
website being the central feature. The campaign excelled at
grassroots organization and mobilization using web 2.0
technologies. Facebook was used to organize, Twitter was
used to send news, and YouTube was used to communicate
with the voters. Emails were used for campaign donations.
A within-tweet balancing strategy (front and backstage) is
better on Twitter. The strategy leads to positive outcomes
for the party. Politicians should add a personal touch to their
professional communications.
The study shows that around debates (or campaign events),
the political marketplace (or use of) of keywords increases.
The use of these rhetorical keywords helps define the
context of the event. The more users tweet these keywords,
the favorable it is for the candidate utilizing them. This
worked in Trump’s favor, helping him bypass party elites.
Trump’s strategy included: celebrity feuding, callouts to
legacy media allies, retweeting fan comments, a blunt
vernacular, and confrontational branding.
Visiting candidate websites and following candidates on
social media had a negative impact on attendance in Iowa
Caucus, whereas liking/sharing political content was
positively related to attendance. Posting comments had no
effect. Therefore, active and passive social media usage
have different impacts. Age, attention to the campaign, cable
news, and local TV news had a positive effect, whereas latenight comedy, radio, and posting political comments were
negatively related to the attendance.
Videos using targeting (using Spanish or having Spanish
text in the first five seconds) were more effective in getting
‘new likes’ from Hispanics. Live videos performed better
than produced/edited videos. Videos that included Hispanic
individuals performed better, particularly when Hillary
Clinton was absent from the video.
Online political activity predicts offline political
participation. Young voters (white and minorities (African,
Asian, Latino)) are equally likely to engage in online
political activities. African-Americans are more likely to use
SNSs for political activities. Older white voters are more
likely to engage in online political activities compared to
older members of the minorities.
The study shows that Instagram posts were not depictive of
personalization in politics. However, party leaders’ images
were used frequently. Professional images were preferred
over personal or private images. The latter were used
infrequently. Interactivity was limited.
Sweden campaigns are professionalized and this increased
from 2010 to 2014. Regardless of size or ideology, all
parties were conducting digital campaigns. Social media are
important form of communication within and outside the
party. Relation between professionalization and
digitalization is strongest among catch-all, bigger parties.
Cyber-mums was a coveted segment (UK general election
2010). The targeting of women included marketing on
parenting websites. The leaders of the parties had webchats
and Q&As with users of parenting websites.
The UK general election (2010) was not an internet-driven
election. The use of social media was limited. A systematic
approach towards social media was absent. Twitter was
Hultman et al., 2019
2019
The effect of social media marketing on candidate’s image and
relationship equity
UK
N/A
Candidate
National
politics
General
Branding, Social Identity Theory
1
Survey
Young voters
1
1
Iyer et al., 2017
2017
The effect of WOM/eWOM (from various mediums) on
political attitudes and behavior of older and younger voters
USA
N/A
N/A
N/A
Facebook
Reference group influence,
Elaboration-Likelihood Model,
generational cohorts
0
Experiment
Voters
1
1
Jain & Ganesh, 2020
2020
How Narendra Modi built credibility
India
2014
Candidate
National
politics
General
Credibility, traits, crisis
management, collaboration
0
Case-study
Secondary data
0
0
Jensen & Bang, 2017
2017
The analysis of candidates’ campaign (Facebook) based on
connectivism and populism
USA
2016
Candidate
Presidential
politics
(primaries)
Facebook
Populism, connectivism
1
Content analysis
Facebook campaigns of
Bernie Sanders and Donald
Trump
0
1
Joathan & Lilleker,
2020
2020
A meta-analysis on permanent campaigning
Not specific to a
country
N/A
N/A
N/A
General
Permanent campaigning
0
Meta-analysis
Published studies
0
1
Kalliny et al., 2018
2018
The role of public policy, media, advertising, and social media
in stimulating the Arab revolution
Arab region
N/A
N/A
Revolution and
protest
General
Hofstede's dimensions, media
0
Survey and social
network analysis
Magazines and online
comments
0
0
Kenski et al., 2017
2017
The use of ideological and party labels to demonstrate ingroup
membership
USA
2016
Candidate
Presidential
politics
(preprimaries
and primaries)
Twitter
Social Identity Theory, Reference
Dependence Theory
1
Content analysis
Tweets of candidates
1
1
Lin & Himelboim,
2019
2019
The structure of the political brand communities (candidatecentered) including the social mediators (bridges-hubs)
USA
2016
Candidate
Presidential
politics
(primaries)
Twitter
Brand communities
1
Content analysis and
social network analysis
Twitter activity of
candidates
1
1
Lin, 2017
2017
The ability of candidates’ profile characteristics to predict
election results
Taiwan
Local election
Candidate
Local politics
Facebook
Social media and political
campaigns
1
Content analysis
Facebook pages of
candidates
1
1
Lucarelli et al., 2020
2020
The emergence of cyber political brands and Information
Technology’s role in political brand development
Italy and Czech
Republic
N/A
Party
National
politics
General
Information Technology, culturalpolitical branding
0
Case-study and content
analysis
Data from online sources
0
0
Marder et al., 2018
2018
The role of undesired social-self and visibility of likes
(conspicuous vs inconspicuous) in driving the ‘intention to
like’ a political brand
USA and UK
N/A
Party
National
politics
Facebook
Self-Concept Theory
1
Survey
Voters
1
1
Marquart et al.,
2019
2019
The role of emotional reactions to political messages in driving
political attitudes and behavior
Netherlands
EU election
Party
EU politics
Facebook
Emotions in political
communication and marketing,
Affective Intelligence Theory
1
Experiment
Voters
1
1
24
perceived to be better for relationship building, allowing
engagement and calls to action. Facebook was better for
recruiting students and reaching a wider audience. It enabled
more effective internal information sharing and
collaboration between local party members. Resource
limitations matter. The presidential, federal system might be
more conducive to social media.
Social media marketing (content, social media credibility,
WOM, interaction) impacts the creation of a candidate’s
image, which leads to voter-candidate relationship equity.
Political ideology has no effect on the relationship between
a candidate’s image and relationship equity.
Younger people prefer political WOM through digital/text
channels. Older voters prefer traditional communication.
Younger voters exhibit greater message involvement with
shallow messages (short and brief) and older voters prefer
deeper messages (complex and detailed). Message
involvement mediates their relationships with believability
and voting intentions. When a message is sent by an
acquaintance, the medium matters. However, if it is from
someone in one’s reference group (classmate/coworker), it
does not matter.
The authors build a credibility model using Narendra Modi
(India’s Prime Minister) as a case study. Modi used various
social media tactics to build credibility: #achedin (translated
as good days), globalized language, Modi kurta, Modi
masks (we are all Modi), virtual/3d holographs, queries or
conversations with Modi on social media, high energy
slogans, and speeches in multiple languages on his website.
Trump’s campaign was populist. Although Bernie’s
campaign had a populist element, it was connective. Both
Trump and Sanders were anti-establishment. Bernie
discussed citizen power and recognized differences, whereas
Trump focused on collective identity and strong authority.
The mentions of democracy and citizen power were greater
in the comments on Sander’s posts. The tone of comments
on Sanders’ page was positive, whereas the tone on Trump’s
page was that of fear.
The study highlights the use and importance of social media
for permanent campaigning. Certain social media activities
indicate the use of permanent campaigning. An excess of
personal posts, frequency of publication on social media,
use of interactivity, and posting rate above peer’s average,
indicate permanent campaigning.
Arab magazines do not reflect the Arab culture. The
comments represent Arab culture but the differences with
the Americans are decreasing. Social media, advertising,
etc., created an environment that led to a new consumer
culture, which may have contributed to Arab Spring.
There was limited evidence that candidates used partisan
labeling. Rather than inter and intra-party attacks, candidates
focused on self-advocacy. Outsider candidates were not
more likely to utilize labeling to prove their in-group
membership.
The common social mediators include other candidates,
media, and journalists. News media and journalists are more
prominent in lower-tier candidates’ communities. Brand
communities of trailing candidates show the highest level of
cluster density followed by middle-tier and leading
candidates (same for reciprocity). Social mediators are
different for winning and trailing candidates. Trailing
candidates are likely to turn to social media to gain the
attention of the same institutions that ignored them in MSM.
Social media presence, account type, verification badge, and
the number of fans predicted voter share and electoral
success. Incumbency had an impact. Candidates using
personal Facebook profiles did better than those with pages
and groups. The number of posts had no impact.
Cyber political brands can come into creation differently.
M5S stemmed from a blog of an Italian humorist, Beppe
Grillo. Czech Pirate Party, an existing party, found support
in Czech techno-culture. New political brands are benefiting
from the diffusion of IT.
Proximity to the undesired social-self that results from
liking a political brand is related to social anxiety. This
proximity is negatively linked to the intention to
conspicuously like a political brand, with social anxiety
mediating this relationship. People are more likely to like a
political brand privately (inconspicuously). With diverse
audiences, people are likely to avoid liking a political brand
visibly (explained by context collapse and the Lowest
Common Denominator effect). People with high political
self-consciousness are more likely to prefer secret likes and
get social anxiety. Americans exhibited greater anxiety and
a lower intention to like.
Emotional (not neutral) messages generate emotions.
Emotions mediate the effect of political messages on
attitudes and intentions. Only positive emotion affect
outcomes. Respondents reacted more positively to messages
that reinforce their existing opinions. The results apply to
Milewicz & Saxby,
2013
2013
What drives the use of social media for inbound
communication among US Congressional candidates
USA
2010
Candidate
US Congress
General
Technology Acceptance Model,
Theory of Planned Behavior
1
Survey
Candidates
1
1
Miller, 2013
2013
Are tech-savvy and traditional campaigns a dichotomy? What
are the factors that dictate the selection?
USA
Local election
Candidate
Local and state
politics
General
Technology and social capital,
technology and civic
participation, Obama campaign
0
Case-study
Secondary data/Author's
consultancy data
0
0
Muñoz & Towner,
2017
2017
The use of Instagram by primary candidates in 2016
USA
2016
Candidate
Presidential
politics
(primaries)
Instagram
Visual framing in politics, visual
communication
1
Content analysis
Candidates' Instagram
pages
0
1
Newman, 2016
2016
The lessons that can be learned from 2008, 2012, and 2016
presidential campaigning (mainly Obama campaigns)
USA
2008/2012/2016
Candidate
Presidential
politics
General
Big-data, microtargeting, social
media, branding
0
Viewpoint,
commentary
Secondary data
0
0
Page & Duffy, 2018
2018
Analysis of candidates’ use of visuals on Facebook and
Twitter (public diaries) with respect to credibility
USA
2012
Candidate
Presidential
politics
(primaries)
Facebook and
Twitter
Symbolic Convergence Theory,
credibility, visual rhetoric
1
Content analysis
Posts and tweets of
candidates
0
0
Parsons & Rowling,
2018
2018
Challenges faced by Welsh politicians in implementing a
relationship marketing approach
UK - Wales
N/A
Candidate
Local and
provincial
politics
General
Political relationship marketing
1
Interviews
Politicians
0
0
Paul & Sui, 2019
2019
The extent of emotion used by political candidates (winners)
on Twitter and the response of followers
USA
2016
Candidate
US Congress
(Senate,
Congress)
Twitter
Emotion in politics
1
Content analysis
(sentiment analysis)
Twitter accounts of winning
candidates and user
comments
1
1
Penney, 2016
2016
Why voters engage in viral politics (sharing online videos)
USA
2012
Candidate
Presidential
politics
Twitter
Deliberative democracy, civic
engagement, virality
1
Interviews
Sharers of political content
0
0
Penney, 2017
2017
The official and unofficial digital campaign promoting Bernie
Sanders presidential candidacy
USA
2016
Candidate
Presidential
politics
(primaries)
General
Two-Step Flow of Information
Theory, connective and collective
actions
1
Interviews
Leaders of official and
unofficial campaigns
0
0
Peres et al., 2020
2020
The universality and media compatibility of the political
communication of 61 world leaders
International
leaders
N/A
Candidate
World leaders
Twitter
Politician brand, communication
styles, particularism Versus
universalism, compatibility of
media outlets
0
Content analysis and
sentiment analysis
Tweets and press articles
1
1
Pich et al., 2018
2018
Young voters’ engagement with the Brexit referendum
UK
Brexit Referendum
N/A
National
politics
Twitter
Customer engagement
1
Survey and social
network analysis
Young voters
0
1
25
left-wing and moderate voters as right-wing voters were not
manipulated by the experimental conditions.
Perceived usefulness of social media for inbound
communication, perceived social media ease of use, and
perceived usefulness of social media for outbound
communication impact satisfaction with social media for
marketing communications, which has an impact on a
leader’s intention to use social media for inbound
communication. Perceived subjective norms have an impact
on a leader’s intention to use social media for inbound
communication. The perceived usefulness of social media
for outbound communications has an effect on the intention
to use social media for inbound communication, which is
mediated by satisfaction with social media.
A candidate’s adoption or use of social media depends on
factors like one’s own traits, situation, and audience traits.
Technology-based campaigns do not guarantee success
Local and state-level candidates are attracted to the low
costs associated with social media and enhanced image
among younger voters. However, young voters hardly
participate in state or local elections.
The ideal candidate frame was used more often, with its
statesmanship dimension being more frequent. The
compassion dimension was used by including images of
family and affinity gestures. Populist frame campaigning
focused on large crowds and ordinary people. Filter usage
was infrequent. The text was used, especially in quotes. The
ideal candidates’ statesman dimension gathered more likes.
Differences existed in drivers of likes and comments.
Advanced statistical techniques are being taken over by big
data and customer analytics. The author introduces the
Obama model (the use of social media, microtargeting, big
data, analytics, feedback, testing, etc.).
Visuals were more influential. The interpretive framework
to assess images was built on the actor (hand, face, clothes,
focal point ), setting (locations, distance, props, forms), and
plot framework (activity, objects). Each of these elements
has an impact on credibility (expertise and trustworthiness).
Candidates used visuals differently. Rick Santorum was the
most balanced in highlighting both dimensions of
credibility.
Not all politicians agree that social media are an effective
engagement tool. Even if they think so, they fear a lack of
control which can damage their relationships. There is a lack
of symmetrical, horizontal communication. Politicians do
not understand a wide range of relationship marketing ethos
and have a limited understanding of social media’s ROI.
Neutral and joy appeals were most frequent in candidates’
use of Twitter. The same is true for the responses of the
public. The results show that the public responds with the
same emotion as the one contained in the candidate’s tweet.
This is particularly true for anger and disgust.
Congresswomen received more joy and disgust comments
compared to Congressmen. Republican candidates elicited
more disgust and anger comments, and fewer joy comments
compared with Democrats. Candidates try to cultivate a
positive image on social media, avoiding negative appeals.
The study focused on users who had shared a particular Mitt
Romney video. Respondents who understood the importance
of marketing and volunteerism acted as unabashed
promoters and persuaders for political parties. Some of the
interviewed participants saw their sharing not as an attempt
to persuade but to further the discourse by adding
information or stimulating dialogue.
The effort to promote Bernie Sander’s candidacy was a
hybrid of top-down online campaigning and a digitallyenabled grassroots movement. The official digital campaign
reflected centralized collective action. The unofficial
campaign (run by Occupy Wallstreet veterans) followed a
model of organization-enabled connective action, where
supporters acted as amplifiers of messages or grassroots
intermediaries. Their main role was to make content go
viral. Crowd-sourced community pages, independent of
campaign influence, reflected a model of self-organized
connective action. The role of citizen marketers is becoming
increasingly important on social media.
World leaders adopt a uniform style on Twitter, supporting
the influence of mediatization, which is a driver of
globalization. A high level of universality and media
compatibility was observed across political leaders. Leaders
are similar to each other in their sentiment, topic mixture,
and use of messages. Overall, the sentiment is positive, and
the discussed topics are mainly diplomacy, economy,
corruption, and the Arab world. Minor country and culturebased differences exist in the leaders’ language.
Young voters’ engagement was greater for the Brexit
referendum than traditional elections. The study proposes a
segmentation typology of young voters based on cognitive,
affective, and behavioral engagement. Young voters were
knowledgeable and actively following politicians on
Ryoo & Bendle, 2017
2017
The topics and themes used by 2016 primary candidates in the
USA and changes in these themes over time
USA
2016
Candidate
Presidential
politics
(primaries)
Facebook and
Twitter
Twitter and politics
1
Content analysis
Candidates’ Twitter and
Facebook pages
0
1
Safiullah et al., 2017
2017
Does social media buzz (number of Tweets) predict election
results?
India
2014
Party
National
politics
Twitter
Buzz marketing
1
Content analysis
Number of tweets on
Twitter
1
1
Schneiker, 2019
2019
The branding of Donald Trump
USA
2016
Candidate
Presidential
politics
(primaries)
Twitter
Celebrity politicians, political
branding
1
Content analysis
Donald Trump's Twitter
page
0
0
Sherman et al., 2012
2012
Young voters’ trust in politicians and information sources like
media, social media, and family.
USA
2008
N/A
Presidential
politics
General
Young voters, trust in social
media
0
Survey
Young voters
1
1
Shmargad, 2018
2018
Influencer politics (influential politicians and politicized
influencers)
USA
2016
Candidate
US Congress
(Senate,
Congress)
Twitter
Influencer politics
0
Content analysis and
social network analysis
Congressional candidates
and users who retweet them
1
1
Temple, 2013
2013
The role played by media in shaping UK general election
UK
2010
Candidate and
party
National
politics
General
UK General Election (2010)
0
Case-study
Secondary data
0
0
Towner & Dulio,
2011
2011
The effect of various web 2.0 tools (candidate website,
YouTube, ABC news) on trust in government, politician, and
performance of internet
USA
2008
Candidate
Presidential
politics
YouTube and
Facebook
Web 2.0 and politics
0
Experiment
Young voters
1
1
Towner & Dulio,
2012
2012
Analysis of US presidential election (2008) and prediction for
2012
USA
2008/2012
Candidate
Presidential
politics
General
New media effect, Obama
campaign
0
Case study, viewpoint
Secondary data
0
0
Towner & Muñoz ,
2018
2018
Media consumption and political participation among baby
boomers
USA
2012
Candidate
Presidential
politics
Facebook,
YouTube, and
Twitter
Political communication effects,
message-response involvement
0
Survey
Baby boomers
1
1
Tsvetkova, 2020
2020
Obama and the practices of public diplomacy
USA
N/A
Candidate
International
General
Public diplomacy, strategic
communication
0
Commentary/viewpoint
Secondary data
0
0
Veenstra, 2017
2017
The formation of subgroups during primaries and the
reconciliation of intraparty divisions before the general
election.
USA
2016
Candidate and
Party
Presidential
politics
(primaries)
General
Hostile media effect, Social
Identity theory
0
Survey (multiple)
Voters
1
1
Vesnic-Alujevic &
Bauwel, 2014
2014
The use of YouTube campaigning in EU elections
France/Italy/
Ireland/Slovenia
2009
Party
EU Parliament
YouTube
Political advertising, Web 2.0,
YouTube
1
Content analysis
Videos posted by political
parties
0
1
26
Twitter. Young voters were active in the lead to Brexit.
Young voters’ engagement is multifaceted.
Candidates favored broadcasting their own messages and
adopting a sales orientation while offering limited
interactivity. The Democrats (Bernie Sander and Hillary
Clinton) had a higher level of activity compared to Donald
Trump. The activity declined when candidates dropped out.
Variance in tweet word count was significantly higher for
Trump denoting a more natural use of social media.
Variations in the reliance on specific themes were witnessed
between and within parties.
Social media buzz can predict parliamentary seats won by
different parties in the general elections. Social media are
gaining importance in Indian politics. Politicians are not
well versed in social media so IT cells provide training.
Donald Trump is an amalgamation of two kinds of celebrity
politicians: superstar political celebrity (one-way
communication, stage events, strong and decisive, special,
unique) and everyday political celebrity (two-way
communication, gaffes, ordinary, anti-establishment).
Trump’s new kind of celebrity politician brand is termed
‘superhero anti-politician celebrity’. This kind sees the
problem, knows how to solve it, has god-gifted abilities to
solve the problem, solves the problem, and breaks the rules.
Young voters’ trust in politicians ranged between negative
and neutral. Social networking sites were not considered
reliable sources of information. Information from immediate
family, friends, press, television, politicians’ websites, etc.,
was considered more trustworthy compared to SNS.
Hierarchy of trusted sources of voting influence was (high to
low): intimate sources, traditional media, political sources,
new media sources.
Influential politicians are those that receive many retweets
while politicized influencers are Twitter users who engage
with politicians’ messages and receive many retweets
themselves. Politicized influencers do not have any official
role in the campaign (everyday makers). For richer
candidates and incumbents, receiving many retweets is
associated with higher vote percentages. For poorer
candidates and challengers receiving retweets from highly
retweeted users is associated with higher vote percentages.
Better-off candidates should thus strive to be influential
politicians, whereas worse-off candidates should aim to get
engagement from politicized influencers.
The UK general election (2010) was not a social media
affair. Politicians were not convicted in their social media
use. The authors refer to social media as the dog that did not
bark. Social media’s role was responsive like spoofing of
slogans and posters.
The viewers of online videos, on the YouTube channel
YouChoose 2008, had low trust in government. The use of
ABC News’ politics page was linked to greater trust in
internet sources and higher perceived performance of
internet sources in regard to political coverage. Facebook
use was linked to political efficacy. The use of candidates’
websites led to a higher perceived performance of internet
sources in regard to political coverage. The study
demonstrates that different technologies, falling under the
web 2.0 umbrella, have distinct effects.
The study highlights the Obama campaign’s focus and
success on new media. It predicts how new media, mobile
marketing, SMS, etc., will interact with politics. It forecasts
greater personalization.
The use of Facebook, Twitter, and YouTube has a limited
impact on the offline political participation of baby
boomers. Websites, however, are effective. Campaign
consumption on Facebook (and to a limited extent
YouTube) is linked to online political participation.
Websites are the most effective indicator of online political
participation. Attention to traditional media does not have
any positive effect on offline and online political
participation. Political interest, gender, race, and party
identification predict various forms of offline and online
political participation.
Barack Obama exploited political marketing in public
diplomacy and introduced social media to foreign policy,
diplomacy, and public diplomacy. Obama connected with
foreign citizens directly using social media.
Identity centrality and candidate favorability predicted
greater perceived bias against one’s candidate. Perceived
network homogeneity influences media bias perception
against one’s party during the general election. The study
suggests that users are source-blind.
YouTube videos were longer than TV ads. The use of a
narrator decreased along the right-left continuum.
Candidates were prominent (some personalization) in many
videos but issues were the main focus. Production
techniques varied. Talking heads were more prominent in
less professionalized countries (France (28%), Italy (40%),
Ireland (60%), Slovenia (100%). Most videos attempted to
inform and mobilize.
Walker et al., 2017
2017
What makes tweets go viral?
UK
2015
Candidate
National
politics
Twitter
Communication effects,
Transmission Model of
Communication
1
Content analysis
Twitter data of MPs
1
1
Wallsten, 2011
2011
Online video sharing among bloggers
USA
2008
N/A
Presidential
politics
Blogs and
YouTube
Political blogging
0
Content analysis
Blogs and online videos
0
1
YLI-KAITALA,
2014
2014
The American influence on the Egyptian revolution exerted
under its public diplomacy strategy
Egypt and USA
N/A
N/A
Revolution and
protest
General
Public diplomacy, political
empowerment, democracy
promotion
0
Case-study
Secondary data
0
0
27
Media (picture/video) and hashtags are the primary
structural drivers of retweets. Source characteristics like
campaign tweets/day, the number of accounts followed, and
total tweet count also drive retweets. Attack, supporting
others, the campaign trail, personal, and position-taking are
the more viral content themes. The number of followers,
sentiment, total campaign tweets, and percentage majority in
prior election also drive retweets. Negative tweets are more
likely to be retweeted. Combining fear and attack or
combing support for others and campaign trail increase
retweets substantially.
Political bloggers were ideologically driven and shared any
video that portrays the opposition candidate in a negative
light. Negative valence dominated the blogosphere. Political
bloggers did not act as media watchdogs. Bloggers seldom
linked to videos from prominent news sources.
Actors trained by American-funded programs to empower
citizens politically and build goodwill/connections with the
middle-east public played significant roles in the Egyptian
revolution.
2.5 Findings
We summarize the chronological evolution of political marketing research devoted to social
media. In doing so, we highlight the progress made over the last decade. Figure 2.1 highlights
the number of articles published yearly. A special issue on the topic explains the high article
frequency in 2017 (Williams, 2017).
2.5.1 A chronological review
Article frequency
18
16
14
12
10
8
6
4
2
0
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Figure 2.1 Political social media marketing (2011-2020)
2011-2015 (13 studies)
Political social media marketing gained prominence following Barack Obama’s presidential
bid in 2008. Consequently, Obama’s campaign features in several studies during this time (e.g.
Miller, 2013; Cogburn & Espinoza-Vasquez, 2011). Studies during this era primarily rely on
the case-study methodology to explore the role of social media in elections or campaigns (e.g.
Harris & Harrigan, 2015; Harmer & Wring, 2013; Temple, 2013). Few studies test
relationships between variables (Milewicz & Saxby, 2013; Towner & Dulio, 2011; GarciaCastañon et al., 2011), and almost all studies focus on the USA and the UK. Trust and political
participation interested scholars during these years (e.g. Sherman et al., 2012; Towner & Dulio,
2011; Garcia-Castañon et al., 2011).
2016-2020 (54 studies)
Sophisticated methodologies, reliance on big data (e.g. Berman et al., 2019), dependence on
theory (e.g. Colliander et al., 2017), and integration of marketing concepts (e.g. Pich et al.,
28
2018) have increased over the last five years. Publication outlets have increased from two in
2015 to twenty-two (Table 2.4). Donald Trump and the US presidential election (2016) have
revived the interest in the domain, which was declining. Finally, the literature has grown to
include various geographical contexts.
Table 2.4 Publication outlets
Journal
Marketing journals
Journal of Political Marketing
Psychology and Marketing
European Journal of Marketing
Journal of Consumer Marketing
International Journal of Market Research
Journal of Marketing Research
Journal of International Marketing
Journal of Strategic Marketing
Australasian Marketing Journal
Marketing Intelligence and Planning
Journal of Marketing Communications
Journal of Research in Marketing and Entrepreneurship
Journal of Promotion Management
Non-marketing journals
Society
Computers in Human Behavior
International Data Privacy Law
Convergence: The International Journal of Research into New Media Technologies
Journal of Communication
Asia Pacific Management Review
Management Research Review
Political Studies Review
Media, Culture & Society
Frequency
37
3
2
2
2
1
1
1
1
1
1
1
1
3
2
1
1
1
1
1
1
1
2.5.2 Research settings
Geographic
The focal point of the literature is the US, which features in thirty-six studies. This is followed
by studies based in the EU (9), UK (7), Asia (4), Australia/New Zealand (3), and Africa (2).
Three studies explore the US in relation to or in comparison with the UK, EU, and Egypt.
Social media platforms
Twenty-nine studies in our review discuss social media in general. Twitter appears in nineteen
studies. Facebook, Instagram, and YouTube feature in eight, two, and two articles respectively.
Three articles focus on Facebook and Twitter, two studies examine Facebook, YouTube, and
Twitter, and one study investigates Facebook and YouTube.
29
Comparative studies
Several studies in our selection take a comparative perspective. Comparative analyses focus on
various aspects like comparisons between different media (Antoniades, 2020; Sherman et al.,
2012), young and old voters (Iyer et al., 2017), American and British voters, (Marder et al.,
2018), political and commercial brands (Boerman & Kruikemeier, 2016), political candidates
and campaigns (e.g. Buccoliero et al., 2020; Peres et al., 2020; Vesnic-Alujevic & Bauwel,
2014), brand communities of candidates (Lin & Himelboim, 2019), social media followers of
politicians and regular citizens (Bode & Dalrymple, 2016), characteristics of user-generated
content during and after political events (Berman et al., 2019), and the US and EU laws
governing political marketing on social media (Bennett, 2016).
Elections
Roughly three-fourths of the articles are embedded in campaigns, lead-up to elections, and
elections. Particularly, the US presidential elections of 2016 (14), 2008 (7), 2012 (5), and the
UK general election of 2010 (3) are examined frequently. Most studies explore elections at the
national level, however, Congress, state, and local elections are considered in several studies
(e.g. Antoniades & Mohr, 2020; Lin, 2017; Milewicz & Saxby, 2013).
Voter segments
Young voters are frequently researched in the literature. They have low trust in government
and politicians (Sherman et al., 2012; Towner & Dulio, 2011) but hold favorable perceptions
of the manner in which minor political parties and local politicians use social media (Abid et
al., 2020). Political marketing on social media leads to stronger relationships with young voters
and increases their political efficacy (Anim et al., 2019; Hultman et al., 2018). Young voters
prefer brief eWOM from their peers (Iyer et al., 2017) and desire personal and social content
from politicians (Abid & Harrigan, 2020). They are generally less engaged in traditional
elections (Pich et al., 2018).
Other segments investigated in the literature include minority voters (Elder & Phillips, 2017;
Garcia-Castañon et al., 2011), aged voters (Towner & Muñoz, 2018; Iyer et al., 2017), and
female voters (Harmer & Wring, 2013).
30
2.5.3 Themes
Table 2.5 Research themes
Theme
PSMM during elections/campaigns
General approach towards PSMM
Adoption of social media for political activities
Social media’s effect on voters’ behaviors
Social media’s ability to predict elections
Political marketer-generated content
Social media and political relationship marketing
Social media and political branding
User-generated content
Digitalization and professionalization
Sample article
Cogburn & Espinoza-Vasquez, 2011
Muñoz & Towner, 2017
Milewicz & Saxby, 2013
Dimitrova & Bystrom, 2017
Cameron et al., 2016
Walker et al., 2017
Harris & Harrigan, 2015
Billard & Moran, 2020
Berman et al., 2019
Grusell & Nord, 2020
Political social media marketing during elections/campaigns
Studies in this theme dissect the strategies of political parties and candidates or the extent and
manner of their social media use during elections. Several articles explore Barack Obama’s
groundbreaking use of new media in 2008 (Newman, 2016; Towner & Dulio, 2012; Cogburn
& Espinoza-Vasquez, 2011) and Donald Trump’s Twitter-savvy strategy (Buccoliero et al.,
2020; Schneiker, 2019; Cornfield, 2017; Jensen & Bang, 2017). While Obama executed a
grassroots movement that connected like-minded voters via social media and his website,
Trump was able to use social media to bypass party elites and cultivate an authentic brand.
Studies outside the US explore social media marketing during the UK (Harris & Harrigan,
2015; Harmer & Wring, 2013; Temple, 2013) and Indian elections (Amoncar, 2020; Jain &
Ganesh, 2020), with social media playing a significant role in the Indian context where Prime
Minister Modi and his Bhartiya Janta Party were able to use social media effectively. The UK
general election (2010), however, showed that Obama’s campaign did not trigger an instant
adoption of political social media marketing in the UK.
General approach towards political social media marketing
Our review shows that politicians do not adopt an interactive or relational approach on social
media and utilize them in a traditional manner, i.e., for political broadcasting, self-promotion,
and self-advocacy (Grussel & Lord, 2020; Kenski et al., 2017; Ryoo & Bendle, 2017; Harris
& Harrigan, 2015). Politicians tend to inform and mobilize through political marketing
(Vesnic-Alujevic & Bauwel, 2014) and their posting frequencies increase or decrease based on
campaign needs (Ryoo & Bendle, 2017). Moreover, the distinctions between politicians are
diminishing as social media mature, with a vast majority of politicians adopting a similar
31
communication style, self-presentation strategy, visual framing, and emotional appeals (Peres
et al., 2020; Grusell & Nord, 2020; Paul & Sui, 2019; Muñoz & Towner, 2017).
A few studies acknowledge differences also (Buccoliero et al., 2020; Page & Duffy, 2018). For
instance, Clinton’s social media marketing was professional compared to Trump’s spontaneous
‘amateurism’ (Buccoliero et al., 2020). Furthermore, candidates are increasingly turning to
social media for political marketing during government (Joathan & Lilleker, 2020). For
example, Obama’s innovative use of social media during government benefitted him
domestically (Cogburn & Espinoza-Vasquez, 2011) and internationally (Tsvetkova, 2020).
Adoption of social media for political activities
Candidates: Factors like the candidate’s age, target market’s age, and the type of election
dictate the adoption and usefulness of social media marketing (Miller, 2013). Further, a
candidate’s adaptation, relationship-building, leadership, and innovation capabilities influence
their adoption of social media (Antoniades & Mohr, 2020; Antoniades & Mohr, 2019). Other
factors that are pertinent to the adoption of social media include the digitalization of national
politics and the candidate’s education and understanding of social media (Grusell & Nord,
2020; Safiullah et al., 2017). Besides politicians’ general adoption of social media, their
adoption for eliciting inbound communications has also been investigated (Milewicz & Saxby,
2013).
Voters: Political interest, gender, race, and party identification determine the general political
use of social media (Towner & Muñoz, 2018; Garcia-Castañon et al., 2011). Specific social
media activities, however, vary in the factors driving them. Following, for instance, is driven
by gender (male), higher income, race, and college education, whereas political tweeting is
driven by low education, disagreements, political ideology, and political motivations (Bode &
Dalrymple, 2016). Sharing political videos is primarily driven by personal motivations,
political motivations, and political ideology (Penney, 2016; Wallsten, 2011). Finally, the
research shows that some platform-based factors like the visibility of likes on Facebook inhibit
engagement with political content (Marder et al., 2018).
Social media’s effect on voters’ behaviors
Research demonstrates that political marketing on social media has an impact on online and
offline political participation (e.g. Anim et al., 2019; Towner & Muñoz, 2018; Dimitrova &
32
Bystrom, 2017; Garcia-Castañon et al., 2011). Although, the effect is more pronounced when
voters are active followers (liking and sharing) rather than passive (Dimitrova & Bystrom,
2017). Social media also facilitate grassroots activism and political protests, an ability
witnessed during the Arab Spring (Kalliny et al., 2018; Yli-Kaitala, 2014).
Social media and their ability to predict elections
The predictive capabilities of social media have also received attention (Shmargad et al., 2018;
Lin, 2017; Safiullah et al., 2017; Cameron et al., 2016). These studies show that various social
media-based indicators like the number of Facebook friends, changes in Facebook friends preelection, retweets, account type, and verification badge can be used to predict electoral results
(Shmargad et al., 2018; Lin, 2017; Cameron et al., 2016).
Political marketer-generated content
The content posted by politicians and political parties is the subject of several studies in our
review. Whereas some studies characterize the content posted by politicians or political parties,
such as the level of personalization, production techniques, emotional appeals, themes, word
count, communication styles, and credibility cues used in the content (Peres et al., 2020;
Grusell & Nord, 2020; Paul & Sui, 2019; Page & Duffy, 2018; Ryoo & Bendle, 2017; Muñoz
& Towner, 2017; Vesnic-Alujevic & Bauwel, 2014), other studies explore the effects of various
content characteristics on content virality and voter outcomes. The virality of tweets, for
instance, is dependent upon various structural elements of the tweet, source characteristics,
sentiment of the tweet, and its content (Walker et al., 2017). Emotional content drives favorable
attitudes and behavioral intentions (Marquart et al., 2019). Lastly, certain factors affect the
composition of political marketer-generated content. Gender (Paul & Sui, 2019), country or
culture (Peres et al., 2020), and party or individual characteristics are some of the few variables
discussed in the literature (Buccoliero et al., 2020; Abid et al., 2019).
Social media and political relationship marketing
Relationship marketing is the only feasible orientation towards political marketing on social
media (Harris & Harrigan, 2015). It increases political participation among citizens (Anim et
al., 2019). Qualitative studies in the domain explore the extent to which politicians practice
relationship marketing on social media (Abid et al., 2020; Harris & Harrigan, 2015), the factors
hampering the application of relationship marketing (Parsons & Rowling, 2018), and the nature
of social media-enabled voter relationships (Abid & Harrigan, 2020). The latter study
33
highlights that young voters desire a personal and social relationship with political brands,
whereas the former studies conclude that politicians and political parties do not adopt a
relational orientation. However, local politicians and minor parties enjoy a positive perception
(Abid et al., 2020). Quantitative studies confirm that certain social media marketing activities
and content cues influence relationship equity and relationship quality respectively (Hultman
et al., 2019; Abid et al., 2019).
Social media and political branding
Donald Trump’s branding strategy has received substantial attention (Billard & Moran, 2020;
Schneiker, 2019; Newman, 2016). Studies highlight that social media are transforming political
branding, which is evolving into a co-created, technology-driven phenomenon as demonstrated
by the rise of Donald Trump and ‘cyber political brands’ in the EU (Billard & Moran, 2020;
Lucarelli et al., 2020). The research on political brand communities shows that communities
devoted to lower-tier candidates are dense and exhibit greater reciprocity (Lin & Himelboim,
2019). Further, brand communities play an important role in political co-branding (Billard &
Moran, 2020). Other studies demonstrate that the behavior towards political brands on social
media is different from that towards commercial brands (Marder et al., 2018; Boerman &
Kruikemeier, 2016).
User-generated content
Like marketer-generated content, user-generated content (UGC) is also a frequently examined
topic. The virality of user-generated content is determined by certain content characteristics
like the tweet’s surface features, linguistic style, emotion, and topic (Berman et al., 2019).
Studies show that young voters prefer brief user-generated content on social networking sites
(Iyer et al., 2017).
The nature of user-generated content is affected by a politician’s gender, party, and the nature
of marketer-generated content itself (Paul & Sui, 2019; Jensen & Bang, 2017). For instance,
Congresswomen receive more comments with joy, whereas Republican candidates prompt
more disgust and anger. Emotional MGC leads to emotional UGC (Paul & Sui, 2019). Studies
examining politician’s and political party’s responses to user-generated content recommend
proactive behavior (Becker, 2020; Calderón-Monge, 2017). Besides user-generated content,
users’ roles as citizen marketers (Penney, 2017) and gatekeepers of citizen-led Facebook
newsgroups (Amoncar, 2020) are also discussed in the literature.
34
Digitalization and professionalization
Lastly, a few studies highlight that modern politics is becoming professionalized and
digitalized. This is the case in the US (Caciotto, 2017) and Europe (Grusell & Nord, 2020).
Political consultancy is becoming tech-driven, creating an immense need for specialist
consultants. In India, political parties are creating IT cells and hiring advertising agencies to
train and assist politicians, or even post content on their behalf (Safiullah et al., 2017).
2.6 Discussion
The review shows that PSMM is gathering significant momentum. Contemporary marketing
paradigms are finding a place in PSMM. However, the literature still lacks depth and is
fragmented, which prevents focus and theory building. Like similar domains, such as political
campaigning and political participation, PSMM can benefit from a systematic arrangement of
literature (Boulianne, 2020; Jungherr, 2016).
2.6.1 Conceptual framework
Besides describing the prevalent themes, we identify, organize, and categorize the numerous
variables or factors that are studied or proposed in the literature (see Table 2.6). Based on these
categories and the context in which the variables are studied, we propose a framework that
offers a visualization of the current state of research and the political marketing process on
social media.
35
Table 2.6 Details of variables included in the conceptual framework
Factors that influence if and how politicians or political parties use social media
Drivers of political social media marketing use or
• A politician’s leadership, adaptation, relationship, and innovation capabilities (Antoniades
adoption among politicians
& Mohr, 2020; Antoniades & Mohr, 2019)
• A politician’s satisfaction with social media, its use of ease, and subjective norms
(Milewicz & Saxby, 2013)
• Candidate’s age, voters’ age, and low-cost (Miller, 2013)
• Social media’s relational use is driven by their perceived ROI, perceived loss of control
(inhibits), engagement, and financial and human resources (Parsons & Rowling, 2018)
• Professionalization and digitalization of politics (Grusell & Nord, 2020)
• Candidate’s education and understanding of social media (Safiullah et al., 2017)
Factors that influence how politicians/political
• Ethnicity, gender, party, incumbency, and competitiveness of race (Paul & Sui, 2019)
parties use political social media marketing
• Campaign period v/s non-campaign period and time in the campaign (Ryoo & Bendle,
2017)
• Candidate-based or party-based differences (Buccoliero et al., 2020; Abid et al., 2019;
Jensen & Bang, 2017)
• Country and culture (Peres et al., 2020)
• Level of personalization or candidate focus in politics and level of professionalization
(Grusell & Nord, 2020; Vesnic-Alujevic & Bauwel, 2014)
• Minor v/s major parties (Abid et al., 2020)
• Candidate tier (Lin & Himelboim, 2019)
• Level of politics (local/national) (Abid et al., 2020)
Factors that have an effect on voters and are controlled by politicians or political parties
Marketer-generated content
• Interesting, informational, and useful content (Hultman et al., 2019)
• Targeting in video and live v/s edited/produced videos (Elder & Phillips, 2017)
• Balanced self-presentation (Colliander et al., 2017)
• Certain themes (e.g. attack, personal, and position taking), the sentiment of content, links,
and media (Walker et al., 2017)
• Emotion in content (e.g. fear and anger) (Marquart et al., 2019; Walker et al., 2017)
• Valence and visuals (Abid et al., 2019; Walker et al., 2017)
• Length and created v/s curated content (Abid et al., 2019)
• Visual framing of the candidate (Muñoz & Towner, 2017)
• Keywords (Cornfield, 2017)
• Time of posting (Buccoliero et al., 2020)
• The volume of comments and content popularity (not controlled by the political brand but
are MGC characteristics that affect its reception) (Abid et al., 2019)
36
• Number of followers (Cameron et al., 2016)
• Change in followers (before elections) (Cameron et al., 2016)
• Verification badge (Lin, 2017)
• Account type (Lin, 2017)
• Total posts (Walker et al., 2017)
• Posts/day (Walker et al., 2017)
• Number of accounts followed (Walker et al., 2017)
Social media presence and visibility
• Presence (Lin, 2017),
• Visibility (Anim et al., 2017),
• Use of social media (Antoniades & Mohr, 2019)
Social media marketing orientation
• Interactive (Hultman et al., 2019)
• Relational v/s sales-oriented (Abid et al., 2020; Anim et al., 2017)
Official brand communities
• Official brand communities (Penney, 2017)
Political advertisements (promoted content)
• The promoted banner has a negative impact (Boerman & Kruikemeier, 2016)
Factors that are not controlled by politicians and political parties
Unofficial brand communities
• Unofficial and crowd-sourced brand communities (Billard & Moran, 2020; Penney, 2017)
Social media ecosystem
• Political influencers (Shmargad et al., 2018)
• Political bloggers (Wallsten, 2011)
• Citizen-marketers (Penney, 2017)
• Social mediators (bridges and hubs) (Lin & Himelboim, 2019)
• A candidate’s direct and indirect communities on social media (Lin & Himelboim, 2019)
Political UGC and eWOM
• Certain political UGC characteristics like surface features, linguistic style, emotion, and
drive increase virality (Berman et al., 2019)
• Brevity/detail and intimacy with the source of eWOM has an impact on its reception (Iyer
et al., 2017)
• Social media buzz can predict election outcomes (Safiullah et al., 2017)
Factors that drive voters towards political use of social media
Drivers of social media use for political activities
• Age, political interest, gender, race, and party identification drive online political
participation (Towner & Muñoz, 2018; Garcia-Castañon et al., 2011)
• Twitter use, media variables (use and trust in media), education, political interest, and
political knowledge drive online political participation (Bode & Dalrymple, 2016)
• Perceived social influence, political motivation, and disagreements drive active political
tweeting (Bode & Dalrymple, 2016)
• Entertainment, informational, social, and instrumental gratifications drive voters to follow
political candidates (Abid & Harrigan, 2020)
• Customer engagement level (Pich et al., 2018)
Profile page
37
Factors that have an effect on the reception of content
Moderators
•
•
•
Mediators
Variables studied as outcomes
Electoral outcomes
Virality
Offline political outcomes
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
User’s consumption: active (like, share) vs passive (following/reading) (Bode &
Dalrymple, 2016)
User’s perceived social network homogeneity (Veenstra et al., 2017)
User’s political self-consciousness, the conspicuousness of liking on Facebook, and social
anxiety (Marder et al., 2018)
User’s personal/political motivations (Penney, 2016; Wallsten, 2011)
User’s perceived social influence (Bode & Dalrymple, 2016)
User’s prior knowledge (Becker, 2020)
User’s political ideology (Marquart et al., 2019)
Emotional reaction to message (Marquart et al., 2019)
Candidate image (Hultman et al., 2019)
Message involvement (Iyer et al., 2017)
Persuasion knowledge (Boerman & Kruikemeier, 2016)
Political efficacy (Anim et al., 2017)
Liking the communication (Colliander et al., 2017)
Number of votes (Cameron et al., 2016)
Election outcomes (Lin, 2017)
Percentage of the vote (Shmagrad et al., 2018)
Number of parliamentary seats (Safiullah et al., 2017)
Retweets (Walker et al., 2017)
Likes from new followers (Elder & Phillips, 2017)
Intention to like (Marder et al., 2018)
Tweet and retweets (UGC) (Berman et al., 2019)
Offline political participation (Towner & Muñoz, 2018; Marquart et al., 2019; Bode &
Dalrymple, 2016)
Source and political attitudes (Marquart et al., 2019; Boerman & Kruikemeier, 2016)
Source trustworthiness (Boerman & Kruikemeier, 2016)
Favorability of candidate (Becker, 2020)
Interest in party and intention to vote for the party (Colliander et al., 2017)
Perceived bias in news (Veenstra et al., 2017)
Participation in a political event (Iowa Caucus) (Dimitrova & Bystrom, 2017)
Trust in government (Towner & Dulio, 2011)
Voter-candidate relationship equity (Hultman et al., 2019)
Political performance, political reputation performance, popularity, and voter loyalty
(Antoniades & Mohr, 2020; Antoniades, 2020; Antoniades & Mohr, 2019)
38
Online political outcomes
•
•
•
•
•
•
Online political participation (Towner & Muñoz, 2018; Bode & Dalrymple, 2016)
Online behavioral intentions (Boerman & Kruikemeier, 2016)
Online relationship quality (Abid et al., 2019)
UGC creation (political tweeting) Bode & Dalrymple, 2016
UGC characteristics - emotion (Paul & Sui, 2019), keywords (Cornfield, 2017)
Cluster density and reciprocity of online community (Lin & Himelboim, 2019)
39
Figure 2.2 illustrates the framework. Per our framework, political brands can directly influence
voters through their official channels (Abid et al., 2019), political advertisements or promoted
content (Boerman & Kruikemeier, 2016), and official brand communities (Penney, 2017). The
latter two mechanisms remain essentially unexplored but the official channels of political
brands can have compound effects. For instance, the content that politicians or political parties
post on their official channels (Hultman et al, 2019; Walker et al., 2017), their approach or
orientation (e.g. interactive and relational) towards these channels (Abid et al., 2020; Hultman
et al, 2019; Anim et al., 2019), and various characteristics of their profile have an effect on
voters (Walker et al., 2017; Lin, 2017). Moreover, the official channels generate word of mouth
(Hultman et al, 2019). Simply by being present and visible on social media, politicians and
political parties prompt positive outcomes (Antoniades & Mohr, 2019; Anim et al., 2017).
Figure 2.2 Conceptual framework of PSSM literature
Besides directly influencing voters, politicians and political parties impact voters indirectly
through user-generated content (eWOM), social media influencers, and crowd-sourced brand
communities (e.g. Lin & Himelboim, 2019; Shmargad et al., 2018), elements of social media
that are not controlled by the politicians or political parties but are influenced by them. Our
review shows that content posted by politicians and their actions impact the nature of usergenerated content (e.g. Paul & Sui, 2019; Jensen & Bang, 2017), which is known to influence
voters (Berman et al., 2019). Interestingly, studies show that user-generated content also affects
40
political parties’ and politicians’ content or activities (Becker, 2020, Calderón-Monge, 2017).
This explains the two-way effects of uncontrolled social media.
Factors affecting politicians’ adoption of social media for political marketing (e.g. Antoniades
& Mohr, 2019; Miller, 2013) are perhaps less relevant today since most politicians are present
on social media. A more pertinent topic is how politicians or political parties use social media
marketing. How they use social media is contingent upon various party or politician-based
factors like the politician’s branding strategy, gender (Paul & Sui, 2019; Jensen & Bang, 2017),
country, and culture (Peres et al., 2020). Additionally, situational variables like race’s
competitiveness or level of personalization in politics also impact the manner in which
politicians use social media (Paul & Sui, 2019; Grusell & Nord, 2020).
Voters’ adoption of social media for political activities is driven by factors like age, gender,
political knowledge, trust in media, and political involvement (e.g. Towner & Muñoz, 2018;
Garcia-Castañon et al., 2011). Importantly, whether voters are influenced by the political
marketer-generated content or user-generated content is contingent upon various content-based
and user-based factors. Content characteristics like visuals, certain themes, negative valence,
and emotion magnify the effect of the content (e.g. Berman et al., 2019; Walker et al., 2017).
Similarly, user-based factors like political self-consciousness, political ideology, and prior
knowledge moderate the effects of political content on various outcomes (e.g. Marquart et al.,
2019; Marder et al., 2018; Bode & Dalrymple, 2016).
Finally, our framework incorporates four categories of voter outcomes identified in the
literature. These are online political outcomes, offline political outcomes, election outcomes,
and content virality (likes, shares, and retweets). Online political participation comprises
variables like online relationship quality (Abid et al., 2019), online behavioral intentions
(Boerman & Kruikemeier, 2016), and online political participation (Towner & Muñoz, 2018),
among others. Offline political participation encompasses desirable outcomes like offline
political participation (Towner & Muñoz, 2018), intention to vote (Colliander et al., 2017),
political attitudes (Marquart et al., 2019; Boerman & Kruikemeier, 2016), and voter-candidate
relationship equity (Hultman et al., 2019), among others. Election results and votes received
are also studied as outcome variables, although, this is only true for studies assessing social
media’s predictive abilities (Shmagrad et al., 2018; Safiullah et al., 2017; Lin, 2017; Cameron
41
et al., 2016). Finally, the virality of political marketer- and user-generated content is the fourth
outcome that appeals to researchers (Berman et al., 2019; Walker et al., 2017).
Several studies note that the effect of political content and political activities on the
aforementioned outcomes is mediated by variables such as candidate image (Hultman et al.,
2019), emotional reactions (Marquart et al., 2019), and persuasion knowledge (Boerman &
Kruikemeier, 2016).
In conjunction with the thematic and contextual analyses, the framework, along with Tables
2.3 and 2.6, offers researchers a structured and comprehensive view of research exploring
political marketing on social media. Therefore, the framework can help avoid duplication of
research and facilitate the discovery of research gaps (Paul & Criado, 2020). We stress once
again that the framework only utilizes the articles included in our review. Therefore, it is
limited by the current state of the literature. For instance, besides user- and content-based
variables, situational variables also impact a content’s reception. However, these variables have
received limited attention in the literature.
2.6.2 Future research agenda
There remains a need to explore PSMM across diverse geographic contexts. However, studies
need to go further and integrate factors like political systems, political culture, political and
technical capabilities, forms of democracy, and voter participation levels to understand how
these macro variables influence political social media marketing and voters’ behavior. For
instance, the US politicians rely on large databases, favorable laws (Bennett, 2016; Newman,
2016), advance technical capabilities (Cacciotto, 2017), and a presidential form of democracy,
which makes the US more conducive to social media marketing (Harris & Harrigan, 2015).
Cultural dimensions that lead to variations in voters’ responses to political social media
marketing also merit investigation (Marder et al., 2019). Future studies should consider the role
of political, social, and cultural variables for a more nuanced understanding of the domain.
Practitioners will value research that looks beyond young voters. Demographic segments like
older voters and women remain under-researched (e.g. Towner & Muñoz, 2018; Harmer &
Wring, 2013). Considering women are prolific users of social media, the segment should be
42
explored further. Similarly, undecided voters, an important behavioral segment, are not
examined in the literature.
Literature is predominantly focused on Twitter and Facebook. Future researchers should study
other platforms like Instagram, YouTube, and Snapchat (Bossetta, 2018). With Gen Z
becoming a sizeable voter segment, TikTok merits investigation since its political use is on the
rise (Galer, 2020). Particularly, the differences in PSMM practices and voter behavior across
social media platforms merit investigation. For example, the visibility of a user’s ‘likes’ has a
negative impact on Facebook (Marder et al., 2018), but does this hold true for Twitter? Is one
platform better for PSMM or achieving specific goals like building voter relationships than
others? The effect of platform characteristics on users is a relevant and timely topic in social
media marketing (Dwivedi et al., 2020).
There is a need to study PSMM outside elections because an overt focus on the short-term (or
sales) is contrary to the contemporary marketing thought. Voter allegiance should be nurtured
over the long term rather than rented during elections. Researchers can also explore if the level
of politics (local, state, national) influences PSMM or voters’ behavior. Finally, the number of
studies embedded in established theories remains limited. Although, some consensus is
emerging on relevant theories and models like Social Identity Theory (Hultman et al., 2019;
Kenski et al., 2017; Veenstra, 2017), Two-Step Flow of Information Theory (Penney, 2017;
Bode & Dalrymple, 2016), Source Credibility Theory (Jain & Ganesh, 2020; Page & Duffy,
2018), Self-Presentation Theory (Grusell & Nord, 2020; Colliander et al., 2017), and the
Elaboration-Likelihood Model (Abid et al., 2019; Iyer et al., 2017). Future scholars can benefit
from a range of theories from the disciplines of personal behavior, social behavior, and mass
communication (see Ngai et al., 2015), most of which remain underutilized in PSMM
The impact of PSMM on voters and voting behavior
The review shows that scholars want to know how PSMM is used by practitioners (political
advisors and politicians), with an emphasis on its impact in the short-term, i.e., elections.
Therefore, future research could further investigate PSMM in relation to voter behavior.
Various outcomes have been explored (e.g. Hultman et al., 2019; Anim et al., 2019; Dimitrova
& Bystrom, 2017), but pertinent outcomes like voting, volunteering, WOM, and financial
contributions have received limited attention. The latter is important given seventy-five percent
43
of Facebook ad spending in the US election cycle of 2020 aimed to raise funds, solicit
contributions, or sell merchandise (Homonoff, 2020).
Future researchers could further refine our understanding of social media’s predictive
capabilities (e.g. Lin, 2017; Cameron et al., 2016) and identify other indicators of an electoral
win across different platforms. Identifying such metrics or analytics will help political
marketing managers understand the ROI of PSMM and that of each platform. A comparison
of the predictive capabilities of various platforms also merits further investigation (e.g.
Cameron et al., 2016).
General approach to PSMM
Candidates are becoming homogenous in their PSMM, which indicates an ideal approach (e.g.
Peres et al., 2020). However, Donald Trump’s unique approach and his success negate this
view. This presents an interesting dilemma for future researchers to resolve. Politicians and
political parties offer limited interactivity and engagement opportunities on social media (e.g.
Abid & Harrigan, 2020), but there is inadequate guidance as to what these interactive and
engagement opportunities are or entail, and whether they have a positive impact on desired
outcomes. For instance, is it feasible for a politician to engage in a dialogue with voters on
social media?
Personalization is a growing trend in politics and few studies explore its’ effectiveness (e.g.
Colliander et al., 2017). Is it more effective than an issue-dominant approach? Should
candidates post personal content frequently? Similarly, it will be worth knowing if certain
candidate characteristics magnify their appeal or effect on social media. Do attractive
candidates perform better on social media? Do politicians that manage their social media
themselves (Donald Trump) fare better? Is a populist stance better suited to social media? Such
research has practical significance.
Barack Obama showed that PSMM is important beyond elections and campaigns, that is, once
politicians have been elected and are in government (Cogburn & Espinoza-Vasquez, 2011).
However, researchers have ignored this important aspect of PSMM. Are distinct strategies,
content, and orientations required when in government and opposition? Is a positive valence
more effective when marketing for incumbents? Such questions demand consideration.
44
PSMM and branding
PSMM’s effects on political brand equity, brand personality, brand knowledge, and brand
image remain unexplored and present viable directions for future research. Since behavior
towards political brands differs from commercial brands (Marder et al., 2018; Boerman &
Kruikemeier, 2016), research is needed to understand when, how, and why these deviations
occur. Current research on brand communities, both official and unofficial, offers limited
insights into how these communities operate. The relationship marketing perspective is ideal
to explore official brand communities and understand how political candidates and parties can
create and engage brand communities (Steinhoff et al., 2019; Sheth, 2017).
Social media and political relationship marketing
Relationship marketing is the advocated approach to political marketing (Ormrod et al., 2013)
and PSMM (Harris & Harrigan, 2015). There remains a need to understand what a relationship
marketing approach towards PSMM entails. Future studies should identify effective casestudies of using PSMM for long-term political relationship marketing. Additionally,
quantitative research is yet to establish if a relationship marketing orientation is more effective
than a traditional approach to political marketing. Various customer relationship typologies
(e.g. Guo et al., 2017) can be used to explore how PSMM aids the development of voter
relationships. Further, how PSMM can facilitate inter-voter relationships is an important
question (Steinhoff et al., 2019). Finally, political marketing literature highlights a relational
approach towards society and various stakeholders (Hughes & Dann, 2009). Future researchers
can add value by exploring PSMM beyond voters. For example, Donald Trump regularly
communicated with stakeholders like Fox News and National Rifle Association via social
media.
Political user-generated content and eWOM
Berman et al.’s (2019) study on Twitter is the only attempt to understand the effect of content
cues on virality. In light of their work, future research can attempt to understand the virality of
UGC using different UGC cues and social media platforms. Further research could understand
how and why voters create political content and the implications this has for the overall political
marketing strategy (e.g. Penney, 2017). Limited research explores the effect of UGC/eWOM
on voters’ attitudes and behaviors (e.g. Iyer et al., 2017), which is worth exploring given the
effect of UGC is distinct from MGC (Müllerb & Christandl, 2019). Future researchers can also
45
explore the different types of UGC (e.g. influencer-generated, celebrity-generated, citizengenerated).
PSMM and political MGC
Politicians need to provide content that is relevant, valuable, and enriching to the voter
experience (Steinhoff et al., 2019). Future researchers can use various content classifications
and characterizations highlighted in marketing literature to understand the effect of various
MGC characteristics (e.g. Tafesse & Wien, 2017; Barger et al., 2016). Importantly, the effect
of political MGC has mostly been studied via content analysis, which does not allow for an
understanding of the interplay between political MGC and source, situational, or user
characteristics. How source characteristics, situational variables, and voter’s personality traits
impact political MGC’s reception are topics that demand attention. Experimental studies can
add value here (e.g. Colliander et al., 2017; Boerman & Kruikemeier, 2016).
PSMM and political advertising
Only two studies in our selection investigate political social media advertising (Boerman &
Kruikemeier, 2016; Vesnic-Alujevic & Bauwel, 2014). Evidence suggests that promoted
tweets have a counteractive effect (Boerman & Kruikemeier, 2016). This is surprising since
political advertising on social media, particularly Facebook, constitutes the largest portion of a
campaign’s digital marketing budget. However, the limited number of studies may be attributed
to our search terms which do not cover political advertising.
Other topics
Value creation and co-creation, prominent themes in social media marketing (Kapoor et al.,
2018; Arrigo, 2018), are rarely explored in PSMM. Value creation is an element of political
marketing’s definition (Hughes & Dann, 2009). How can social media facilitate the co-creation
of value between political brands and voters? What are the antecedents and consequences of
this co-creation? What are the factors that impede value co-creation? These basic questions
remain unanswered. Similarly, customer engagement, an important concept in contemporary
marketing (Hollebeek et al., 2019; Brodie et al., 2011), is pertinent to social media (Ajiboye et
al., 2019; Arrigo, 2018). However, Pich et al. (2018) remains the only study on the topic.
Research that explores how politicians can generate voter engagement is valuable to political
actors and democracy in general.
46
Another important topic in marketing, influencer marketing (Vrontis et al., 2021; Appel et al.,
2020), has received little attention (Shmargad et al., 2018). The motivations driving political
social media influencers, characteristics and types of influencers, and their impact on voters
are valid areas of research.
There is also a need to understand PSMM’s role within and in relation to the overall media mix
and digital marketing strategy. Social media do not function in isolation and are impacted by
or impact other media (Dwivedi et al., 2020). Therefore, an understanding of PSMM as a
component of a holistic political marketing strategy is beneficial. This evaluation is frequently
highlighted in recent social media marketing literature also (Appel et al., 2020; Dwivedi et al.,
2020; Voorveld, 2019). Finally, despite ethical concerns around PSMM (Appel et al., 2020),
we have a limited understanding of ethical issues related to PSMM. This provides a meaningful
avenue for further research.
2.7 Conclusion
Our systematic review synthesizes and presents an overview of the literature in the field of
PSMM. The review illustrates that PSMM is gaining traction globally, particularly among
marketing scholars. Research is starting to assimilate contemporary marketing paradigms.
Similarly, the prevalent themes, which are emphasized in our review, reflect growing
synchronization with social media marketing literature (Alalwan et al., 2017). The growing
number of publication outlets, a nascent domain, high practical significance, and the many
promising areas of research offer an opportune time to undertake research in PSMM.
Despite a thorough and systematic approach, the review has limitations. The search term, social
media, is not the only relevant term. The terms ‘new media’ and Web 2.0 are also used in the
literature. Similarly, we do not include the term ‘social networks’. However, marketing studies
almost exclusively rely on the term ‘social media’ and we include the names of all prominent
social media platforms (including SNSs) used in political marketing. Therefore, these issues
have limited bearing on our review. The number of databases limits our selection. Similarly,
the exclusion of conference papers and book chapters limits the findings. We compensate for
this by having slightly low cut-off values for IF and H-Index. Finally, the review is limited by
its marketing-dominant view. This is also reflected in the themes identified..
47
CHAPTER 3: AN EXPLORATION OF SOCIAL MEDIAENABLED VOTER RELATIONSHIPS THROUGH USES AND
GRATIFICATION THEORY, PSCYCHOLOGICAL
CONTRACT, AND SERVICE-DOMINANT ORIENTATION
Abid, A., & Harrigan, P. (2020). An exploration of social media-enabled voter relationships
through uses and gratifications theory, psychological contract and service-dominant
orientation. Australasian
Marketing
Journal, 28(2),
71–82.
https://doi.org/10.1016/j.ausmj.2020.02.002
Chapter 3 is the first study of the research project that relies on primary data. It is a qualitative
study that attempted to answer three questions regarding the nature of social media-enabled
voter relationships from the perspective of young voters.
How can political brands
strengthen their relationships
with voters on social media?
CH3. What is the nature
of social media enabled
voter relationships?
RQ1. What are the
gratifications that drive
voters to follow political
brands?
RQ2. What are the drivers
of social media-enabled
voter relationships?
RQ3. What are the
interactions that underpin
social media-enabled
voter relationships?
CH4. What is the
marketing orientation
that political brands
adopt towards social
media marketing?
CH5. What is the role of
marketer-generated
content in driving online
voter relationships?
RQ1. Do political
brands adopt a
relationship marketing
orientation towards
social media
marketing?
RQ2. Are there any
dimensions that are
not represented in
RMO framework?
RQ1. What is the effect
of various content cues
on online relationship
quality?
RQ2. Does content
curation have an
impact on online
relationship quality?
Figure 3.1 Chapter 3 (Paper 2)
48
CH6. What are the roles of
marketer-generated
content and behavioral
engagement in driving
online voter relationships?
RQ1. What are the
effects of various
content cues on
engagement and online
relationship quality?
RQ2. Does behavioral
engagement mediate the
effect of content cues
on online relationship
quality?
3.1 Abstract
The relationship that develops between voters and political entities on social media is underresearched. This study offers insight into this contemporary phenomenon by exploring the
factors that initiate and drive social media-enabled voter relationships. Data were collected by
conducting focus groups with young voters. The Uses and Gratifications Theory was used to
explore the motivations that stimulate young voters to follow political entities on social media.
Secondly, the drivers of the resulting relationship were explained using the concept of the
psychological contract. Lastly, the various online interactions that underpin this relationship
were investigated using the concept of service-dominant orientation. The findings reveal that
social, informational, and entertainment gratifications are the primary drivers of this
relationship. Further, developmental, individuated, relational, and ethical interactions fortify
online voter relationships. However, a lack of trust, unmet expectations, and an absence of
individuated interactions are major challenges. The study recommends humanizing politicians
through social and emotional content.
49
3.2 Introduction
The declining voter engagement and participation in Australian politics is a consistent trend
(Hasham, 2016). This voter disengagement, particularly in traditional and electoral politics, is
especially prominent in young voters (Loader et al., 2014). A relationship-based approach
towards political marketing has the potential to re-engage voters and revitalize democracy
(Ormrod et al., 2013; Henneberg & O’Shaughnessy, 2009). Social media are ideal to build
relationships with young voters. However, political entities (PE) have been hesitant or
incapable of adopting a relational approach to social media (Parsons & Rowling, 2015; Harris
& Harrigan, 2015). More than a third of Australian social media users follow elected officials
or political candidates (Newman et al., 2017), which allows PEs to build a direct channel of
communication and interaction with voters. Surprisingly, little is known about the nature of the
online relationship that develops between young voters and PEs because of the former
following the latter on social media. This study undertakes a comprehensive exploration of this
social media-enabled relationship based on marketing frameworks that have received limited
attention in the political marketing literature.
Marketing’s dominant paradigm shifted from marketing mix towards relationship marketing
more than two decades ago (Grönroos, 1994; Gummesson, 1994). However, the field of
political marketing is yet to reflect this. Apart from the initial discussion on PRM’s potential
and compatibility with the political context (Henneberg & O’Shaughnessy, 2009; Bannon,
2005; Scammell, 1999), calls for further research have not yielded significant results.
Furthermore, the few studies in the area do not offer a first-hand voter perspective (e.g. Parsons
& Rowling, 2018; Harris & Harrigan, 2015). Similarly, despite a substantial amount of
literature on online customer relationships (Verma et al., 2016; Malthouse et al., 2013; Barnes
& Cumby, 2002), online voter relationships have received limited attention. The present study
attempts to remedy this. Firstly, it explores the motivations that drive young voters to initiate
an online relationship with PEs by following them on social media. The Uses and Gratifications
Theory (UGT) was deemed appropriate for this purpose. UGT is commonly utilized to examine
the gratifications that users seek from different media and the various activities associated with
these media (Sheldon & Bryant, 2016; Flanagin & Metzger, 2001; Katz et al., 1973). Secondly,
the concept of the psychological contract was employed to study the resulting relationship and
its drivers (Rousseau, 1995). Lastly, since PEs provide services, the online interactions that
underpin this relationship were studied using the concept of service-dominant orientation
50
(SDO) (Karpen et al., 2015). The goal of the study was to achieve an understanding of social
media-based voter relationships from the perspective of a specific group (young Australian
voters). A qualitative methodology that uses focus groups is ideal in this scenario (Calder,
1977). The research was carried out by conducting three focus groups with twenty-four young
voters. The selection and union of the three frameworks allowed for a holistic, logical, and
step-wise exploration of what initiates this relationship, the factors driving it, and the
interactions that underpin this online relationship. UGT predominantly considers the
individual’s viewpoint, whereas SDO and the psychological contract are mainly focused on
organizational and interactive perspectives respectively. Furthermore, the psychological
contract allows the study of the relationship as a whole, whereas SDO facilitates a granular
approach by breaking down the relationship into its many constituent interactions.
The study contributes to political marketing theory by further integrating it with the
relationship marketing paradigm. It contributes to social media literature by explaining a
relatively unexplored online relationship. Additionally, the research is the first to extend
psychological contract and SDO to the political context. Finally, the study provides actionable
insights to PEs and their social media teams.
3.3 Literature review
3.3.1 Uses and Gratifications Theory
The Uses and Gratifications Theory has provided valuable insight at the nascence of each
media including radio, TV, newspapers, cable TV, WWW, internet, cell phones, and social
media (e.g. Whiting & Williams, 2013; Flanagin & Metzger, 2001; Katz et al., 1973). As per
UGT, users select media to fulfill their specific needs which leads to ultimate gratification
(Katz et al., 1973). UGT has origins in the study of political information-seeking behavior
(Lariscy, Tinkham, and Sweetser, 2011). Blumler and McQuail (1969) devised the Political
Media Gratification Scale to investigate gratifications driving the consumption of political
content on TV. Their scale investigated the gratifications of guidance, reinforcement of
decisions, surveillance, excitement, and expected utility of political information in future
interpersonal communications. Garramone et al. (1986) found that the use of political computer
bulletin boards was driven by surveillance, diversion, and personal identity (a combination of
social utility and self-expression).
51
More recently, gratifications associated with various online political activities like subscribing
to a political party’s e-newsletters, consuming political news on social media, reading political
blogs, and visiting candidate’s social media pages have been investigated (Macafee, 2013;
Macafee & De Simone, 2012; Kim & Johnson, 2012; Ancu & Cozma, 2009; Jackson &
Lilleker, 2007; Kaye & Johnson, 2002). These studies highlight that informational, social, selfexpression, entertainment, and guidance are the main gratifications that drive online political
activities. Parmelee and Roman (2019) confirm the presence of these gratifications among
users following politicians on Instagram. Their study shows that information, guidance, and
social gratifications were the predominant gratifications for following politicians on Instagram.
Additionally, the gratifications of entertainment and self-expression were only relevant for
younger (under 40) and less-educated followers respectively. Gratifications are rarely universal
and vary with activity (Macafee, 2013). For instance, political comments are driven by
information-sharing and social needs, whereas politicians are ‘liked’ for self-expression.
Similarly, gratifications are determined by audience characteristics also (Parmelee & Roman,
2019; Park, 2013). Moreover, gratifications differ with medium i.e., newspaper, TV, and radio.
Although some gratifications are universal, others are media and activity-dependent
(Timmermans & De Caluwé, 2017; Sheldon & Bryant, 2016).
Although the gratifications associated with following politicians on Instagram have been
studied recently, the quantitative approach adopted by Parmelee and Roman (2019) does not
allow for the emergence of new activity-specific gratifications. New gratifications emerge with
new media technologies and activities. However, a quantitative approach that uses instruments
designed for earlier media technologies (or activities) and a predetermined list of gratifications
do not allow the emergence of any new gratifications. For instance, Ancu and Cozma (2009)
and Park (2013) investigated three gratifications, whereas Macafee (2013) studied four
gratifications acknowledging that these were not exhaustive. As Sundar and Limperos (2013,
p. 4) note “measures designed for older media are used to capture gratifications from newer
media; and gratifications are conceptualized and operationalized too broadly (e.g., informationseeking), thus missing the nuanced gratifications obtained from newer media”.
The authors propose that an exploratory, qualitative approach is ideal to identify new and
nuanced gratifications associated with following politicians on social media, along with
confirming the already known gratifications. In this regard, Timmermans and De Caluwé’s
(2017) study provides a sound methodology. Prior to adopting a quantitative approach, they
52
explored the gratifications associated with Tinder using a qualitative approach that consisted
of focus groups with Tinder users. This allowed for identification and incorporation of activity
and platform-specific gratifications, which they subsequently confirmed using quantitative
analysis. Other UGT studies have also taken a qualitative, exploratory approach to identify new
gratifications associated with social media use (e.g. Whiting & Williams, 2013).
UGT’s origin in media effects on political behavior (Ruggiero, 2000) makes it an ideal
framework to explore the motivations that drive young voters to initiate an online relationship
with PEs by following them on social media. This study employs UGT to explore the first
research question:
RQ1. What are the uses and gratifications that drive young voters to follow PEs on social
media?
3.3.2 Psychological contract
Traditionally, the psychological contract is used to explain employer-employee relationships
(Rousseau, 1995). However, it is also applicable to consumer relationships. Customers are
never fully aware of the explicit rules of commercial contracts but do possess a mental schema
of a seller’s obligations (Pavlou & Gefen, 2005). From a marketing perspective, Guo et al.
(2017) define the psychological contract as the mental structure of the consumer’s knowledge
about their relationship with a business.
Rousseau (1995) categorized psychological contracts as transactional or relational contracts.
Transactional contracts are episodic, specific, and instrumental in nature, whereas relational
contracts are flexible, long-term, and trust-based. Additionally, a third contract has also been
discussed. This transpersonal contract is based on an organization’s stance on a cause or an
ideology that is valued by the customer (Mason & Simmons, 2012). A belief that the other
party has failed to fulfill its obligations leads to a psychological contract breach, which is
cognitive in nature. It is distinct from psychological contract violation, an emotional state that
may or may not follow a breach (Morrison & Robinson, 1997). PCV has a negative impact on
customers, particularly, on committed customers (Montgomery et al., 2018; Malhotra et al.,
2017). In the e-commerce context, the buyers’ psychological contract regarding a seller’s
obligations includes timely delivery, no discrepancy in advertised and delivered product, and
53
adherence to payment and refund policy (Pavlou & Gefen, 2005). Guo et al. (2017) consider
reciprocity, mutual benefit, and the nature of the exchange (social vs. economic) as major
factors that decide the type of psychological contract in place. Their research identified four
psychological contracts: relational, standard, transitional, and captive contracts. As per Mason
and Simmons (2012), service providers offer different service value propositions (functional,
economic, psychological, and ethical), and the consumer’s acceptance of the proposition results
in either transactional, relational, or transpersonal contracts.
The authors consider the psychological contract to be particularly relevant to studying voterPE relationships. Hannah et al. (2016) describe the three main characteristics of a psychological
contract. The first characteristic is the existence of multiple contract makers like salespeople,
customer service representatives, marketing departments, etc. This is replicated in the electoral
market where multiple entities, such as the party, its leaders, and its candidates serve as contract
makers. The second characteristic is the incompleteness of the psychological contract. Like
consumers, who cannot immediately identify the complete parameters of their long and
complex relationship with a seller, voters are also unaware of the complete implications and
consequences of their relationship with a PE. Circumstances change and new issues emerge,
creating new expectations and obligations. This highlights that at any given point, a voter’s
psychological contract is incomplete. Lastly, voter-PE relationships satisfy the condition of
mutuality. Voters and PEs have a shared understanding of the terms of their relationship. This
mutual understanding regarding expectations and obligations is derived from past experiences,
the party’s policies, the candidate’s promises, and the party’s or candidate’s legacy. Although
the psychological contract has not been applied to voter relationships, it has been utilized to
study the relationship between members and labor unions (Braekkan, 2013; Turnley et al.,
2004). This study uses the concept of psychological contract to investigate the second research
question:
RQ2. What are the factors that drive social media-enabled voter-PE relationships?
3.3.3 Service-dominant orientation
Service-dominant logic is based on the ‘logic of togetherness’. It envisions multiple actors
working together to achieve mutual benefit in an environment that is characterized by
relationships and trust (Joiner & Lusch, 2016, p. 26). SDL’s vagueness has been noted in the
54
literature (Grönroos & Gummerus, 2014). This has resulted in the development of middle-level
theories that operationalize SDL (Lüftenegger et al., 2017; Karpen et al., 2015). One of these
is service-dominant orientation, a collection of organizational capabilities that facilitate
resource integration and value-co-creation among various actors through individuated,
relational, ethical, developmental, empowered, and concerted interactions (Karpen et al.,
2015). The definitions are listed in Table 3.1. Excellence in the facilitation of these interactions
has a positive impact on customer trust, satisfaction, commitment, and an organization’s
performance (Karpen et al., 2015; Wilden & Gudergan, 2015). The significance of interactions
has been acknowledged in relationship marketing also (Prahalad & Ramaswamy, 2004;
Grönroos, 1997). Relationship building requires meaningful interactions and experiences with
customers. Notably, Grönroos (2004) placed interactions at the center of the relationship
marketing process.
Table 3.1 Service-dominant orientation
Type of interaction capability
Individuated
Relational
Ethical
Empowered
Developmental
Concerted
Definition ((Karpen et al., 2015)
An organization’s ability to:
Understand the resource integration processes, contexts, and desired outcomes
of individual actors within the service system.
Enhance the connection of social and emotional links with individual actors
within the service system.
Act in a fair and non-opportunistic way toward individual actors within the
service system.
Enable individual actors within the service system to shape the nature and
content of the exchange.
Assist individual actors’ own knowledge and competence development within
the service system.
Facilitate coordinated and integrated service processes with individual actors
within the service system.
A services perspective emphasizes skills, knowledge, information, and relationships rather than
products, making it an attractive proposition for political marketers (Butler & Harris, 2009).
Additionally, the service-dominant logic provides an overarching framework for delivering
service, providing unique offerings, and creating value in the political context, which will lead
to a competitive advantage for PEs and satisfied voters (O’Cass, 2009). Keys to voter
satisfaction and engagement under this paradigm are meaningful interactions with voters and
party capabilities that facilitate voter relationships (O’Cass, 2009). Since all politicians and
political parties essentially provide services, SDO is an ideal theoretical basis to answer the
study’s third research question. Further, it allows for the breakdown of the online voter
relationship into its constituent interactions.
55
RQ3. What are the interactions that underpin social media-enabled voter relationships?
As discussed in the introduction, the collective use of these paradigms allows a comprehensive
understanding of the motivations for following PEs, the drivers of the resulting voter-PE
relationship, and the various interactions that underpin this relationship. Voters seek to satisfy
distinct gratifications by following PEs on social media, thus their motivations for establishing
an online relationship vary. Since the relationship transpires due to a media activity, UGT is
ideal for exploring the initiators of this relationship (Sundar & Limperos, 2013). The authors
recognize that these initiators are not the same as drivers. Various theoretical perspectives have
been used to explain customer relationships and their drivers (e.g. Dick & Basu’s (1994)
attitudinal perspective, Gruen’s (1995) relational perspective, Mende et al.’s (2013) consumer
trait perspective, etc.). However, the psychological contract’s focus on the exchange (economic
or social) and mutual benefits (rather than organizational benefits) (Guo et al., 2017) resonates
with the various definitions of political marketing (Hughes & Dann, 2006; Henneberg, 2002).
Additionally, promises, expectations, and perceived obligations, which are essential
ingredients of psychological contracts, apply to the political context also. Therefore, the
authors selected the psychological contract to explore the drivers of online voter-PE
relationships. Lastly, relationships are composed of many interactions over an extended time.
This is true for social media-based relationships also. These interactions have various
dimensions that can facilitate or impede online relationships. Knowledge of these dimensions
allows PEs to reinforce online voter relationships. Not only does SDO distinguish the various
dimensions of interactions, SDO’s grounding in services literature makes it pertinent to PEs.
The three frameworks are ideally suited to answer the three research questions. Further, an
overlap between the three frameworks facilitated the data collection and interpretation process.
3.4 Research methodology
An exploratory qualitative approach was adopted to address the research questions. The
approach is appropriate for under-researched areas (Sofaer, 1999), which is the case in political
marketing and contemporary marketing concepts (Ormrod et al., 2013). Specifically, focus
groups were conducted as they are suited to the acquisition of an in-depth understanding of a
group’s attitudes, experiences, feelings, perceptions, beliefs, norms, and values (Carey &
Asbury, 2012; Rabiee, 2004; Kitzinger, 1994). Importantly, focus groups not only present the
group’s dominant view on an issue but also the different perspectives that exist within the group
56
(Rabiee, 2004). The focus groups were conducted using a phenomenological approach
(Bradbury‐Jones et al., 2009). This approach is ideal when an everyday phenomenon or an
activity is to be described from the perspective of a certain group (Calder, 1977). Further, focus
groups are ideal for understanding student participants due to the insights gained from the
discussion (Palomba & Banta, 1999). Additionally, focus groups are a common form of
research in political marketing (Cyr, 2016; Lees-Marshment, 2014) and have origins in media
effect studies (Kamberelis & Dimitriadis, 2013). A phenomenological approach requires three
to four focus groups with a homogenous audience to capture data, with each focus group
running between one to two hours in a relaxed environment and comprising of four to ten
participants (Krueger & Casey, 2014; Rabiee, 2004; Kitzinger, 1994). The focus groups
satisfied all these criteria.
The focus groups elicited responses using prompts that were sourced from existing research
instruments (Flanagin & Metzger, 2001; Karpen et al., 2015; Guo et al., 2017). This allowed
for theoretical consistency that is required when extending a theory (Hsieh & Shannon, 2005).
Table 3.2 provides example items and their sources. Notably, a semi-structured approach
towards the focus groups was preferred over a rigid format. This facilitated the emergence of
unplanned data that was not being captured by the instruments being utilized for this research.
Researchers conducted three focus groups comprising twenty-four participants. Research
shows that between 80%-90% of the research themes may be uncovered by conducting three
focus groups (Guest et al., 2017). A minimum of three focus groups is sufficient as per Krueger
and Casey (2014) and Morgan (1996). The sample was drawn from students at an Australian
university. Along with being a highly targeted segment, young voters receive significant
academic attention also (e.g. Haenschen & Jennings, 2019; Winchester et al., 2014). All
participants were enrolled on a full-time basis. Thirteen males and eleven females constituted
the sample. The sampling criteria were the requirements that participants must currently follow
an Australian political party or politician on social media and be between the ages of 18 and
26 (it was found that all participants followed international PEs also). The focus groups were
conducted in September of 2017. At the time, there were approximately 1.1 million domestic
students in Australian universities. Of these, 61 percent are under the age of 25, 65 percent are
studying full-time, and 71 percent are pursuing an undergraduate degree (Universities
Australia, 2019). An average Australian finishes high school between the ages of 17 and 18
(Highschool Australia, n.d.).
57
Table 3.2 Examples of items used as a guide for focus group discussions
Uses and Gratification (adapted from Flanagin & Metzger, 2001)
I follow political entities to:
Get information
Feel important
Be entertained
Make decisions
Stay in touch
Psychological Contract (adapted from Guo et al., 2017)
What this political entity and I expect from each other is clearly specified
If good performance is provided, I will vote for this entity
This political entity rewards voters who support it
This political entity and I would help each other without expectation for any return
I have learned to look out for myself in this relationship
Interactions (adapted from Karpen et al., 2015)
This political entity:
Encourages two-way communication with me
Shares useful information with me
Invites me to provide ideas or suggestions
Provides messages to me that are consistent with each other
Makes an effort to understand my individual needs
An invitation email was sent to students enlisted in a first-year, undergraduate marketing
course. This allowed for cordial groups that had some prior familiarity (Rabiee, 2004).
Participants were incentivized with a coffee voucher and refreshments. The focus groups lasted
seventy-five minutes on average. The participants’ introductions (in their own words) are
presented in Appendix A. Following the transcription of audio recordings, the data were
transferred to NVivo for subsequent analysis. A codebook was developed before data analysis.
This codebook was created using deductive category application, which uses existing theory
and research questions to develop initial codes (Mayring, 2000). This approach is prevalent in
directed content analyses (Braun & Clarke, 2006; Hsieh & Shannon, 2005). Further, the
codebook was verified by a senior researcher. The primary coder was a doctoral student trained
in qualitative research methodologies. A directed content analysis was preferred over a
thematic analysis since the research aim was to extend established theories to the political and
social media contexts. As a structured and theoretically driven process, directed content
analysis is ideal for validating or extending theory and studying it in a new context (Elo &
Kyngäs, 2008; Hsieh & Shannon, 2005). Data that did not fit in pre-determined codes were
identified and analyzed to check if new data codes were needed (Hsieh & Shannon, 2005).
The coding process focused on individual and group data (Onwuegbuzie et al., 2009). Specific
attention was paid to words, frequency, sequence, connections, and extensiveness during data
analysis (Rabiee, 2004; Reed & Payton, 1997). To increase the study’s validity and reliability,
58
the study adhered to various instructions highlighted in the literature (Morse, 2015; Brod et al.,
2009). For instance, content validity was ensured by using tested items from established
theoretical frameworks to guide data coding, focus group discussions, and presentation of
findings. Rich and thick descriptions and verbatim extracts (corrected for mistakes) were used
to enhance the study’s validity (Noble & Smith, 2015). Member checks were undertaken. Four
participants were revisited to verify the researchers’ interpretation of their contribution (Morse,
2015; Baxter & Eyles, 1997). A comprehensive coding scheme with clear definitions was
developed before the transcript analysis to ensure reliability (Morse, 2015; Hsieh & Shannon,
2005). The primary coder’s coding was checked by a senior researcher to ensure reliability and
validity. Additionally, internal consistency and reliability are higher when a single coder is
used (Kidd & Parshall, 2000).
3.5 Findings
3.5.1 What are the uses and gratifications that drive voters to follow PEs on social media?
Informational and instrumental gratifications
A dominant motivation that emerged from data pertained to the participants’ informational
needs. Information from the PE and information about the PE were the main drivers of this
social media activity. It allowed respondents to stay up-to-date and get information about PE’s
opinions, activities, and policies. Local politicians were considered a good source of
information about local and community matters; however, national politicians and major
political parties were followed for personal and policy-related information respectively.
Participants who followed smaller parties and single-issue groups were more concerned with
information about specific issues rather than the PE. It was noted that a more dynamic political
context magnified the significance of this gratification.
(I follow politicians) so we know what they are doing, what they are up to, what they are
thinking about, information about what they are saying. It is not necessarily where I get all my
stats from, but information about them, their party, and policies. [Participant 1, Focus Group
3]
(Compared to Australian PEs) I generally follow US politicians because they tend to have
discussions about abstract ideas. They have better talent than the Australians since we only
have a mere 20 million people, so it’s not as good of a political contest. [P8, F3]
59
It’s more like you want to have news. You want to see what they are doing and have news from
them. So, like I follow Malcolm Turnbull’s Facebook page, I don’t necessarily support him.
[P4, F1]
The participants acknowledged that this unfiltered and direct information was important in
building their opinions, particularly on issues where they perceived media bias or limited media
coverage. Although this information was not shared outside a small circle of friends, a few
participants found this information to be valuable when attempting to persuade friends and
family. However, political discussions on social media were deemed a counterproductive
process. To a lesser degree, the participants connected with PEs to evaluate them and consider
their voter-politician fit.
At the moment, I am following the Australian Christian Lobby because with the upcoming
plebiscite I find it really interesting because the media has a very one-sided perspective I think.
So seeing the other side is really interesting. [P4, F2]
I think I will start to form an opinion if I get to hear one person, but I want to hear from the
others. It helps form an opinion and start a conversation. [P2, F2]
I personally think that what you see and do over time is important to know what they are about,
whether you like them or not. I just follow them to get engaged with their character. Especially
with local politicians, I will follow them to see if I want them to be the person who is
representing me. [P1, F1]
Social, entertainment, self-discovery. and civic gratifications
A major factor that compelled participants to follow politicians on social media was the need
for a personal and social connection. Social media allowed the ‘humanizing’ of politicians, and
the participants were keen on personal content depicting the everyday and regular life of
politicians. This acted as a bonding aid between the participants and the politicians. To a lesser
degree, following similar PEs acted as a bonding element between like-minded individuals. A
few respondents felt that having up-to-date information about PEs made them feel important
in their peer group.
60
Even if I am loyal to the party, I don’t like social media content that is very factory generated,
it is boring and that’s not why you are on social media, like, I can read that on the news. We
come to social media because we want a personal connect, and if I am not getting that, it’s not
interesting. [P3, F2]
A political party will only post political stuff but the person will be posting personal stuff and
I want to see how they handle it. [P1, F2]
(Following same PEs) I think it’s like a bonding strain, like, you are a feminist as well. [P4,
F3]
Respondents were in consensus that following PEs was entertaining and fun. Some politicians
were more likely to be followed due to this gratification. Politicians posting content that
presented policies and complex ideas in easy, fun, and accessible ways were appreciated. The
entertainment value of the users’ comments on posts and tweets from certain PEs was
highlighted, along with their use to gauge the temperature on issues. Some politicians who had
reached celebrity status were followed for entertainment reasons also.
I won’t vote for him, but I think that the way he (Sam Dastyari) speaks about issues and
articulates them in entertaining ways is really accessible and obviously very entertaining. [P6,
F2]
Barnaby Joyce is another good one for fun. His Facebook page is hugely entertaining. A few
memes having a dig at Johnny Depp, his is always good value. [P7, F3]
I am also interested in looking at the comments, especially for Trump. I am kind of interested
in how people kind of judge him. It is pretty entertaining. [P8, F1]
Participants believed that following PEs with opposing views was a good way to develop bipartisan views, informed opinions, empathy, and tolerance. Participants discussed how
following PEs allowed them to ‘make up their own mind’. Some were of the view that
responsible citizens must follow PEs to become informed about their leaders and not doing so
was detrimental to the country.
61
(I follow PEs I dislike) To see the other side. You don’t want to be too biased in your thinking,
so you always try to get both sides. [P10, F3]
I think it’s also your responsibility as a citizen in society to keep up with what the current
affairs are and what both sides of the issues are. [P2, F1]
These guys are our leaders. These guys are making decisions that affect our lives in many ways
and in so many different portfolios, and I think a normal student doesn’t care as much as I do,
but it will still be important for them to engage in terms of understanding what plans they are
trying to make, and that’s why I follow (PEs on social media). [P5, F2]
3.5.2 What are the factors that drive social media-enabled voter relationships?
Socio-emotional exchanges
The predominant facilitators of this relationship were social and personal content. This
humanizing of the politician allowed participants to relate and connect with the politicians at a
personal level. In this regard, content that showed similarity with the audience (young voters)
was well received. Minor and non-career politicians were deemed more proficient at this. They
were considered more relatable than career politicians and better at building personal
connections with voters on social media. Participant’s acknowledged that a drawback of this
personal approach to social media was an overt focus on PE’s social media image in place of
policy and performance.
I think that politicians are really similar and everyone can see through it. That’s why I think
that minor parties are better on social media because, usually they are not career politicians
from Liberals and Labor, who are the same person even though they have different stances.
That’s why minority parties do better because they are people who aren’t from politics. [P5,
F1]
Last night in the football grand final between Subiaco and Peel, Mark McGowan posted a
selfie with his two sons, and weeks ago Malcolm Turnbull did the same with his grandson at
the footy. It definitely humanizes these people. It’s harder to get mad at someone when you
think they are someone like you. It is easy to get mad at someone who is like disjointed from
what your life is like, but he seemed like a family man, so you feel a little less angry. [P3, F3]
62
I think what politicians do is they kind of try to project themselves as perfect people, but they
should try to be a normal person on social media who can make mistakes. [P6, F1]
Meeting expectations
The expectations of personal content did not diminish the need for a professional approach.
For instance, spelling mistakes and inappropriately timed tweets were seen negatively. In terms
of political content, bipartisan statements and appeals based on factual narratives were seen in
a positive light, as was content depicting community involvement and awareness. The
participants expected the PEs to take the lead in igniting bipartisan debates on social media and
to be proactive in engaging and teaching youth about politics. Silence on contentious issues
was seen undesirably. Lastly, concerning their political participation, participants were not
keen on expanding this social media-based relationship into offline behavior and going out of
their way for PEs (e.g. calling friends on behalf of PE).
Can you trust the person that gives spelling mistakes? He hasn’t put much thought into it. [P2,
F2]
I would publicly express my approval of a post if it was sort of impartial, whereas I find that if
I see something from the Labor or the Liberal party that I agree with, I am less inclined to
share it because I don’t want to give this image that I have some sort of partisanship. [P9, F3]
Our generation is very lazy and it is very hard to get us to do anything. Social media is one
thing but translating that activity is very difficult. Honestly, I don’t know how they would do it.
If they could find something, that will be amazing. [P7, F1]
Trust
A major inhibitor of this relationship was the lack of trust in PEs and the social media
environment. The social media environment was perceived to be characterized by bias,
manipulation, and misinformation. Further, respondents expressed low levels of trust in their
relationship with PEs on social media and recalled the use of neutral sources to verify
information occasionally. Adding to this lack of trust was the ability of social media to
document interactions whereby past interactions damaged the PE’s credibility in case of
changes in political stances. The use of social media teams to manage politicians’ accounts was
63
considered contrary to the spirit of social media. Nevertheless, most participants were
sympathetic and understood that PEs have resource limitations, such as time and money.
Finally, only two participants had ever unfollowed PEs, which was due to excessive noise and
insensitively timed personal posts.
Sometimes it could be good to follow politicians directly, but you also need to follow people
that are analyzing what they (PEs) are saying because you need to have a check on them
basically. [P1, F3]
Anything on social media has been manipulated to push a certain agenda, to tell the truth
attractively. I do think it’s a bit manipulative. [P5, F2]
With him, you can see that he is not behind his social media account. It’s so obvious that it’s a
team. I think that’s the part which pisses me off. [P3, F2]
The participants perceived PEs’ activities on social media as self-serving in nature and salesoriented. Phrases like “they just want to sell their brand”, “they are after votes”, “promote
their agenda”, etc., exemplified the participants’ view on the social media use of PEs.
However, those who had offline relations with the politician they followed and those who
followed local politicians and smaller PEs characterized their relationships as mutually
beneficial. Another observation was a greater level of trust in smaller political parties’ and local
politicians’ use of social media.
Anything to push their ideology. They (PEs) say what needs to be said. Not from the heart. Can
be from the heart, but a lot of it is how it is going to affect my ratings, how this is going to
improve my gain. [P6, F3]
I am in the same boat (political volunteer) as [P2, F3]. My example is Andrew Hasting. I have
gotten investment in what he is saying online but that’s because I had a pre-existing
relationship. [P3, F3]
I have friends who are at council positions and they are posting throughout the year. Like this
is what I am doing, this is where I was. People in federal or state politics will only post in
months before the election for traction. I feel like they are using it for two different things. Like
64
one is using it for maintaining relations and figuring out what the community wants, and the
other one is like I need votes quickly to get elected. [P5, F1]
3.5.3 What are the interactions that underpin social media-enabled voter relationships?
Ethical, individuated, and empowered interactions
As discussed earlier, the exchanges on social media were embedded in an environment of
skepticism and distrust. Furthermore, participants felt that the PEs were not giving sufficient
attention to youth-related issues on social media. The participants thought that the PEs did not
understand their needs and offered little consideration to the youth’s concerns when devising
policies. They believed that politicians were accessible but did not offer enough opportunities
to involve young voters on social media. The participants discussed how PEs from other
countries use social media to guide community development by generating ideas, conducting
live Q&A sessions (on Facebook), and promoting charity causes.
Social media is such a new concept. It is such a great outreach tool. I don’t think they (PEs)
know how to use it. I think the majority of the politicians are sort of baby boomers. It is hard
to say that they understand us. [P4, F2]
You have no super young people at super high levels that really connect with the majority of
people our age. [P4, F1]
A politician opened a donation account for Rohingya in Indonesia, and it worked well. They
got a lot of money. [P1, F2]
Developmental, relational, and concerted interactions
Respondents acknowledged the developmental impact of following PEs on social media. The
ability to get undistorted communication from PEs, exposure to both sides of the argument,
and real-time information were the benefits that were highlighted by the participants.
Additionally, the participants believed that an increase in political knowledge would lead to a
better citizenry. However, participants felt that PEs could do more to educate voters in a
bipartisan manner. The participants were doubtful whether PEs were looking for actual
dialogue and genuine engagement, and considered their communications to be politically
65
driven. A few participants highlighted that it was impractical for politicians to try and engage
in dialogue on social media.
(What PEs should do on social media) Maybe they should push education of politics. Maybe
they should say that you need to have an opinion and we don’t care if it is left or right, but
having a strong opinion is part of a first-world country and educated voters. [P2, F2]
I think that they try to push dialogue but in the end, they always go back down the voting track.
[P1, F1]
(Whether PEs are seeking dialogue) I think that since social media is seen in a traditional way
like an email (by PEs), the politician would not engage in a slightly different political stance
because they feel that on social media somebody can just do a screenshot and use that as
evidence for when they do come up towing the party line. [P5, F1]
Most participants gave greater importance to party communications when it came to policy
matters. An inconsistency between the politician’s and the party’s communications reduced the
belief in the materialization of the agenda. However, most respondents found consistency in
the party’s and its politicians’ communications. Finally, politicians who took inconsistent
positions in their social media communications were likely to have their credibility questioned.
We are Australia, We are not America. The way our system works here is that it’s done on a
party basis and not one person. He might be the face of the party but it’s ultimately the party
that makes decisions. [P4, F2]
I generally look at the people in that party and if they are behind what this particular person
is promising. If their opinions are different I lose hope. [P3, F2]
I think one thing is that when they (politicians) say something online, they don’t know that
three years ago they said something that was completely the opposite. [P1, F3]
66
3.6 Discussion and implications
This study provides in-depth information about the relationship between young voters and PEs
on social media. The discussion is structured as per the three research questions. A brief
summary of the findings is presented in Table 3.3.
Table 3.3 Summary of findings
Research questions
RQ1. What are the uses and gratifications
that drive young voters to follow PEs on
social media?
RQ2. What are the factors that drive social
media-enabled voter-PE relationships?
RQ3. What are the interactions that
underpin social media-enabled voter-PE
relationships?
Summary of findings
Primary gratifications include informational, social, and
entertainment. Instrumental, self-discovery, and civic
gratifications also exist.
Socio-emotional exchanges, trust, and meeting expectations
drive social media-enabled voter relationships.
Developmental, individuated, relational, and ethical interactions
underpin social media-enabled voter relationships.
3.6.1 Discussion
Uses and gratifications for following PEs on social media
The research found considerable evidence for informational, social, and entertainment needs,
along with some evidence of instrumental (guidance) and self-discovery needs. Overall, the
findings are in line with Parmelee and Roman’s (2019) recent research. Informational
motivations, i.e., information seeking and information sharing, are the predominant uses of
various political activities on social media (e.g. Macafee, 2013; Kim & Johnson, 2012). Our
study found much evidence for the former, particularly since our audience was educated
(Blumler, 1979). However, different types of information were sought from various PEs. For
instance, local councilors were followed for community-related information, whereas national
politicians and political parties were followed for personal and policy information respectively.
Despite information sharing being a common use of social media, our study found limited
evidence for this. The public nature of social media and the private nature of politics hindered
information sharing. Instrumental needs are the use of the medium for making decisions,
solving problems, and negotiating (Flanagin & Metzger, 2001). Most young voters
acknowledged that the activity of following PEs on social media, though not critical to their
voting outcomes, was important in forming opinions, evaluating PEs, and understanding
current issues. This demonstrates the instrumental and guidance value of this relationship
(Blumler & McQuail, 1969).
67
Self-discovery “involves understanding and deepening salient aspects of one’s self through
social interactions” (Dholakia et al., 2004, p. 244). Although this gratification is unexplored in
the political context, the research indicates that participants followed PEs to develop bi-partisan
thinking, tolerance, and empathy. Other motivators of engaging with PEs on social media are
entertainment and relaxation (Kim & Johnson, 2012). This is an important motivation for
following PEs on social media but only among younger voters (under 40) (Parmelee & Roman,
2019). The study found considerable evidence for this gratification. However, this need was
the driver for connecting with certain politicians rather than a universal motivation for
following PEs. Similarly, the entertainment value of the comments is recognized in the
literature and was validated by the participants (Whiting & Williams, 2013). Self-expression
is another motivation that drives the political use of social media (Park, 2013; Macafee, 2013).
Macafee (2013) found that ‘liking’ a candidate on social media was a way of self-presentation.
However, this study found little evidence for symbolic or actual self-expression as a motivation
for following PEs. This is in line with Parmelee and Roman’s (2019) research that found that
following politicians was a means for self-expression among less-educated voters, unlike our
participants. Although other researchers have also highlighted self-expression as a political
gratification, for instance, ‘liking’ as a means of self-expression (Dimitrova & Bystrom, 2017)
and self-expression as a motivation for opinion leaders on Twitter (Park, 2013), it is not a
motivation for following PEs. This confirms that gratifications vary with political activity and
audience.
Social gratifications involve the creation and maintenance of contact or relationship with
others. It includes other-related needs, such as getting to know others or staying in touch with
others (Flanagin & Metzger, 2001). Our conclusion that participants satisfied social needs
through this relationship is consistent with Ancu and Cozma’s (2009) findings that visitors of
politicians’ social media pages sought to gratify social needs. Finally, status enhancement, i.e.,
feeling important and impressing others, is also a gratification that social media users seek
(Ifinedo, 2016; Flanagin & Metzger, 2001). There was limited evidence that this motivation
drove engagement with PEs on social media. Only a few participants agreed that having realtime political information made them feel valued in their social circles.
In comparison with studies done on earlier media technologies (newspaper, TV, bulletin
boards), our study finds that entertainment has become a more prominent gratification for
political activities on media. Also, our study highlights that political self-expression and utility
68
of political information in inter-personal communications are not the motivations that drive the
political use of social media in young people (Garramone et al., 1986; Blumler & McQuail,
1969). These findings are plausible given the decline of political involvement and voter
engagement in young voters. Further, social media are more geared towards entertainment and
young voters are not willing to express themselves due to the public nature of social media and
the fear of a backlash.
Since uses and gratifications are media, context, and activity-specific, new gratifications
emerge as various activities are examined (Sundar & Limperos, 2013). This study found
evidence that some participants followed PEs because they perceived it to be the right thing.
This gratification of doing the right thing was due to a sense of civic obligation and
responsibility towards the country and the democratic process. This civic motivation has not
been investigated in the context of political activities on social media previously.
Factors driving social media-enabled voter relationships
Meeting expectations and fulfilling obligations are essential to the development of a
relationship (Rousseau & Tijoriwala, 1998). For instance, online brands that do not meet the
perceived expectations and obligations of their customers fail to develop a relational
psychological contract (Pavlou & Gefen, 2005). Our study identifies expectations and
obligations associated with major politicians from the perspective of young voters. These
include professionalism, honesty, leadership on issues, a personal approach, and an emphasis
on political education rather than political promotion. Although an average level of economic
exchange is a prerequisite to both standard and relational contracts, it is a high level of social
exchange that distinguishes relational contracts (Guo et al., 2017). Social exchanges stress the
social and emotional aspects of a relationship (Bal et al., 2010; Rousseau, 1995). The present
study confirms their importance in the development of relational contracts on social media.
Participants were highly desirous of personal and social content that depicted politicians’
everyday life and humanized them. Social and emotional elements have been acknowledged in
political marketing (Bruter & Harrison, 2017; Ormrod et al., 2013). Our findings confirm that
the presence of both the economic and the social exchange is essential for relationship survival,
however, social exchanges are its predominant driver (Gassenheimer et al., 1998). In addition
to social exchanges, trust and the perception of mutual interest are the main drivers of online
voter relationships. However, the findings indicate that major PEs were perceived to be driven
by self-interest and their activities on social media were characterized by distrust and
69
manipulation. The absence of trust and mutual benefit are denotative of captive contracts, a
deduction that is reflected in the lack of reciprocity offered by the participants (Guo et al.,
2017).
Minor parties and local politicians enjoy a better reputation among citizens (Murphy, 2017;
Fitzgerald & Wolak, 2016; Rockler, 2013). Their positive psychological contract with voters,
as demonstrated by our findings, is not only consistent with the literature but also explains their
stronger relationships with voters. Higher trust and perception of mutual interest reflect a
relational contract with minor parties and local politicians. Geographic proximity, familiarity,
and a community orientation helped local politicians develop stronger voter relationships than
mainstream politicians. As for smaller political parties, they were almost exclusively followed
for values-based propositions (environmental, anti-immigration, Christian values, rural
development, etc.). This alludes to the presence of a transpersonal psychological contract
between voters and smaller political parties. Transpersonal psychological contracts are strong
because they are based on shared values (Mason & Simmons, 2012).
Another plausible explanation of the voters’ comparatively stronger relationships with smaller
political parties is complexity (or lack of). Mainstream parties and politicians have more
obligations than minor PEs. They carry immense expectations whether they are in government
or opposition, and deal with a wider array of issues than minor PEs. It may be inferred that the
psychological contract becomes more complex as the organizational size and scope increases.
Studies show that the size of an organization has an impact on the psychological contract and
a small firm size intensifies relationships (Nadin & Cassell, 2007). Therefore, it is a logical
assumption that numerous party voices (leaders, officials, and candidates), greater expectations
and obligations, a wider scope of issues, long-established legacies, and practical realities of
politics and governance increase the complexity of the psychological contract and impede the
creation of relational contracts between voters and major PEs. Contrarily, with their niche focus
on a single issue, minor parties like Greens, One Nation, etc., are able to form a simple and
strong psychological contract.
It should be noted that the online relationships between young voters and political entities
reflect the characteristics of a weak relationship. For instance, unmet expectations and
unfulfilled obligations did not result in psychological contract violations or termination of the
online relationship (unfollowing PE). This highlights the negligible impact that psychological
70
contract breaches had on emotions and relationship continuity, which indicates an uncommitted
and weak relationship (Montgomery et al., 2018). Literature shows that following a PE is a
passive activity that has a low cost associated with it (Dimitrova & Bystrom, 2017; Cameron
et al., 2016). This offers some explanation regarding why PEs do not lose followers despite
major scandals and swings in popularity.
Interactions underpinning social media-enabled voter relationships
Interactions form the basis of customer relationships and service delivery (Karpen et al., 2015;
Grönroos, 2004). Accordingly, PEs that are able to generate meaningful online interactions
with voters are better positioned to create strong relationships. Overall, our study reveals that
major PEs are not offering young voters the interactions that they value. Minor PEs, however,
are more proficient at providing valued interactions.
Major PEs were perceived to have an insufficient understanding of the needs of young voters.
In the absence of a personalized or an individuated understanding, relationships are hard to
foster (Bojei et al., 2013; Gordon et al., 1998). Considering that young voters are disengaged,
not a reliable segment, and have a lower likelihood of taking part in elections (Wright &
Koslowski, 2019; Hannan-Morrow & Roden, 2014), major PEs might be more inclined
towards segments that are more likely to vote on election day. Additionally, the long-held belief
that PEs join social media to target young voters might not be accurate today. Social media
demographics have changed over time along with social media’s importance in politics. A
positive perception of minor PEs’ individuated capabilities on social media is entirely plausible
as these PEs have been known to target and attract young voters (Murphy, 2017; Rockler,
2013). Social media are known to facilitate relational interactions. The findings suggest that
socio-emotional interactions supported online relationship building. Personal and social
content that sketched the human aspect of major politicians was favored by participants. The
importance of enhancing the social and emotional value of interactions is documented in
marketing and political marketing literature (Vargo & Lusch, 2016; Ormrod et al., 2013). PEs’
perceived failure in offering two-way interactions and engagement, an essential element of
relationship (Brodie et al., 2011), is also consistent with the literature since most PEs do not
respond to or interact with voters on social media (Lees-Marshment, 2014).
Ethical interactions create trust, the primary ingredient of relationships (Morgan & Hunt,
1994). Since, major PEs were deemed opportunistic, manipulative, and self-serving in their
71
online interactions, strong relationships that evoke reciprocity are unlikely. Local politicians,
as well as minor parties and their candidates, were deemed trustworthy. Studies support our
findings since local politicians are deemed more trustworthy and distrust in major PEs is a
major reason for favoring minor PEs (Wood & Daley, 2018; Fitzgerald & Wolak, 2016). Social
media are also being used to empower citizens. Local governments are using it to invite
feedback from citizens and politicians are hosting live Q&As on it. The participants felt that
Australian PEs do not provide opportunities for participation and cooperation. Likewise,
participants strongly felt that major PEs are failing to develop young voters. Many commercial
entities use their social media presence to create strong online customer relationships by
developing their customers (Eisingerich & Bell, 2006). For example, MAC Cosmetics offers
tutorials on applying make-up on their YouTube channel. Our study found substantial evidence
for the developmental potential of social media interactions. Lastly, multiple contact points
exist in businesses (e.g. sales staff, customer service staff, etc.) as well as politics (e.g. local
members, federal members, party officials, etc.), which raises the significance of concerted
interactions (Karpen et al., 2012). A concerted effort is needed to cultivate relationships (Sin
et al., 2005). The study indicates that coordinated interactions and communications increase
the belief in the achievability of the agenda.
3.6.2 Research implications and limitations
The study highlights that various gratifications drive voters to initiate online relationships with
political entities. More importantly, the research implies that before adopting a quantitative
approach in UGT studies, a qualitative study should be carried out to uncover the activity,
platform, or audience-specific gratifications that are not being captured by established research
instruments, since most of these have been devised for earlier media technologies or activities
(Sundar & Limperas, 2013). For instance, the gratification of self-discovery is not included in
similar studies. Also, despite utilizing Flanagin and Metzger’s (2001) comprehensive twentyone item research instrument as a guide, the study uncovered a civic gratification that had not
been explored previously. Furthermore, the study highlights that other than identifying new
gratifications, findings from qualitative UGT studies provide nuanced and refined insights. For
example, the literature suggests that getting information is a dominant gratification for online
political activities, however, the type of information (policy, personal, issue, or community)
voters expect from various PEs is different, as demonstrated by this study.
72
The extension of the psychological contract to understand voter relationships is unprecedented
in literature. Like employee and consumer relationships, voter relationships are also governed
by the psychological contract. The findings confirm that although the economic aspect of the
exchange is important, social exchanges and trust are the main factors in developing
relationships (Guo et al., 2017). Further, the study affirms that the type of psychological
contract voters develop may be dependent upon contextual factors like party size, personal
values, complexity, and scope of obligations (Nadin & Cassell, 2007). Likewise, the voter’s
psychological contract may be a function of PE’s service value proposition (Mason &
Simmons, 2012). Minor parties and local politicians are more likely to invoke voters using
values or ethics-based appeals, thus increasing the probability of developing relational and
transpersonal psychological contracts with voters. Each voter-PE online interaction has the
potential to strengthen the relationship. However, certain types of interactions, i.e., relational,
developmental, individuated, and ethical interactions, are of greater significance when
cultivating online relationships with voters.
Considering the small sample size, the use of a student sample, and the qualitative methodology
used, the generalizability or transferability of the study is its primary limitation. The findings
might not be as relevant to other voter segments as they are for young educated voters.
However, we provide thick descriptions for those who seek to transfer the findings to another
context or group (Morse, 2015). Future researchers intending to explore this activity using a
quantitative methodology should consider the gratifications proposed in this study, especially
those researchers who are employing a student sample since gratifications are audiencespecific. Additionally, the impacts of a PEs’ service-dominant orientation and a voter’s
psychological contact on voter behavior merit further investigation.
3.6.3 Managerial implications
The study offers practical suggestions to PEs that are attempting to engage young voters on
social media. Firstly, PEs should take a strategic and relational approach to social media use.
This entails continuous use of social media outside the election cycle. The authors recommend
a personal, social, and bipartisan approach that attempts to develop voters. Secondly, followers
have different gratifications and PEs need to cater to this by offering relevant and diverse
content. Additionally, young voters prefer a more intimate and social relationship with
politicians. Content that conveys politicians as everyday people and humanizes them may lead
73
to such a relationship. Although a personal and social approach is preferred by young voters,
professionalism is expected. For instance, personal posts should not coincide with tragic events
(natural disasters, major accidents, etc.) and should be free from errors. Additionally, PEs need
to be more interactive. For example, they need to respond or react to followers on social media
(Donald J. Trump regularly did this). Major PEs’ social media strategies should aim to cultivate
trust and the perception of mutual benefit by offering ethical and developmental interactions.
Notably, a churn of social media followers might not be a good measure to gauge popularity
as voters rarely unfollow PEs. Also, following a PE is not an expression of preference. The
impact of grass-root politics should not be ignored. Local politicians enjoy a positive
relationship with voters on social media due to geographic proximity, personal familiarity, and
personal similarity. Major political parties can benefit from this positive relationship by
integrating local politicians in their social media strategies and by providing training and
resources to local politicians. Local politicians’ strategies should highlight their community
ties and provide updates on the community. Minor PEs enjoy a relatively strong relationship
with voters due to their values-based propositions. The authors recommend a social media
strategy that emphasizes these values.
3.6.4 Societal implications
Given the pervasive effects of social media and politics on society, our study has implications
that go beyond political marketing. Young voters are a substantial segment of the electoral
market. Their distrust in major political entities, reluctance to reciprocate, and the perception
of not being an important voting segment are bad omens for the future of Australian democracy,
which is dependent upon voter participation and engagement for legitimacy. Recent times have
seen a worrisome decline in political trust, satisfaction with democracy, and commitment to a
single political party among Australian voters (Evans et al., 2018; Wood & Daley, 2018),
variables that represent a relationship’s strength or quality (Clark et al., 2017). Although our
study recommends that politicians should adopt a personal and social approach to engage and
build strong relationships with young voters on social media, this has implications for our
society, since politics, elections, and democracy are not popularity contests. Finally, young
voters consider social media as developmental avenues. This should reflect in the strategies of
electoral commissions and political institutions. The authors recommend these authorities
adopt an educational, developmental, and individuated approach to engage young voters.
74
CHAPTER 4: A RELATIONHIP MARKETING
ORIENTATION IN POLITICS: YOUNG VOTERS’
PERCEPTIONS OF POLITICAL BRANDS’ USE OF SOCIAL
MEDIA
Abid, A., Harrigan, P., & Roy, S. K. (2021). A relationship marketing orientation in politics:
Young voters’ perceptions of political brands’ use of social media. Journal of Strategic
Marketing, 29(4), 359–374. https://doi.org/10.1080/0965254X.2020.1777457
Chapter 4 is the third study of the research project. It is a qualitative study that attempted to
understand the perceptions of young voters regarding social media activities of political brands.
The two research questions that the study sought to answer are presented in Figure 4.1.
How can political brands
strengthen their relationships
with voters on social media?
CH3. What is the nature
of social media enabled
voter relationships?
RQ1. What are the
gratifications that drive
voters to follow political
brands?
RQ2. What are the drivers
of social media-enabled
voter relationships?
RQ3. What are the
interactions that underpin
social media-enabled
voter relationships?
CH4. What is the
marketing orientation
that political brands
adopt towards social
media marketing?
CH5. What is the role of
marketer-generated
content in driving online
voter relationships?
RQ1. Do political
brands adopt a
relationship marketing
orientation towards
social media
marketing?
RQ2. Are there any
dimensions that are
not represented in
RMO framework?
RQ1. What is the effect
of various content cues
on online relationship
quality?
RQ2. Does content
curation have an
impact on online
relationship quality?
Figure 4.1 Chapter 4 (Paper 3)
75
CH6. What are the roles of
marketer-generated
content and behavioral
engagement in driving
online voter relationships?
RQ1. What are the
effects of various
content cues on
engagement and online
relationship quality?
RQ2. Does behavioral
engagement mediate the
effect of content cues
on online relationship
quality?
4.1 Abstract
This study explores young voters’ perception of social media use and activities of political
brands in light of their general apathy towards traditional politics. Focus groups were
conducted with young voters to collect data. Relationship marketing orientation served as the
study’s underpinning theoretical framework. Findings suggest that young voters perceive that
mainstream political brands are not utilizing social media to build online relationships.
However, smaller political brands enjoy a positive perception. This is due to them excelling on
the dimensions of bonding, trust, and shared values. The research provides executable
recommendations to political brands and their social media teams. The research also identifies
and proposes ‘customer education’ as the eighth dimension of Relationship Marketing
Orientation.
76
4.2 Introduction
Social media are an essential component of contemporary political marketing. Social media
allow politicians and political parties to communicate, campaign, and connect with voters.
Likewise, following political parties, political candidates, or elected officials (referred to as
political brands from here on) is among the most common activities on social media (Newman
et al., 2017). Unlike commercial brands, most political brands have been ineffective in
exploiting social media’s potential to build online relationships and continue to adopt a
traditional approach to political marketing on social media (Harris & Harrigan, 2015). The
aforementioned conclusion has been confirmed across various North American and European
contexts (e.g. Abid et al., 2019; Parsons & Rowling, 2018; Lees-Marshment, 2014). However,
due to the research subjects chosen or the methodologies employed in these studies, the voter’s
point of view is absent. This study fills this gap by exploring young voters’ perceptions of the
social media activities of political brands. This is done by utilizing Relationship Marketing
Orientation (RMO), an established framework that is yet to be utilized in non-commercial and
online contexts.
Declining political participation, distrust of political institutions, and voter apathy is a
consistent trend across many countries, and this is particularly true for young voters (Dalton,
2013; Martin, 2012). Unlike political marketing, political relationship marketing (PRM) has
the potential to reduce political distrust and revive political engagement. This is because PRM
is concerned with the long-term welfare of society (Ormrod et al., 2013; Bannon, 2005).
Furthermore, when compared with traditional political marketing, PRM is more compatible
with social media (Lees-Marshment, 2014; Jackson & Lilleker, 2009), justifying an intricate
understanding of online voter relationships and the ways to strengthen them.
This study employed focus groups to explore young voters’ perceptions of the social media use
of political brands based on the RMO framework (Sin et al., 2002; Callaghan et al., 1995). The
study sought to investigate whether the RMO dimensions are indeed exhibited in the social
media activities of political brands. Also of interest were any new dimensions that helped
explain the perceptions of young voters. Social media’s ubiquity among youth and the
influence social media exert on their relationships with brands make young voters an ideal
subject for this study. The study aims to further research in the field of online PRM.
Additionally, it extends the RMO framework to novel contexts. Relevant literature is reviewed
77
next. This is followed by a description of the research methodology. The subsequent findings
and discussion are structured as per theory. Finally, the various implications and limitations of
the research are considered.
4.3 Literature review
4.3.1 Political relationship marketing and social media
Literature advises the development of long-term relationships with various stakeholders that
operate in the political arena (Hughes & Dann, 2009), and relationship marketing can achieve
this by transforming the voter-political brand exchange from an economic exchange to a social
exchange (Ormrod et al., 2013). Additionally, relationship marketing is more apt when the time
between two purchases (i.e., elections) is substantial (Egan, 1999). At the macro level, PRM
can reduce electoral volatility and stimulate voter involvement, loyalty, and trust. At the microlevel, activities like target marketing, CRM, merchandising, and fund-raising make PRM
relevant (Henneberg & O’Shaughnessy, 2009).
Social media require political brands to adopt an approach based on PRM. The literature
recommends a relational, interactive, and dialogue-based approach to social media in politics
(Lees-Marshment, 2014; Jackson & Lilleker, 2009). Harris and Harrigan (2015) note that social
media are more effective for relationship building than campaigning. Despite this, political
brands are not utilizing social media to build relationships. For instance, Small’s (2012) online
content analysis revealed that Canadian political brands were inconsistent in the application of
relationship marketing. Likewise, Jackson et al. (2012) determined that the websites of British
political brands failed to adhere to the principles of relationship marketing. Studies in the
Greek, UK, and US contexts also confirm the absence or limited application of relationship
marketing principles (Harris & Harrigan, 2015; Williams & Gulati, 2014; Papagiannidis et al.,
2012). Parsons and Rowling (2018) concluded that political brands fail to adopt a relationshipbased approach to social media marketing because they fear the lack of control on social media,
have limited resources, and are skeptical of social media’s return on investment. From a
methodological perspective, almost all of the aforementioned research is either conceptual,
presents the brand’s perspective, or is based upon a content analysis of political brands’
websites or social media pages. Thus, a first-hand voter perspective is missing in the domain,
which limits the guidance that is available to politicians and political parties. The authors
78
contend that RMO provides an easily executable and theoretically rigorous framework to
evaluate the social media activities of political brands and explore their social media marketing.
4.3.2 Relationship Marketing Orientation
The concept of Relationship Marketing Orientation (RMO) was introduced and solidified to
operationalize the broad philosophy that underpins relationship marketing (Sin et al., 2002;
Callaghan et al., 1995). RMO denotes the effort a company puts into developing long-term
relationships with customers (Tse et al., 2004). Companies that adopt a relational orientation
excel in the dimensions of empathy, reciprocity, bonding, trust, shared values, and
communication (Sin et al., 2002). Recently, a seventh dimension, harmonious conflict
resolution, was proposed by Kwan and Carlson (2017). RMO’s positive relationship with
various desirable brand, financial, organizational, and marketing outcomes is established in the
literature (e.g. Kwan & Carlson, 2017; Yoganathan et al., 2015; Hau & Ngo, 2012; Sin et al.,
2006). The impacts of the different RMO components are dissimilar and context-dependent.
For instance, Hau and Ngo (2012) found no evidence for communication’s and empathy’s
influence on customer satisfaction, whereas Yoganathan et al. (2015) found communication
and empathy to have a positive impact on brand equity.
RMO’s applicability in various service contexts (e.g. Kwan & Carlson, 2017; Yoganathan et
al., 2015) makes it pertinent to politics since a politician or a political party is essentially a
service provider (O’Cass, 2009). Specifically, major political brands (major political parties
and national politicians) in government are more likely to benefit from adopting a RMO as it
is more critical for market leaders (Tse et al., 2004; Yau et al., 2000). Additionally, RMO is
compatible with social media, which are relational tools (Harris & Harrigan, 2015). Literature
from commercial marketing suggests that a relational orientation will not only assist political
brands in realizing political objectives (like market share or growth) but will also increase voter
retention, satisfaction, and trust (Sin et al., 2006). Therefore, RMO is an attractive proposition
that benefits political brands, voters, and democracy. Lastly, the components of RMO apply to
the political context and several have been explored in the political science literature.
Components of Relationship Marketing Orientation
Following is a brief explanation of the seven components of RMO and their relevance to the
political context.
79
Bonding. Bonding results in two parties acting in a unified manner towards a desired goal
(Callaghan et al., 1995). It involves creating a sense of belonging and affection towards the
relationship, keeping in constant touch, close cooperation, and long-term orientation (Sin et al.,
2006; Sin et al., 2002). Social media facilitate bonding by expediting cooperation (Hajli et al.,
2017). Further, the self-disclosing nature of social media interactions increases intimacy and
bonding and fosters parasocial relationships between followers and those they follow (Chung
& Cho, 2017). Although bonding has not been studied in the political marketing context, the
renowned ‘Obama 2008’ presidential campaign exhibited many elements of bonding like
constant updates, dialogue, and long-term orientation (Cogburn & Espinoza-Vasquez, 2011).
It is noteworthy that Barack Obama’s account remains the most followed account on Twitter
with over 113 million followers.
Trust. Trust is one’s “confidence in the exchange partner’s reliability and integrity” (Morgan
& Hunt, 1994, p. 23). Trust is a widely studied variable in political science. Political trust is on
the decline, which is worrisome since trust is a precondition for democracy and government.
The trend is more pronounced in Australia where political brands enjoy a low level of trust
among the public (Koziol, 2017). Research suggests that relational activities (like building
online brand communities) lead to higher customer trust (Habibi et al., 2014). Likewise, social
media may be used to build trust in public institutions (Warren et al., 2014).
Communication. Effective communication leads to stronger relationships, customer loyalty,
relationship success, trust, commitment, and satisfaction (Palmatier et al., 2006). Political
communication is an important theme in political science and political marketing research.
Being an effective communicator might be more important than a politician’s governance,
management, and policies (Flanagin & Metzger, 2017). However, unlike traditional mediums,
social media require politicians and political parties to adopt an interactive approach that
encourages user-generated content and dialogue (Schivinski & Dabrowski, 2016; LeesMarshment, 2014). In fact, Facebook encourages politicians and political parties to get
involved in ‘Q and A’ sessions, engage voters, and reply to constituents (Facebook, 2018).
Shared values. Shared values are the “extent to which partners have beliefs in common about
what behaviors, goals, and policies are important or unimportant, appropriate or inappropriate,
and right or wrong’’ (Morgan & Hunt, 1994, p. 25). Shared values are an important part of
80
modern politics, even more so than issues and policies (O’Shaughnessy, 2004). Each political
brand, much like a consumer brand (Zhang & Bloemer, 2008), represents distinct values. An
ideal example is that of the two American political parties, both representing a distinct set of
shared values. The Republican Party represents traditional, Christian, and conservative values
and free-market capitalism, whereas the Democratic Party emphasizes economic fairness,
racial and gender equality, and pro-environmental values.
Reciprocity. Reciprocity requires that a party makes allowances for the other in return for
allowances that may be received later (Callaghan et al., 1995). Reciprocity may be as simple
as giving a gift to customers (Alrubaiee & Al-Nazer, 2010). Reciprocity is difficult to
operationalize in political marketing as the political exchange is an open exchange that involves
imbalanced reciprocity because what voters and political brands exchange is neither equivalent
nor immediate (Hoppner et al., 2015; Ormrod et al., 2013).
Empathy. The use of empathy in marketing and advertising is common (Kemp et al., 2017).
Empathy allows the exchange partners to see the relationship from the other party’s perspective
by seeking to understand their goals and desires (Sin et al., 2002; Callaghan et al., 1995).
Empathy is a desirable trait in political candidates also (Guzmán & Sierra, 2009; Kinder, 1986).
Furthermore, empathizing is an effective course for political persuasion (Feinberg & Willer,
2015).
Harmonious conflict resolution. Conflict resolution is defined as “the extent to which
disagreements are replaced by agreement and consensus” (Robey et al., 1989, p. 1174).
Although conflict resolution has not been studied in the political context, political brands will
benefit by devising conflict resolution mechanisms as conflict resolution is linked to a
relationship’s success, partners’ commitment, and performance (Blumenberg et al., 2008).
Social media are fast becoming a medium where consumers share their reviews, file
complaints, and expect resolutions (Gunarathne et al., 2017).
4.4 Research methodology
The study adopted a qualitative approach that utilized focus groups. Given the limited research
in the field of online voter relationships, a qualitative approach is ideal (Sofaer, 1999). Focus
groups are a common form of research in political marketing (Lees-Marshment, 2014) and
81
work well with student samples (Palomba & Banta, 1999). The questions investigated the
various dimensions of online voter relationships and broadly captured the different items
highlighted in the RMO framework (Kwan & Carlson, 2017; Sin et al., 2002).
Three focus groups were conducted at an Australian university. The focus groups satisfied the
various duration, size, and homogeneity requirements highlighted in the literature (Krueger &
Casey, 2014). Twenty-four participants took part in the study. Males comprised 54 percent of
the sample. Participants were under the age of 26 and were incentivized with a coffee voucher
and refreshments. Since the study aimed to investigate the perceptions of young voters,
university students were considered an ideal sample. Although the use of student samples is a
debated issue, it remains popular in top social sciences and marketing journals (Kees et al.,
2017; Pelletier & Sharma, 2015). More importantly, the concerns regarding the use of student
samples, i.e., lack of external validity and generalizability of research, were less relevant to
this qualitative study which does not seek generalizability (Kees et al., 2017; Ok et al., 2008).
The only sampling criteria was the requirement that participants must be following one or more
political brands on social media.
More specifically, a theory-driven, deductive approach was used to analyze data using
predetermined codes based on theoretical constructs (Braun & Clarke, 2006). A deductive
approach is appropriate in situations where an established theory is being tested in a novel or a
different context (Elo & Kyngäs, 2008). This approach is commonly used for qualitative
research in marketing (e.g. Sanghvi & Hodges, 2015; Chatzidakis et al., 2004). RMO’s seven
components formed the basis of the pre-determined codes. This was followed by directed
content analysis, a structured process that is appropriate for extending or supporting the
existing theory (Hsieh & Shannon, 2005). New data codes were established to analyze data
that did not fit into the predetermined codes. Finally, the presentation of findings and
subsequent discussion were steered by the theory (Hsieh & Shannon, 2005). The authors
ensured the study’s validity and reliability by following guidelines on theme formation, checks,
coding, and presentation of findings (Brod et al., 2009; Miles & Huberman, 1994). For
instance, validity was achieved by deriving focus group prompts from validated research
instruments. Rich and thick extracts were used to enhance the study’s credibility (Noble &
Smith, 2015). Reliability was ensured by having a second researcher verify the primary
researcher’s coding.
82
4.5 Findings
The present study sought to understand young voters’ perceptions of the social media activities
of Australian political brands using the RMO framework. The findings are presented in Table
4.1 and are structured as per the components of RMO. However, an additional dimension was
needed to capture the unplanned data that emerged during the analysis. This dimension is
termed ‘customer education'. Smaller political brands refer to local politicians and minor
political parties.
83
Table 4.1 Focus group findings
Trust
Bonding
Communication
Shared values
Findings
Participants demonstrated low trust in major political
brands on social media.
Participants were doubtful of major political brands’
intentions on social media.
Participants characterized the social media
environment as deceptive and manipulative.
Participants noted self-serving and partisan use of
social media by major political brands.
Participants exhibited trust in the social media
activities of smaller political brands.
Participants noted the short-term focus of major
political brands on social media.
Participants expressed dissatisfaction over the lack of
cooperation and involvement opportunities offered by
major political brands on social media.
Participants praised the long-term, mutually valuable
approach of smaller political brands.
Participants described various ways in which online
cooperation can take place.
Participants perceived political brands as using social
media for one-way communication.
Participants acknowledged that the social media
environment does not allow for a thoughtful
conversation between followers and political brands.
Participants described the communication of major
political brands as partisan and manipulative.
Participants felt that major politicians might not share
the same values as young voters due to an age gap.
Participants deemed smaller political brands to be
better at highlighting shared values.
Participants preferred content that symbolized
universal or Australian values.
Participants liked personal content because it helped
them see major politicians as regular people.
Sample quotes
I know social media, that behind one single post there are ten people reviewing it and making
sure that it is what attracts the target audience. That’s why I do not trust politicians on social
media because you can’t tell if something is done by PR or politician.
On the local level, there is this one person that I follow who always gives out their mobile
number, like if you ever need anything call me immediately. I think it’s very trusting or they
have a second phone but it’s a display of trust.
It is (political brands on social media) like everything on the web. In its very soul, you can’t
trust what you see. As I said before, I try to educate myself as much as possible, ill crossreference them.
Some federal politicians who are very community-focused, you will see them post things
throughout their term, things like ‘oh I went to this fair’ or whatever, but some other
politicians I see only post in the months leading up to the election because they see it as a way
of engaging with the voters during election time and not about maintaining the relationship,
and I think that’s the big difference from local politicians.
For me, it is a mutually beneficial relationship (following Greens) because I believe in the
ideology and I want to support them and they need me to support them.
For the local government, we can have online competitions or polls for like what is good for
the city.
On top of the backlash (for posting an opinion), how would you have a public discussion
online? Post a Facebook post: I love Trump, please comment.
And they will, like, set up their own website and use social media to get you to go there, then
they might even use email marketing and things like that. Like spam.
They want to persuade the voters to vote for their ideas. I think that’s their only goal.
It really makes him relatable as a person, like all the first names. He actually talks about Lucy
and actually shows like fun things, and I really like it.
Sometimes politicians have failed because they come across as impersonal. Julia Gillard was
really disliked because she seemed too professional. And then you get a video of someone who
is riding a Harley Davidson and they are like vote for me. I don’t know, it is cool.
I liked it. It was the one where he had the baby in his one hand and a beer in the other, and he
talked about how he could sit down and finally watch the game.
84
Empathy
Reciprocity
Conflict resolution
Education
Participants felt that major political brands did not
understand the concerns of young voters.
Participants empathized with politicians and
understood their resource limitations on social media.
Participants blamed political parties’ legacies and
stances for the apparent absence of empathy in
politicians.
Participants believed that major political brands are
driven by main voting segments and provide limited
attention to students.
Participants were not willing to reciprocate either.
Participants agreed that social media are
not a good avenue to resolve conflict and considered
offline options better.
Participants considered responding to a political
brand on social media to be futile.
Participants rarely unfollowed political brands on
social media due to conflicts.
Participants recognized the learning potential of
social media and agreed that major political brands do
not attempt to educate voters. However, smaller
political brands were seen in a positive light.
Participants also acknowledged that social media
might not be conducive to an in-depth education on
complex economic or policy matters.
Participants praised local politicians and smaller
parties for education on community and issues
respectively.
Things like politicians seem to think like it is affordable to live. There have been examples like,
just stop drinking coffee. I grew up in hypocrisies like the government want to increase fees for
higher education but when they were going through universities it was free. So there is a bit of
anger also.
I mean the youth vote is not big, especially on a federal level, maybe on the council level, but
on a federal level, it doesn’t make a difference.
It was a party policy around ten years ago that the Labor Party was opposed to same-sex
marriage and now party policy has changed and they are in favor of same-sex marriage. The
Liberal Party was quick to catch on to that… remember when you made Penny Wong (a
Labor politician who was in a same-sex relationship) say that she was against same-sex
marriage.
If it is a locality where a lot of students live they are more likely to go towards the students, if
there are more old people they will go to old people. They will target the majority.
It depends. I will only share their post if it’s a serious issue like a vote for a war, but it has to
be damn important. It would be for my own interest, regardless.
I think there are too many opinions and conflicting stuff on social media so if you did have
some sort of an issue with the politician and you want to talk about it, social media is not the
best way to do it.
Well if you are a top comment, I do expect a response. If you have a thousand likes behind
your comment, they should respond. Because it makes a difference, your opinion is now
validated by a thousand people.
I unfollowed Clive Palmer out of principles. As much as I like his memes and all, the things
that he has done in Queensland, I cannot continue to like his pictures of his dogs.
I have certainly used information that I have got from them to get to an opinion and kind of
discuss with other people.
Sometimes it doesn’t translate that well on social media because like no one is going to be
scrolling on their Facebook feed and be like I am going to read it. Most people are not going
to read that. And because the posts are like pdfs usually, links to giant reports or articles. No
one is really… everyone just reads the headlines on Facebook.
Politicians are not developing us. They want votes, they want to sway people to their side. I
think that when you form an opinion when you are at this age, you know in your 20s, chances
are that it will stick. We don’t know much of the world yet.
85
4.6 Discussion and implications
The findings suggest that the major Australian political brands do not apply a relational
approach to social media. This is consistent with earlier research (Harris & Harrigan, 2015;
Small, 2012). The findings show that young voters perceived major political brands to be
engaged in a sales-based approach on social media and driven by their self-interests.
Contrariwise, smaller political brands enjoyed a positive perception, particularly in regards to
bonding, shared values, and trust. The findings are reflected in the performance of minor
political parties in the recent Australian federal election that saw one in four Australians vote
for minor parties (Salisbury, 2019). Not only are minor political parties frequently credited
with effective marketing (Lees-Marshment, 2014; Maier & Tenscher, 2009), they exert greater
effort on social media and are more interactive, which is understandable as they lack the
resources required for traditional media (Kalsnes, 2016). Likewise, a positive perception of
local politicians is consistent with the literature (Fitzgerald & Wolak, 2014).
4.6.1 Dimensions of RMO
A relationship is implausible in the absence of trust. The same stands true for online
relationships (Verma et al., 2016). Considering the historically low levels of trust in politicians
(Tillett, 2019), it is unsurprising that their social media activities are also distrusted. Apart from
a general skepticism of information on social media, perceptions of opportunistic behavior, use
of half-truths, and misleading information eroded trust in major political brands. Smaller
political brands, however, were able to generate trust through their social media activities. The
conclusion is validated by the literature that finds local politicians are trusted more than
national ones (Fitzgerald & Wolak, 2014). Similarly, the rise of Australian minor parties (since
2007) coincides with an increase in distrust towards major political brands (Grattan Institute,
2018). Along with trust, a major reason for a generally negative perception of the major
political brands was the absence of bonding. Lack of cooperative opportunities and a shortterm focus (on ratings and elections) hindered the formation of genuine bonds (Callaghan et
al., 1995). Online cooperation is important as the resulting bonds transfer into offline support
and reduce doubt in the relationship (Quinton & Harridge-March, 2010; Chattananon &
Trimetsoontorn, 2009). Major Australian political brands can learn from examples like Donald
J. Trump’s #MAGACHALLENGE and Bernie Sander’s #MyBernieStory. Smaller Australian
political brands fared better in this dimension also. Participants perceived them to be long-term
86
orientated and driven by mutual benefit. The narrow community focus of local councilors and
a single-issue approach of minor political parties facilitated this perception.
Political marketing communication is most effective when used to form emotional and human
connections (Newman, 2001). The findings demonstrate that this is correct for political
communication on social media also. Participants expressed aversion to persuasive messages
(Lilleker & Jackson, 2014) and sought neutral communication. A communication approach
that was dominated by social and personal content was preferred by participants as it allowed
the cultivation of emotional connections and made politicians seem relatable. Expectedly,
political brands failed to indulge in two-way communication. Multiple studies offer a similar
conclusion (Small, 2012; Small, 2008). Smaller political brands have human and resource
limitations, however, major political parties can learn from commercial brands which respond
to their customers on social media. Although not witnessed in our study, research shows that
minor political parties are more likely to respond to their followers (Kalsnes, 2016). Shared
values are an important aspect of politics and modern elections are increasingly becoming a
contest of competing values (Ormrod et al., 2013; O’Shaughnessy & Henneberg, 2002). Online
communities, like a political brand and its followers, are built upon shared values (Habibi et
al., 2014; Zhou et al., 2012). Content that portrayed shared values and activities was well
received by the participants. However, it was the smaller political brands that excelled in this
dimension. As discussed earlier, by emphasizing one shared value, these political brands
appeared to have created an emotional relationship with young voters who are more likely to
be driven by single-issue politics (Quintelier, 2007).
Despite being a desirable trait in politics (Guzmán & Sierra, 2009), politicians and political
parties failed to empathize with young voters on social media. Empathy requires an
understanding of the young voters and the participants perceived this to be absent in political
brands. It was further observed that the political party’s values and political stances made it
difficult to exhibit empathy. This is congruent with past research that shows empathy to be a
taxing task that involves role conflicts (Varca, 2009). An illustration of this is Republican
politicians (USA) finding it difficult to demonstrate empathy with victims of gun violence.
Reciprocity is difficult to operationalize in the political context for many reasons, including
candidate or party failure (Egan, 2005). The participants felt that political brands did not offer
any consideration to young voters and students. Furthermore, reciprocity is two-way, and
participants were not willing to reciprocate beyond social media either. Lastly, harmonious
87
conflict resolution requires a joint problem-solving approach (Mohr & Spekman, 1994),
however, social media were deemed incompatible for this due to clutter and conflicting
opinions. There was a consensus that conflicts rarely led to the termination of online
relationships with political brands. This is explicable since following political brands on social
media is a passive activity (Dimitrova & Bystrom, 2017).
Customer education
The data revealed a dimension that was not being captured by the current RMO framework.
Relationship marketing literature has emphasized educating customers (Eisingerich & Bell,
2006). Gummesson (1998) notes that educating the customer is not only a viable customer
strategy but is a privilege. Customer education may be defined as the provision of knowledge
and skills that are needed to understand and use information critically (Bell et al., 2017). It
remains an under-researched theme in relationship marketing despite having a positive impact
on customer loyalty and trust (Suh et al., 2015; Eisingerich & Bell, 2006), and this is
demonstrated by its absence from the RMO framework. As companies increasingly involve
customers in service co-creation or delivery, the provision of firm-specific and market-related
education becomes necessary (Bell et al., 2017; Karpen et al., 2015). Additionally, certain
sectors, like IT and financial services, offer increasingly complex offerings, thus mandating
the education of customers. Social media are ideal for customer education. Presently, many
businesses are utilizing social media to educate their customers. For instance, software
companies like QSR International offer NVivo training on YouTube. Additionally, since
customer education leads to higher customer engagement on social media (Grewal et al., 2017),
the authors advise political brands to integrate voter education in their social media marketing.
However, there is a caveat. It is only firm-specific education that leads to customer loyalty.
Market-related expertise may have an adverse impact on loyalty (Bell et al., 2017).
Nevertheless, politicians and political parties ought to focus on market-related expertise also,
since the goal should be an informed public and not just loyal voters.
4.6.2 Research, managerial, and societal implications
The present study extended the RMO framework to political and social media contexts.
Additionally, the research furthered the integration of relationship marketing concepts into
PRM. Social media have the potential to revitalize relationship marketing practice and
literature (Sheth, 2017). The same might be true for political relationship marketing also.
88
Education is expected by modern consumers. Consequently, it should be included in the RMO
framework as its eighth component. Surely, a relationship marketing approach should entail
unbiased education of the customer. Such an approach is in line with the actor-to-actor
orientation that is propagated in the services literature (Wieland et al., 2012) and the
contemporary paradigm of corporate social responsibility also. Secondly, the study reaffirms
the inability of major political brands to adopt a relational approach on social media (Parsons
& Rowling, 2018; Harris & Harrigan, 2015), while confirming the more relational approach of
the smaller political brands. Consistent with prior research, our research highlights that the
relative importance of RMO dimensions is context-dependent, and reciprocity and conflict
resolution might not be pertinent to the political social media setting. From a managerial
perspective, we advise political brands to utilize social media’s educational potential in order
to cultivate trust. Further, the authors advise personal disclosure and demonstration of values.
Frequent postings outside the campaign period, online contests or polls, a neutral approach,
and occasional interaction with followers are also recommended to bond with young voters.
Simple actions like sharing, liking, or replying to a follower’s content are also advised.
Social media might be effective in reducing the political disillusionment in young voters,
although, this will not be through public debate and political participation but through the
creation of emotional relationships between political brands and voters, relationships that have
little to do with political policies and issues. Similarly, social media contests and personal
content may diminish the solemnity of the political process and democracy, a haunting thought
for political scientists (Ormrod et al., 2013), but the low involvement of the young voters
requires such an approach. Ultimately, the future of democracy rests on the engagement and
participation of young voters. The successful use of social media by minor political parties and
major politicians like Donald J. Trump and Bernie Sanders is likely to further push the elections
into the hands of digital marketers. A case in point is Brad Pascale, head of digital marketing
for Donald J. Trump’s 2016 campaign and the campaign manager for his 2020 campaign, who
had no political experience before joining the campaign and treated the campaign of 2016 as a
data-driven B2C campaign (Digital Marketing Institute, n.d). Finally, social media’s agendasetting capabilities give substantial voice to the young voters since they are the largest user
group on most social media platforms. To some degree, this counteracts the heavy influence
that the older segment exerts due to being a coveted and reliable voter segment that possesses
greater resources.
89
4.6.3 Limitations and directions for future research
A limitation of this research was the use of a small student sample. Additionally, the qualitative
nature of the study limits its generalizability. However, the study does provide sound and
promising directions for future research. Firstly, the integration of a new dimension in the RMO
model requires a quantitative approach to develop and validate the construct. Secondly, since
young voters are often considered to be a distinct segment (Winchester et al., 2015),
identification of dimensions that are important to older voters is equally important. For
instance, young voters’ appreciation of a social and bipartisan approach might not be replicated
by the older voters. Young voters are interested in single-issue politics, have low political
knowledge and political involvement, are driven by image, and have limited responsibilities,
whereas older voters’ lifecycle stage (taxes, mortgage, children), higher civic orientation, and
vaster political knowledge might require a more issue-oriented, professional, and partisan
approach (Twenge et al., 2012; Quintelier, 2007). Further, a more in-depth look into factors
that facilitate a positive perception of smaller political brands may also benefit political
marketers. Future researchers may explore how major political brands can use social media to
form bonds with voters and nurture political trust. Lastly, the literature recommends a
cooperative and a dialogue-orientated approach on social media but the actual implementation
of this is riddled with difficulties. In this regard, research is needed to guide political parties
and politicians in the enactment of an interactive and dialogue-based approach.
90
CHAPTER 5: ONLINE RELATIONSHIP MARKETING
THROUGH CONTENT CREATION AND CURATION
Abid, A., Harrigan, P., & Roy, S. K. (2020). Online relationship marketing through content
creation and curation. Marketing Intelligence & Planning, 38(6), 699–712.
https://doi.org/10.1108/MIP-04-2019-0219
Chapter 5 is the fourth study of the research project. It is an online content analysis that
examined the effect of various content cues and characteristics on online relationship quality.
The paper appeared in Marketing Intelligence and Planning’s special issue: Online
Relationship Marketing.
How can political brands
strengthen their relationships
with voters on social media?
CH3. What is the nature
of social media enabled
voter relationships?
RQ1. What are the
gratifications that drive
voters to follow political
brands?
RQ2. What are the drivers
of social media-enabled
voter relationships?
RQ3. What are the
interactions that underpin
social media-enabled
voter relationships?
CH4. What is the
marketing orientation
that political brands
adopt towards social
media marketing?
CH5. What is the role of
marketer-generated
content in driving online
voter relationships?
RQ1. Do political
brands adopt a
relationship marketing
orientation towards
social media
marketing?
RQ2. Are there any
dimensions that are
not represented in
RMO framework?
RQ1. What is the effect
of various content cues
on online relationship
quality?
RQ2. Does content
curation have an
impact on online
relationship quality?
Figure 5.1 Chapter 5 (Paper 4)
91
CH6. What are the roles of
marketer-generated
content and behavioral
engagement in driving
online voter relationships?
RQ1. What are the
effects of various
content cues on
engagement and online
relationship quality?
RQ2. Does behavioral
engagement mediate the
effect of content cues
on online relationship
quality?
5.1 Abstract
Purpose – The purpose of this paper is to test the influence of various content cues and
characteristics on the followers’ online expressions of relationship quality. Second, the
research aims to understand the moderating effects of content curation.
Design/methodology/approach – The sample comprised of 100 posts and 29,000 comments
that were sourced from the Facebook pages of the Democratic Party and the Republican Party.
The content was coded using the prior literature. Comments were manually coded using a
deductive approach and captured the dimensions of relationship quality. Multiple regression
was used to confirm the hypotheses.
Findings – Visuals, content popularity, the volume of comments, and the content’s length have
a positive effect on voters’ expressions of relationship quality. However, source credibility,
argument quality, valence, and interactivity does not have an impact. Additionally, content
curation negatively moderates the effects of length and interactivity on expressions of
relationship quality.
Practical implications – The findings emphasize the use of peripheral cues rather than the
central route. Curating interactive and lengthy content should be avoided, however, curation of
images and videos is well received.
Originality/value – The research contributes to the literature by understanding the role of
marketer-generated content in building online relationships. Additionally, it explores the
distinct impacts of created and curated content.
92
5.2 Introduction
Despite a surge in interest, research on social media’s impact on customer relationships remains
limited (Steinhoff et al., 2019). Although studies discuss the effects of social media
engagement on online relationships (e.g. Clark et al., 2017; Achen, 2016), a granular approach
is needed to identify the content-based factors that strengthen online relationships. This study
focuses on marketer-generated content (MGC) and advances theory by investigating the effect
of content, not on social media’s vanity metrics like shares, comments, and likes, but on online
relationships. Specifically, we look at the effects on the followers’ expressions of relationship
trust, commitment, and satisfaction. A secondary purpose of this research is to explore whether
curated content is perceived differently from created content. Although their distinct impacts
have attracted the attention of practitioners, scholarly work is absent. This research adopts an
online content analysis that is underpinned in the Elaboration-Likelihood Model (ELM) (Petty
& Cacioppo, 1986). In total, 100 posts and 29,184 comments were analyzed to investigate the
inherent online relationships on the official Facebook pages of the Democratic Party and the
GOP (Republican Party). The study provides tactical insights and contributes to theory in
relationship, social media, and political marketing. Finally, in the backdrop of the upcoming
US presidential election, this study is relevant and timely.
Politicians spent $900m on digital marketing in the US mid-term elections of 2018 (Kantar
Media, 2018). Similarly, voters are active on social media also. Two-thirds of users engage in
political activities and a third follow politicians (Newman et al., 2017). The literature
recommends a relationship-based approach to political marketing because it considers the longterm welfare of the society, reduces voter distrust, and increases voter engagement (Ormrod et
al., 2013). Political relationship marketing (PRM) is the ideal approach on social media,
however, politicians fail to adopt it (Harris & Harrigan, 2015).
The purpose of this study is to analyze the various content cues and characteristics employed
by the Facebook pages of American political parties and test their impacts on the expressions
of relationship quality in the comments. Previous research on the ELM and social media has
highlighted argument quality, source credibility, source style, and content popularity as
relevant cues (Teng et al., 2017; Chang et al., 2015). Outside the ELM literature, content
interactivity is a frequently cited characteristic (De Vries et al., 2012). The selection of
relationship trust, commitment, and satisfaction was driven by their extant presence in the
93
literature and their wide acceptance as dimensions of relationship quality (Clark et al., 2017;
Sanchez-Franco & Rondan-Cataluña, 2010). The study contributes to the literature. It explores
the role content cues and characteristics play in influencing followers’ expressions of
relationship quality. Second, it studies relationship quality in an underexplored political context
using a relatively novel approach. Additionally, it furthers the integration of the ELM,
marketer-generated content, and relationship marketing into political marketing. Finally, it
studies the role of content curation in moderating the impact of content on online relationships.
5.3 Literature review
5.3.1 Created and curated content
Voters come across a vast amount of political content on social media. This content is
categorized as MGC and user-generated content. In regards to MGC, marketers can either
create original content or can select and share content from external sources. The latter is
known as curated content. The scant literature limits our understanding of the distinct impacts
of created and curated content, if there are any. The authors posit that curating content will
have a negative impact on online relationships. First, personalization is essential to relationship
marketing (Gordon et al., 1998). Social media channels that create content are better equipped
at offering personalization to their audience. Second, when viewed from the perspective of
Social Exchange Theory, relationship marketing’s earliest theoretical foundation, both parties
have reciprocal obligations. Official Facebook pages provide relevant content, whereas
followers engage with and respond to the content. This exchange happens through reciprocity.
Reciprocity is essential to relationships and social media facilitate reciprocity (Clark et al.,
2017). Since curating reflects a limited investment in content, reciprocity dictates that the
followers might limit their engagement. Additionally, directed and undirected communications
have different impacts. Directed communication is more likely to evoke a response (Goh et al.,
2013). Content created by social media channels themselves is more likely to be perceived as
directed communication. Therefore, the authors propose that content curation will have a
negative moderating effect on the impacts that content cues have on the expressions of
relationship quality.
94
5.3.2 Social media and relationship quality
Online customer relationships can be strengthened by offering relevant and valuable content
(Steinhoff et al., 2019). Social media impact various dimensions of a relationship and have the
potential to foster close relationships with customers (Verma et al., 2016). Social media
influence relationship quality also. For instance, fans who followed the official Facebook pages
of NBA teams enjoyed a higher relationship quality (Achen, 2016). Clark et al. (2017) found
the same to be true in the context of students and universities. Since social media engagement’s
impacts on relationships and relationship quality are acknowledged, an understanding of the
content-based factors that enhance relationship quality is needed.
5.3.3 Relationship quality
Relationship quality is a composite or a multidimensional construct that captures various
dimensions of a relationship and represents its strength (Palmatier et al., 2006). Relationship
quality leads to positive customer outcomes (Roy & Eshghi, 2013). There is no consensus as
to what dimensions constitute relationship quality. However, relationship trust, commitment,
and satisfaction are the most commonly used variables to represent relationship quality (Clark
et al., 2017). In this study, relationship quality represents the percentage of comments in which
the followers have expressed relationship trust, commitment, or satisfaction with the political
party or its leaders.
Relationship trust
Relationship trust is one’s “confidence in the exchange partner’s reliability and integrity”
(Morgan & Hunt, 1994, p. 23). Various operationalizations of trust exist in the literature that
consider the elements of reliability, truthfulness, and sincerity as important dimensions of trust
(Garbarino & Johnson, 1999; Morgan & Hunt, 1994). Relationship trust is pertinent to politics
considering the declining trust in political parties (Dalton, 2013).
Relationship commitment
Commitment is “an enduring desire to maintain a valued relationship” (Palmatier et al., 2006,
p. 138). Party membership and identification are declining and major parties are losing voters
to minor parties (PEW Research Centre, 2018). Consequently, voter commitment has become
essential to the long-term survival of major parties. This study utilizes Garbarino and Johnson’s
95
(1999) conceptualization of relationship commitment, which bases commitment on a sense of
belonging, pride, loyalty, and length of the relationship.
Relationship satisfaction
Relationship satisfaction is a consumer’s overall satisfaction with the relationship (Palmatier
et al., 2006). Verma et al. (2016) determined that relationship satisfaction was the most crucial
variable in cultivating online customer loyalty. Current times have seen decays in the
satisfaction levels of political institutions and democracy (American Institutional Confidence
Poll, 2018).
5.3.4 Political relationship marketing and social media
Although politicians’ content strategies have received academic attention, much of this
research is grounded in the political sciences’ literature (e.g. Muñoz & Towner, 2017).
However, studies provide evidence that content characteristics have an impact on online voter
behavior (Elder & Phillips, 2017). The literature recommends that politicians should adopt a
relational approach on social media, however, politicians have been unable to do so (Parsons
& Rowling, 2018). Social media are incompatible with the one-way communication associated
with politics and work better as relational tools (Harris & Harrigan, 2015). Minor parties have
successfully employed such an approach and have thrived due to the intimate networking
offered by social media (Maier & Tenscher, 2009).
5.3.5 Elaboration-Likelihood Model
The ELM is a dual-process theory of persuasion (Petty & Cacioppo, 1986). It is widely used to
understand the content on social media (Teng et al., 2017). ELM posits a central and peripheral
route. The central route implies that when receivers are motivated and able to process
information, the argument’s quality is effective in changing attitudes. Conversely, if the
receivers are unable or lack the motivation to process information, attitude change will occur
via the peripheral route. ELM has been utilized to understand relationship quality in the context
of interest-based online communities. Chen and Ku (2013) found the content’s argument
quality and source credibility to have a positive impact on community members’ perceived
relationship quality. Similarly, Sanchez-Franco and Rondan-Cataluña’s (2010) study suggests
that both peripheral cues (aesthetics) and central route (usability) influence online relationship
quality. The ELM is a model of persuasion which makes it pertinent to politics. The selection
96
of the cues was driven by the research methodology’s constraints, the social media context, the
extant presence of argument quality and source credibility in the literature, and the current
political climate of the USA.
5.3.6 Content cues and the conceptual framework
Argument quality
Argument quality refers to the persuasive power of the argument that is presented in the
message. A strong argument is comprehensive, accurate, timely, and relevant (Teng et al.,
2017). Argument quality increases online trust and has a positive impact on relationship quality
(Chen & Ku, 2013). Wang et al. (2017) found that argument quality and relationship quality
work in synergy to facilitate favorable online behavior. The current study postulates that:
H1a. Argument quality will have a positive impact on the expressions of relationship quality
H1b. Content curation will negatively moderate the effect of argument quality on the
expressions of relationship quality
Peripheral cues
The decrease in political involvement has magnified the significance of peripheral cues
(Landtsheer et al., 2008). This study explored source credibility, source style, content
popularity, and content interactivity.
Source credibility. Source credibility refers to the perceived believability, competence, and
trustworthiness of the source (Chen & Ku, 2013). Source credibility influences an online
message’s acceptability and confidence in it, and despite being a peripheral cue, it acts via the
central route also (Shu & Scott, 2014). Source credibility influences relationship quality (Chen
& Ku, 2013). Therefore, the authors propose that:
H2a. Source credibility of the content will have a positive impact on the expressions of
relationship quality
H2b. Content curation will negatively moderate the effect of source credibility on the
expressions of relationship quality
97
Source Style. Source style captures five dimensions of online messages: visual, volume, length,
valence, and dispersion (Teng et al., 2017). Dispersion has not been included in the research.
Although source style was created to study online customer reviews, the dimensions apply to
MGC also.
Visuals influence emotions and impact information acceptance, attitude change, and preference
(Lin et al., 2012). Visual content generates greater engagement on social media (Kim et al.,
2015). Similarly, a message’s valence influences the audience also (De Vries et al., 2012; Teng
et al., 2017). Although messages with negative valence are more impactful, considering the
context of relationship marketing, the authors theorize that content with a positive valence will
generate positive comments, and therefore, it will positively influence relationship quality. The
authors posit that:
H3a. Visuals (images and videos) will have a positive impact on the expressions of relationship
quality
H3b. Content curation will negatively moderate the effect of visuals on the expressions of
relationship quality
H4a. The valence of the content will have a positive impact on the expressions of relationship
quality
H4b. Content curation will negatively moderate the effect of valence on the expressions of
relationship quality
A greater volume of reviews reduces the risks associated with a product by increasing the
perception of trust and credibility (Teng et al., 2017). Moreover, a higher number of reviews
corresponds to higher sales (De Maeyer, 2012). It is plausible that the volume of comments
will have a similar effect for MGC. Content’s length influences its perception (Teng et al.,
2017). An advertisement’s length increases recall and attitude formation (Newell & Henderson,
1998), and 30-second online advertisements are more effective at conveying emotions than 15second advertisements (Goodrich et al., 2015). The study postulates that:
H5a. The length of the content will have a positive impact on the expressions of relationship
quality
98
H5b. Content curation will negatively moderate the effect of content’s length on the
expressions of relationship quality
H6a. The volume of the comments will have a positive impact on the expressions of
relationship quality
H6b. Content curation will negatively moderate the effect of the volume of comments on the
expressions of relationship quality
Content Popularity. Zadeh and Sharda (2014) conceptualize content or post popularity as all
the post metrics that are visible to users. Content popularity impacts users’ perceptions. Since
social media users are swayed by social influences and norms (Chang et al., 2015), it is
conceivable that the content’s popularity increases the likelihood of positive expressions of
relationship quality. Thus, the study hypothesizes that:
H7a. The popularity of the content will have a positive impact on the expressions of
relationship quality
H7b. Content curation will negatively moderate the effect of content popularity on the
expressions of relationship quality
Content Interactivity. Interactivity is “the degree to which two or more communication parties
can act on each other, on the communication medium, and on the messages and the degree to
which such influences are synchronized” (Liu & Shrum, 2002, p. 54). Interactivity’s impact on
online engagement is unclear. De Vries et al. (2012) found it to have a positive impact but other
findings are contradictory (Tafesse, 2015). Nevertheless, the interactivity offered by social
media is ideal to foster relationships (Verma et al., 2016). Therefore, the study hypothesizes
that:
H8a. The interactivity of the content will have a positive impact on the expressions of
relationship quality
H8b. Content curation will negatively moderate the effect of content interactivity on the
expressions of relationship quality
99
Content Cues and
Characteristics
Argument Quality
Content
Curation
Source Credibility
Visual
(-)
Valence
Expressions of
Relationship Quality
Relationship
Trust
Relationship
Commitment
(+)
Length
Volume of Comments
Relationship
Satisfaction
Content Popularity
Content Interactivity
Figure 5.2 Conceptual framework
5.4 Research design and methodology
To test the hypotheses, the study employed an online content analysis that adopted a mixedmethods approach. The coding of the content adopted a qualitative approach that was built
upon prior research. In contrast to traditional online content analyses that investigate likes,
shares, and comments as dependent variables, the study utilized the comments to investigate
expressions of relationship quality. The authors calculated the percentage of comments that
were expressions of relationship trust, commitment, or satisfaction for each post. This allowed
for quantitative analysis. Such approaches are essential if researchers want to utilize online
content analysis beyond likes, shares, and comments.
5.4.1 Sample and data collection
The content was collected from the official Facebook pages of the two parties. The analysis
covered 50 consecutive posts for each party preceding a fixed date. However, for the
Democratic Party, this was 53 posts. Three posts were in Spanish and were omitted from the
data. The study involved coding of comments also. To this end, Facebook posts and comments
were captured using Evernote and transferred to NVivo for coding and analysis. These
comments totaled 29,184. Out of these comments, a total of 8,624 comments were classified
100
as expressions of relationship quality. Replies to comments were not included in the analysis.
The authors acknowledge that 100 posts are a limitation of the research. However, studies
employing a similar sample size have yielded insightful findings through quantitative
approaches, such as Cvijikj and Michahelles’ (2011) analysis of 120 posts. Quantitative content
analyses of fewer than 100 posts are present in the political marketing literature (VesnicAlujevic & Van Bauwel, 2014).
5.4.2 Operationalization of variables
First, the content was classified for the media type (text, picture, and video). Subsequently, the
sample of 100 posts was coded using established coding instruments where available. For
instance, various levels of interactivity were coded based on past research (De Vries et al.,
2012; Tafesse, 2015). Content popularity was measured by adding visible statistics (likes,
shares, and comments) (Zadeh & Sharda, 2014). Coding of variables like length, the volume
of comments, visual, and valence was objective (Table 5.1). Subjectivity was required in the
coding of argument quality and source credibility. A senior coder worked alongside the primary
coder in the coding of argument quality. Lastly, source credibility was judged by the overall
visibility of the news organization. NY Times, CNN, Fox News, Washington Post, and official
sources like Whitehouse.gov were coded as credible. Content that was without any source or
was shared from lesser-known news websites was coded as having low credibility. These
included sources like VOX, Mother Jones, and The Daily Beast.
Table 5.1 Operationalization of variables
Construct
Argument Quality
Level
Operationalization
0=Argument quality not used
1=Argument quality used
1= No source/low credibility source
2= Credible source
0= No visual
1= Visual present
0= Short (video < 1 minute, text < 50 words)
1= Long
0= Negative valence
1= Positive/neutral valence
Facebook post’s number of comments
0= No interactivity
1= Low interactivity (link provided)
2= Call to action
3= Call to action and link provided
Sum of post's likes, shares, comments
0= Created Content
1= Curated content
Source Credibility
Visual
Length
Valence
Volume
Interactivity
No
Low
Medium
High
Post popularity
Content Curation
101
Comments were analyzed to identify, code, and count comments that characterized relationship
quality. The comments were coded based on a theoretically driven, deductive approach. The
deductive codes were derived from items that are used in the literature to represent the three
components of relationship quality. Table 5.2 provides details about the coding process.
Comments have been altered. Each comment was categorized in one code only. Some
comments expressed conflicting views regarding party and leader. The comments that
preferred the leader but not the party were excluded, whereas comments favoring the party but
not the leader were included in the analysis. To ensure validity, a second coder, who was not
involved in the study, was recruited, trained, and allocated ten posts. The primary coder was
not involved in this process. The inter-coder reliability for relationship trust, commitment, and
satisfaction were 0.9, 0.7, and 0.7 respectively, with an average of 0.77. Although the ICR
value is slightly less than desirable, 0.7 is often cited as an acceptable value (MacPhail et al.,
2016). The Cohen’s κ for the three variables were 0.62, 0.65, and 0.66. This indicates a
substantial amount of inter-coder agreement beyond chance (Hallgren, 2012). Considering the
high number of comments, the complexity of coding comments expressing multiple thoughts,
and the inherent subjectivity of interpreting online communications, the authors deemed these
values acceptable. Using the number of expressions of relationship quality variables in the
comments and the total number of comments, the authors calculated the percentage of
comments expressing the variables of relationship quality for each post. Subsequently, the data
were transferred to SPSS
102
Table 5.2 Coding items and examples for relationship quality
Construct
Items
Examples
Commitment
Long term
I have always supported the Republican Party and will continue to do so.
Loyal
Sense of belonging
You have my vote #TRUMP2020.
We Republicans need to put up a united front on the shutdown.
Proud
Forget the haters. The whole country is proud of you Mr Trump.
Reliability
Democrats should know that this President means what he says.
Truthfulness
Is it the republican party that only cares about the Americans?
Sincerity
Donald Trump loves his country and Americans. I know this much.
Trust
He says the truth. Too bad the fake media is undermining him.
Satisfaction
Overall Satisfaction
I agree with this decision a hundred percent.
Comparative Satisfaction
Economy is great under this administration. Lowest unemployment!
Rather than investigating his campaign, they should investigate the Democrats.
Good job Mr President.
Noise
Comments that expressed dissatisfaction and distrust towards the content generator
Comments with friends tagged in the comment
Comments with emoticons only
Comments that showed satisfaction/dissatisfaction with both parties
Multiple comments by the same user
Neutral comments
Comments where users said they were not Americans
Comments that provided a link without any conclusive statement
Comments that did not make sense or were ambiguous
103
5.5 A tale of two parties: a preliminary analysis
GOP’s content fared better at generating social media responses. Similarly, the user comments
included a higher number of expressions of relationship quality (almost five percent). In total,
78 percent of the Democratic content was devoted to issues. A clear majority of this content
was informative, of negative valence, and curated from news websites. In contrast, the GOP’s
approach was dominated by values, emotions, and the use of symbolic images. GOP’s content
received 58 percent more likes/post and 141 percent more shares/post when adjusted for the
number of followers. Democrats generated 34 percent more comments/post, however, these
comments included fewer expressions of relationship quality when compared to the GOP.
GOP’s relative success may be attributed to a relationship-based approach. The content mix
used by the GOP was a balanced mix of videos, images, and text. Further, the significantly
higher content creation rate of the GOP implies a more customized undertaking. Higher contact
frequency, as was the case for the Democratic Party (6 posts/day v/s GOP’s 1.3 posts/day),
leads to stronger relationships (Palmatier et al., 2006). However, social media experts
recommend a range of 1–3 posts/day for accounts with a large following. Beyond this, the
content’s ability to elicit responses decreases due to Facebook algorithms (Kolowich, 2017).
Online PRM should include calls to action (Lees-Marshment, 2014). GOP’s consistent use of
these calls exceeded that of the Democrats. Finally, the low instances of negative valence
(attack/comparative content) in GOP’s content characterized a relational strategy also.
5.6 Results
5.6.1 Descriptive statistics
The central route was adopted in 24 posts and credible sources were used in 44 posts. A clear
majority of the posts employed visuals (n = 92). Content with positive valence was
incorporated in 31 posts. Interactivity was present in 67 posts. The average popularity of a post
was 2,385.37 (SD = 2,640.75) and the average number of comments was 291.87 (SD = 165.08).
The average number of shares was 482.08 (SD = 1,151.62) and the average number of likes
was 1,611.42 (SD = 1,748.99). Finally, the average percentage of comments that were coded
as expressions of relationship trust, commitment, and satisfaction (i.e. relationship quality
percentage) was 25.4 percent (SD = 12.76).
104
5.6.2 Test of hypotheses
Prior research on the impact of content cues and characteristics employs regression (De Vries
et al., 2012) or an analysis of variance (Cvijikj & Michahelles, 2011). The former was selected
for testing the hypotheses. Multiple regression analysis was carried out to investigate the
impact of the independent variables on the expressions of relationship quality. Given that the
independent variables are nonmetric in nature with two or more categories, dummy variables
were used, which acted as the replacement independent variables (Hair et al., 2010). Each of
the dummy variables represented one category of a nonmetric independent variable. Any
nonmetric independent variable with k-categories was represented by k−1 dummy variables.
A preliminary analysis was conducted to check for basic assumptions of multiple regression
(Pallant, 2016). There were no correlations above 0.5. The scatterplot of the standardized
residual revealed a few outliers that were not deleted considering the small sample size. The
variance inflation factor (VIF) values of all the independent variables used in the regression
analysis are less than 5 which indicates that multicollinearity is not an issue with this research
(Hair et al., 2010). The VIF values for the independent variables are: argument quality (VIF =
1.21), source credibility (VIF = 1.57), valence (VIF = 1.61), content popularity (VIF = 1.40),
content interactivity (VIF = 2.18), volume of comments (VIF = 2.34), visual (VIF = 3.20), and
length (VIF = 1.31). Interaction terms were created between the moderating variable (i.e.
content curation) and the independent variables, and the interaction terms were included in the
hierarchical regression analysis along with the respective independent variables. Considering
the small sample size, the adjusted R2 is a better measure of the model’s explanatory power
(Tabachnick & Fidell, 2007). The adjusted R2 is 62.6 percent. A dummy variable was included
in the regression equation to check for differences across the followers of both parties. The
results showed no significant impact of the dummy variable. The results are presented in Table
5.3.
105
Table 5.3 Regression Table and ANOVA Results
Independent
Variables
Standardized
Regression
Weights
t-value
p-value
Hypotheses/
Results
Direct effects
Argument Quality
-.09
-.76
.45ns
H1a/Rejected
Source Credibility
.06
.61
.54ns
H2a/Rejected
Visual
.30
2.15
.04**
H3a/Accepted
Valence
-.03
-.29
.77ns
H4a/Rejected
Length
.47
3.41
.001**
H5a/Accepted
Volume of Comments
.48
4.64
.000***
H6a/Accepted
Content Popularity
.67
2.08
.04**
H7a/Accepted
Content Interactivity
.60
.49
.62ns
H8a/Rejected
Moderation results (only significant moderation results shown)
Length * Content
-.58
-3.85
.00***
H5b/Accepted
Curation
Content Interactivity *
-.28
-2.27
.00***
H8b/Accepted
Content Curation
Note: Dependent variable: Expressions of Relationship Quality
R-square=.63; F-value=7.18; Sig. F=.000
ns → non-significant; *** indicates p-value <.001; ** indicates p-value <.05
Length * Content Curation (Interaction term)
Content Interactivity * Content Curation (Interaction term)
ANOVA Results
Model
Sum of Squares
df
Mean Square F
Sig.
1
Regression
9048.89
15
603.26
7.18
.000b
Residual
7062.89
84
84.08
Total
16111.78
99
Dependent Variable: Expressions of Relationship Quality
Predictors: (Constant): Argument Quality, Source Credibility, Visual, Valence, Length, Volume of Comments,
Content Popularity, Content Interactivity
In regards to direct effects, H1a and H2a stated that argument quality and source credibility
would have a positive influence on expressions of relationship quality. However, the effects of
argument quality (β = −0.09, p = 0.45) and source credibility (β = −0.06, p = 0.54) were not
statistically significant. Therefore, H1a and H2a were not supported. The results supported
H3a, H5a, and H6a which theorized the direct effects of visual (β = 0.30, t = 2.15, p<0.05),
length (β = 0.47, t = 3.41, p<0.01), and volume of comments (β = 0.48, t = 4.64, p<0.01) on
the expressions of relationship quality. However, valence did not show any positive influence.
Therefore, H4a was not supported. H7a was accepted as content popularity (β = 0.67, t = 2.08,
p<0.04) was the strongest unique predictor of the expressions of relationship quality. H8a was
rejected as content interactivity did not have any effects. The study also investigated the
moderating effects of content curation. H5b (β = −0.58, t = −3.85, p<0.01) and H8b (β = −0.28,
t = −2.27, p<0.01) were the only hypotheses that were accepted as the results showed that
content curation negatively moderated the influence of content’s length and interactivity on the
expressions of relationship quality. However, it is worth mentioning that H3b (β = 0.37, t =
2.03, p<0.05) showed a significant relationship when visuals’ influence on the expressions of
106
relationship quality was moderated for curation. Since this was not in the direction predicted
by the hypothesis, H3b was rejected.
5.7 Discussion and implications
5.7.1 Discussion
MGC cues and characteristics impact online engagement (Chang et al., 2015). Subsequently,
online engagement impacts online relationship quality (Clark et al., 2017). However, studies
exploring the direct influence of content cues and characteristics on online relationships are yet
to come to fruition. This study sought to fill this gap. Moreover, prior online content analyses
neglect the vast data embedded in the comments. Surely, research can benefit from a
theoretically driven analysis of comments. Finally, the study investigated content curation and
creation in the light of relationship marketing literature.
Content’s argument quality and source credibility did not have an impact. This contradicts the
past literature (Chen & Ku, 2013). The deviation may be attributed to the low political
involvement of voters, stronger attitudes in politics, political polarization, and the emotional
aspect of political behavior. These conditions increase the reliance on peripheral cues rather
than argument quality (Howe & Krosnick, 2017; Petty & Cacioppo, 1986). Additionally, prior
studies were conducted using online interest-based communities (fashion, tourism, etc.) where
members exhibit higher levels of involvement. Following someone on social media is a
comparatively passive and low-involvement activity. Source credibility, despite being a
peripheral cue, influences via the central route also (Shu & Scott, 2014). This strengthens our
conclusion that the central route is less influential since both source credibility and argument
quality had no significant effects. It is a logical deduction that a popular post will attract
positive comments. The impacts of popularity and volume of comments may be attributed to
social influences and bandwagon effects (Chang et al., 2015). The findings demonstrate that a
higher number of comments correspond to a favorable discourse. Additionally, the study shows
that visual and lengthy content is effective. Visuals are the preferred media type on social media
(Kim et al., 2015) and are employed to evoke emotions using political symbolism (Ormrod et
al., 2013). Similarly, longer advertisements increase recall and evoke emotions (Goodrich et
al., 2015). The rejection of the central route and the effectiveness of peripheral cues indicate a
low political involvement.
107
The results indicate that curation might not be ideal in some cases; for instance, when the
content is lengthy or interactive. Followers prefer viewing lengthy content and interacting with
content when it is created. Interactivity’s impact on social media responses is unclear. De Vries
et al. (2012) found that it increases audience responses because it prompts followers to respond.
On the contrary, Tafesse (2015) argues that interactivity increases the post’s structural
complexity, resulting in greater exertion of cognitive resources and low response rates. The
latter appears valid in the context of our research since parties shared content that requested
followers to take actions requiring considerable effort (signing an online petition). Perhaps, the
type and context of interactivity matter. Asking followers to sign a petition is different from a
branded social media contest. Contrary to our hypothesis, curation positively moderated the
impact of visuals on the expressions of relationship quality. Our findings provide sufficient
evidence that content curation and creation influence the content’s perception. Interestingly,
practitioners realize this effect. Hootsuite (2019) recommends that 40 percent of content should
be created because it drives conversion rate and web traffic. Finally, the findings demonstrate
that content factors play a role beyond persuasion, engagement, and social media responses.
5.7.2 Theoretical implications
The study’s first contribution concerns the introduction of a content-centered approach towards
the study of online relationship marketing. Social media revolves around content and there is
a need to fully understand how various content cues and characteristics lead to stronger online
relationships. Such a content-centric approach is common in studies exploring social media’s
impact on engagement and responses but is missing in studies investigating online
relationships. Our findings indicate that certain content cues are more desirable and better
suited to building online relationships.
Second, the distinction between the two types of MGC (i.e. created and curated) has not been
explored in the social media literature, despite numerous studies on MGC’s impact on social
media responses. Our study highlights that content curation has a negative impact in certain
scenarios. The authors recommend its inclusion in future studies since not taking the MGC’s
type into account will lead to imperfect results.
Third, the study contributes to the literature in the field of PRM, adding a quantitative
dimension to a mostly qualitative domain (e.g. Parsons & Rowling, 2018; Harris & Harrigan,
108
2015). The findings show that a relational approach on social media is reciprocated by the
followers. The comparative analysis demonstrates that the GOP’s relational approach led to
higher social media responses and expressions of relationship quality.
Furthermore, the research contributes to political social media marketing by identifying the
various content-based factors that drive positive expressions in comments. We highlight that
content’s interactivity might have a contextual rather than an absolute impact, which depends
on the type of interactivity offered and the followers’ level of involvement. This explains some
earlier findings that contradict each other. We highlight the importance of peripheral cues in
engaging voters. This is aligned with the assessment that modern voters have low political
involvement and are emotionally driven (Bruter & Harrison, 2017).
Finally, social media comments have a vast amount of data embedded in them that is rarely
explored in quantitative studies. Our research refines an alternative method to utilize online
content analysis, beyond the number of social media responses. Although rare, a somewhat
similar approach has been adopted in the study of social media communications (Rowe, 2015).
5.7.3 Managerial implications
The study recommends the use of peripheral cues to increase the expressions of relationship
quality in the comments. This will create a positive e-word of mouth also. Visuals should be a
central element in the content strategy. Also, political parties that curate content excessively
might be at a disadvantage. The industry standard recommends a content creation rate of 40
percent (Hootsuite, 2019). If curating, the study recommends sharing images and short videos.
Parties are recommended to use interactivity in a manner that is less resource-intensive and
more engaging. For instance, branded contests with rewards might be more effective (De Vries
et al., 2012). Importantly, argument quality might not be effective in strengthening online
political relationships and parties should not base their content strategy solely on arguments.
Additionally, the study highlighted the different approaches taken by the parties. This is
insightful from the perspective of an overall content strategy. The difference between the
content cues and characteristics and audience responses is meaningful given the GOP’s strategy
was more relational and yielded greater engagement and expressions of relationship quality.
The Democrats’ lack of a relational approach demonstrates that even major political parties are
yet to utilize social media to their full extent. Lastly, in terms of societal implications, the
109
research reveals that, in an era of low political involvement, high political hostility,
partisanship, and fake news, appealing through argument quality, source credibility, and
content with positive valence has little impact on voter relationships. These are ominous signs
for democracy’s future.
5.7.4 Limitations and future research
The use of two parties from one country and a relatively small sample size are limitations of
this study. However, an increase in the number of posts would have required computer-aided
coding which is not ideal for coding semantically complex texts (Krippendorff, 1989). The ICR
value raises some concerns. Considering the coding complexities discussed earlier, the authors
consider this satisfactory. When using online content analysis, it is not possible to know which
posts are boosted by paying for increased visibility. However, this is an issue that is inherent
in the methodology. Another limitation of our research is not giving equal weight to the three
relationship quality variables. Thus, the results are skewed towards the expressions of
relationship satisfaction since they were the most common occurrence. Also, the political
context limits the study’s generalizability. Further research is needed to determine the impacts
of curated and created content. The authors call for more research that investigates the role of
content in developing online relationships. Future researchers can use a similar methodology
to explore other content cues, characteristics, and appeals.
110
CHAPTER 6: SOCIAL MEDIA IN POLITICS: HOW TO
DRIVE ENGAGEMENT AND STRENGTHEN
RELATIONSHIPS
Abid, A., Harrigan, P., Wang, S., Roy, S. K., & Harper, T. (2020). Journal of Marketing
Management. First round of revision is due in September.
Chapter 6 is the final study of the research project. It examined the effect of various content
cues and characteristics on engagement and online relationship quality. Paper 5 extends Paper
4 to devise a more holistic framework of social media-enabled voter relationships. The editor
and the reviewers have requested revisions in the manuscript.
How can political brands
strengthen their relationships
with voters on social media?
CH3. What is the nature
of social media enabled
voter relationships?
RQ1. What are the
gratifications that drive
voters to follow political
brands?
RQ2. What are the drivers
of social media-enabled
voter relationships?
RQ3. What are the
interactions that underpin
social media-enabled
voter relationships?
CH4. What is the
marketing orientation
that political brands
adopt towards social
media marketing?
CH5. What is the role of
marketer-generated
content in driving online
voter relationships?
RQ1. Do political
brands adopt a
relationship marketing
orientation towards
social media
marketing?
RQ2. Are there any
dimensions that are
not represented in
RMO framework?
RQ1. What is the effect
of various content cues
on online relationship
quality?
RQ2. Does content
curation have an
impact on online
relationship quality?
Figure 6.1 Chapter 6 (Paper 5)
111
CH6. What are the roles of
marketer-generated
content and behavioral
engagement in driving
online voter relationships?
RQ1. What are the
effects of various
content cues on
engagement and online
relationship quality?
RQ2. Does behavioral
engagement mediate the
effect of content cues
on online relationship
quality?
6.1 Abstract
Marketing researchers have devoted considerable attention to marketer-generated content,
social media behavioral engagement (likes and shares), and online relationships. Prior studies,
however, do not integrate these critical aspects of social media marketing. Our study, which is
underpinned in the Elaboration-Likelihood Model, offers evidence that MGC leads to high and
low-involvement SMBE, which in turn has a positive impact on followers’ online relationship
quality. A content analysis of the official Facebook pages of the Democratic and Republican
parties revealed that peripheral cues (source credibility, emotion, valence, and visual
symbolism) are the primary drivers of SMBE, and argument quality impacts high-involvement
SMBE only. We recommend that political marketers should rely on distinct sets of MGC cues
to target high and low-involvement followers, with emotion being the universal element.
112
6.2 Introduction
Proficient use of social media has played a pivotal role in the success of many political brands
across the world. They played an important role in the United States (US) presidential election
of 2020. Financial resources dedicated to digital political marketing have grown exponentially
over the last decade. Around $1.8 billion was spent on digital media in the recent presidential
election, and Facebook and Google were the primary financial benefactors (Radio Info, 2020).
Social media’s influence on politics will continue to grow because social media, unlike
traditional media, offer utility beyond one-way campaigning. Official Facebook pages of
political brands engage in seeking donors, updating followers, facilitating petitions, recruiting
and engaging volunteers, and developing online relationships with followers.
Social media facilitate relationship marketing (Steinhoff et al., 2019; Sheth, 2017). Political
brands are continuously engaging their followers on social media in an attempt to foster online
relationships with them. The gap in knowledge in this area is focused on the mechanism that
leads to stronger relationships on social media (Steinhoff et al., 2019). In particular, it is
important to understand how marketer-generated content (MGC) influences social media
behavioral engagement (SMBE), and how this SMBE leads to stronger online relationships.
Previous research shows that MGC cues and characteristics drive behavioral engagement on
social media (e.g. Dolan et al., 2019; Colicev et al., 2018; Tafesse & Wien, 2018), which in
turn leads to stronger online relationships (Clark et al., 2017; Achen, 2016). Our study offers a
holistic framework that integrates and tests the two aforementioned conclusions. Furthermore,
we classify SMBE as high-involvement (shares) and low-involvement (likes), offering a subtler
understanding of the factors underpinning SMBE. Such an understanding of SMBE is of
significance to political and commercial brands alike.
Specifically, we investigate the impact of argument quality and peripheral cues like source
credibility, emotion, negative valence, and visual symbolism on low-involvement SMBE and
high-involvement SMBE. Notably, we examine whether low and high-involvement SMBE are
indicative of low and high-involvement followers who will rely on peripheral and central routes
respectively. Further, the impact of high and low-involvement SMBE on online relationship
quality is also explored. Theoretically, we draw on the Elaboration-Likelihood Model (ELM),
which is a dual-process theory of attitude change (Petty & Cacioppo, 1986). The ELM has
emerged as a viable framework to understand the effectiveness of MGC in generating social
113
media responses and building relationships (e.g. Chang et al., 2020; Abid et al., 2019; Teng et
al., 2017). As a model of persuasion, the ELM is an appropriate framework to study political
marketing. Additionally, the ELM resembles Aristotle’s rhetorical appeals, the oldest
framework of political persuasion, which makes it pertinent to the political context. Finally,
the vast literature devoted to the ELM accommodates the various persuasive cues that are
utilized by political brands.
To test the proposed relationships, data are collected through a quantitative and qualitative
online content analysis of the official Facebook pages of the Democratic and Republican
parties, the two primary political parties of the USA. Specifically, data comprise 200 posts and
43,500 associated comments. Data were manually coded for ELM cues. SMBE was
operationalized using the number of likes and shares elicited by MGC (Tafesse & Wien, 2018),
and relationship quality, a composite of relationship trust, commitment, and satisfaction, was
operationalized through user comments, which were coded manually (Abid et al., 2019).
This research contributes to both theory and practice. From a theoretical perspective, it offers
a framework that provides a deeper understanding of how MGC leads to engagement and
stronger online relationships. Using recent literature, the study classifies SMBE as high and
low-involvement and uses ELM cues to verify this classification. Such a classification has
significance and implications for future researchers utilizing online content analysis of social
media pages and dual-process theories. Further, the study investigates the efficacy of various
ELM cues in the online political context. We offer insight to political marketers regarding the
tactical use of MGC to engage and foster relationships with social media followers based on
their level of involvement.
In the following section, we introduce the various concepts that underpin this study. These are
the Elaboration-Likelihood Model (ELM), social media behavioral engagement (SMBE), and
relationship quality. Subsequently, the various ELM cues and the conceptual framework are
described. This is followed by the methods section, which elaborates on the coding process
used in the study. Next, the results of the hypotheses testing are presented. The discussion and
implications are presented in the penultimate section. The final section highlights the
limitations of our study and suggests future avenues of research.
114
6.3 Literature review
6.3.1 Elaboration-Likelihood Model
A dual-process theory, the ELM is commonly utilized to understand the impact of social media
content (e.g. Chang et al., 2020; Colicev et al., 2018; Teng et al., 2017). According to the ELM,
persuasion or attitude change can take place via two routes, which are the central and peripheral
routes (Petty & Cacioppo, 1986). Receivers who are highly involved, i.e., they have the
motivation and ability to process a message, engage in a high degree of elaboration.
Consequently, attitude change occurs via the central route. When receivers activate the central
route, the argument quality of the message determines its persuasiveness. The attitudes formed
via the central route are stronger and lasting. The peripheral route is associated with lowinvolvement conditions in which receivers lack the motivation or ability to elaborate on the
message and devote substantial cognitive resources. As a result, peripheral cues, like those
pertaining to the source (e.g. credibility, attractiveness, likeability) or the message (e.g. length,
message’s popularity), determine the persuasive impact of the message. Attitude change via
peripheral route is weaker. Numerous peripheral cues have been identified in studies that
explore online content using the ELM (Teng et al., 2017).
As a model of persuasion, the ELM is an appropriate framework to understand political
marketing. However, studies benefiting from the ELM remain few in the political marketing
literature (e.g. Iyer et al., 2017; Koc & Ilgun, 2010; Landtsheer et al., 2008). These studies
mainly highlight the effectiveness of peripheral cues in political marketing. We believe that the
ELM’s similarity to Aristotle’s rhetorical appeals of ethos, pathos, and logos make it pertinent
to the political context. Per Aristotle, these three modes of persuasion represent an appeal to
the source’s character or credibility (ethos), an appeal to the audience’s emotions (pathos), and
an appeal to reason (logos). A similitude can be drawn between logos and argument quality.
Similarly, ethos and source-based peripheral cues are also comparable. Although emotions, or
pathos, rarely feature in the ELM literature since it is a model based on cognitive elaboration,
emotions have a place in the ELM since emotions influence cognition (Petty & Briñol, 2015).
Morris et al. (2005) stated that cognition has an emotional core. The ELM has also been used
to cognize online relationships (Chen & Ku, 2013; Sanchez-Franco & Rondan-Cataluña, 2010;
Jo, 2005), which makes it pertinent to the present study. The studies on the topic show that
both argument quality and source credibility have an impact on online relationship quality
(Chen & Ku, 2013; Sanchez-Franco & Rondan-Cataluña, 2010).
115
6.3.2 Social media behavioral engagement
Likes, shares, and comments represent the real behavior of followers. Therefore, they represent
the behavioral dimension of customer engagement (Tafesse & Wien, 2018), the other two being
emotional and cognitive (Hollebeek et al., 2014). Consequently, these social media responses
are termed consumer behavioral engagement on social media (Tafesse & Wien, 2018). Other
conceptualizations like post popularity (Chang et al., 2015), social media engagement behavior
(Dolan et al., 2019), and content receptivity (Kumar et al., 2016) also exist in the literature.
Research validates the positive impact of SMBE on various marketing outcomes. For instance,
content receptivity has a substantial effect on spending, cross-buying, and customer
profitability (Kumar et al., 2016), whereas post popularity influences the perceived usefulness
of content as well as preference towards it (Chang et al., 2015). Since mere comment count
does not represent positive consumer engagement, we operationalize SMBE as likes and shares
(Tafesse & Wien, 2018; Tafesse, 2015).
High-involvement SMBE and low-involvement SMBE
The processing of information is driven by an individual’s motivation, opportunity, and ability
(MacInnis & Jaworski, 1989), which are the primary antecedents of involvement (Leung &
Bai, 2013; Andrews et al., 1990). Of particular concern to our study are motivation and ability
since the two are inherent to the ELM, which rests on the premise that motivation and ability
are the main determinants of a higher likelihood of elaboration (Petty & Cacioppo, 1986). The
authors assert that followers who share MGC display a higher level of involvement than
followers who simply like MGC. According to the ELM, such followers have greater
motivation, resulting from personal relevance of the message, their need for cognition, or their
sense of responsibility, and greater ability, which may be due to a better understanding of the
message, their prior knowledge, or low levels of distraction.
Recent literature reveals that not all social media responses are the same. Writing a comment,
for instance, requires greater physical effort and cognitive processing than clicking the like
button, and therefore, it indicates a higher level of involvement or engagement with the content
(Alhabash et al., 2019; Dolan et al., 2019). The assertion that ‘liking’ represents a lower level
of involvement or engagement with the content is grounded in several recent studies. Firstly,
likes are one-click activities that require a low level of effort (Zell & Moeller, 2017). Likes
require a low amount of elaboration or reflection compared to other social media responses
116
since they are ritualistic, instant, habitual, and automatic responses to content that grab our
attention (Alhabash et al., 2109; Hayes et al., 2016). One study affirming this is Alhabash et
al. (2019), which was based on experiments that measured participants’ psychophysiological
measures (heart rate, skin conductance level, orbicularis oculi, and corrugator supercilii muscle
activation) as they browsed through content.
Qualitative studies show that social media users occasionally click the like button aimlessly,
without much elaboration or cognition (Hayes et al., 2016). This is associated with
automaticity, which may be due to a lack of attention or intention, and is linked to repetitive
behavior, such as frequently liking a post from a friend without much consideration for the
content (Hayes et al., 2016). Content generators can perceive the automaticity of those liking
their content and decrease their relationship closeness accordingly (Carr et al., 2016). Likes
can also be ‘pity likes’ and ‘support likes’, which are likes given when feeling sad for or being
supportive of someone, for instance, if their content is not getting likes (Rhoads et al., 2016;
Hayes et al., 2016). Both these represent a low devotion of cognitive resources towards the
actual content. Likes are the most common social media response because humans have limited
capacities of information processing which refrains them from deeply engaging with a high
number of messages on social media (Alhabash et al., 2019; Khan, 2017). In fact, liking and
passive consumption of content (simply viewing) are driven by the same user gratification,
which is entertainment, whereas commenting and sharing are closely linked to social and
informational gratifications (Khan, 2017). The authors recognize that likes are a more active
and desirable form of SMBE than passive viewing or consumption of content, as highlighted
in the literature (Dolan et al., 2019; Alhabash et al., 2109), however, they indicate a lower level
of elaboration and involvement with the content when compared with shares.
Consumer-to-consumer sharing of brand messages is an important aspect of social media
marketing (Ordenes et al., 2019). Sharing political content on social media is driven by selfexpression (Parmelee & Roman, 2019). Not only does the sharer take some ownership of the
content, sharing content with one’s network also reflects a greater involvement in the content’s
topic. People who share are more likely to perceive themselves as opinion leaders, having
social, self-serving, or altruistic motivations (Kümpel et al., 2015). This is particularly true for
sharing political content, which is a serious activity that is negatively linked to impulsiveness
(Hossain et al., 2018). Sharing political content is a form of self-presentation, and the content
acts as symbols through which the sharer represents him or herself (Liu et al., 2017). Sharers
117
of news and political content are interested in influencing and informing others and provoking
discussion (Chadwick & Vaccari, 2019). Research shows that individuals who are political
ideologues and have an interest in hard news are more likely to share, which demonstrates that
shares are the markers of engaged followers (Kalogeropoulos et al., 2017). Sharers propagate
a political brand’s message to their network and the content that is shared is usually
accompanied by an opinion in the form of a statement or caption (Liu et al., 2017), which
demonstrates that sharers engage in some form of content creation, the highest level of
engagement (Dolan et al., 2019). Like likes, shares are a one-click activity, but the followers
who share a content exert or expect to exert resources in responding to those who engage with
the shared content.
Based on the preceding discussion, we contend that shares represent a higher level of
involvement with MGC than likes, particularly in the political context. Therefore, we classify
shares as high-involvement SMBE and likes as low-involvement SMBE. It stands to reason
that followers engaging in high-involvement SMBE are more likely to process information via
the central route as opposed to those who engage in low-involvement SMBE. The latter group
is more likely to activate the peripheral route and process peripheral cues rather than relying
on the quality of the argument presented in the MGC.
6.3.3 Relationship quality
The integration of social media in relationship marketing (Sheth, 2017; Choudhury & Harrigan,
2014) and political marketing is well documented (Williams, 2017). A relational approach to
political marketing is advised in the literature (Ormrod et al., 2013; Henneberg &
O’Shaughnessy, 2009). A relational approach is perhaps the only feasible approach towards
social media, although, political brands remain reluctant to adopt it (Parsons & Rowling, 2018;
Harris & Harrigan, 2015). Voters have been known to develop online parasocial relationships
with politicians (Ancu & Cozma, 2009). Social media allow political brands to foster these
relationships by offering a direct and continuous channel of communication.
Relationship quality is a widely studied concept in relationship marketing. It refers to the
strength of a marketing relationship (Palmatier et al., 2006). Several dimensions have been
used to operationalize relationship quality. The most commonly utilized dimensions are that of
relationship trust, commitment, and satisfaction (Clark et al., 2017; Achen, 2016; Hajli, 2014).
118
Research shows that engagement with a brand’s social media channels leads to higher
relationship quality (Clark et al., 2017; Achen, 2016), which is linked to desirable marketing
outcomes (Hajli, 2014; Roy & Eshghi, 2013). At the individual level, relationship rust,
commitment, and satisfaction are the most studied variables in online and offline relationship
marketing (Verma et al., 2016; Palmatier et al., 2006).
Relationship trust refers to one’s “confidence in the exchange partner’s reliability and
integrity” (Morgan & Hunt, 1994, p. 23). Relationship commitment is “an enduring desire to
maintain a valued relationship” (Palmatier et al., 2006, p. 138). Relationship satisfaction is a
consumer’s overall satisfaction with their relationship (Palmatier et al., 2006). Understanding
relationship quality is crucial in the political context because trust in American political brands,
satisfaction with American democracy, and commitment to the two major parties have seen
significant declines (Pew Research Centre, 2018; American Institutional Confidence Poll,
2018). In line with recent research (Abid et al., 2019), the study operationalized online
relationship quality using comments associated with the content. The present study refers to
online relationship quality as the percentage of comments that were classified as expressions
of relationship trust, commitment, or satisfaction with the content-generator.
6.4 Conceptual framework
Argument Quality
Source Credibility
High-involvement
SMBE
(Shares)
Online Relationship
Quality
Emotion
Valence
Low-involvement
SMBE
(Likes)
Central route
Peripheral Route
Visual Symbolism
Figure 6.2 Conceptual framework
6.4.1 Argument quality
Argument quality is defined as “the persuasive strength of arguments embedded in an
informational message” (Bhattacharjee & Sanford, 2006 p. 811). In other words, it is the
119
receiver’s perception of how convincing the argument is or how complete and accurate is the
information presented in the argument (Chang et al., 2020). A strong argument is
comprehensive, accurate, timely, and relevant (Teng et al., 2017). Based on the ELM, strong
arguments have an impact on the attitudes of receivers who are highly involved or engaged
with the message and willing to elaborate on it (Petty & Cacioppo, 1986).
Strong arguments are effective in the political discourse also (de Zuniga et al., 2018).
Embedded in the rhetorical appeals framework, content analyses of Barack Obama’s Facebook
pages have found that logos has an impact on comments and shares (Gerodimos & Justinussen,
2014; Bronstein, 2013), which indicates that logos is effective in generating high-involvement
SMBE. In line with these studies and the central tenet of ELM, the authors conceive that
argument quality will have an impact via the central route on high-involvement SMBE.
Additionally, the ELM posits that argument quality impacts through the central route only.
Therefore, we hypothesize that:
H1. Argument quality will have a positive impact on high-involvement SMBE.
6.4.2 Source credibility
In situations of low involvement, when motivation and ability are missing, source-based factors
like source credibility become salient. Source credibility is a widely studied peripheral cue
(Teng et al., 2017; Hur et al., 2017; Chang et al., 2015). It refers to the perceived believability,
competence, and trustworthiness of the source (Chen & Ku, 2013). Source credibility
influences an online message’s acceptability and confidence in it, as well as its perceived
usefulness (Kang & Namkung, 2019; Shu & Scott, 2014). The present study, however, does
not refer to the source credibility of the communicator (i.e., Republican or Democratic parties’
Facebook pages) but that of the content they post. Sharing content from news websites is a
common behavior of political brands on social media. It is the credibility of the sources of this
news that we investigate. To objectively study the effect of a content’s source credibility, the
authors judged the content’s source credibility by utilizing the Media Bias Chart, a mechanism
developed by Ad Fontes Media that aims to help news readers establish the reliability of news
sources.
120
Content shared from credible sources like Reuters, New York Times, Wall Street Journal, or
the BBC should stimulate peripheral processing of information, particularly in lowinvolvement conditions since a high-involvement audience is more likely to evaluate the
substance of the content rather than simply relying on the source (Petty & Cacioppo, 1986).
Therefore, content shared from credible sources will activate the peripheral route and impact
low-involvement SMBE, i.e., likes. In addition to this primary effect via the peripheral route,
studies demonstrate that source credibility influences via the central route also. This is due to
the credibility of the source being viewed as an “issue-relevant argument by high elaboration
users” (Bhattacherjee & Sanford, 2006). The effect of source credibility through the central
route has been verified in the literature (Kim et al., 2016; Tseng & Wang, 2016), including in
politics (Chebat et al., 1990). By increasing the perceived usefulness of the message, source
credibility triggers higher elaboration. Considering source credibility has an impact in high and
low-involvement conditions, we hypothesize that:
H2a. The source credibility of content will have a positive impact on high-involvement SMBE.
H2b. The source credibility of content will have a positive impact on low-involvement SMBE.
6.4.3 Emotion
Political marketing consultants recommend affect-laden communications (Serazio, 2017).
Emotions’ positive impact on political behavior is established in political psychology (Brader
& Markus, 2013). Emotional political content is ideal on social media (Samuel-Azran et al.,
2015; Bronstein, 2013). In Aristotelian rhetoric, pathos (emotions) is one of the three primary
modes of persuasion. Research shows that stronger voter relationships are predominantly
contingent upon social and emotional aspects of the voter-political brand exchange (Abid &
Harrigan, 2020). Voting itself is an emotional act (Bruter & Harrison, 2017). Similarly,
emotions are the primary driver of content virality on social media (Tellis et al., 2019; Berger
& Milkman, 2012). Expectedly, emotional news articles are more likely to be shared (Berger,
2011).
The integration of emotions in the ELM framework remains limited due to the ELM being a
model of cognition rather than affect. However, emotional appeals, which feature in a message
(e.g. hope, fear, anger, joy), influence a receiver’s judgments through cognitive processes
(Petty & Briñol, 2015). Consequently, recent studies have integrated emotions into the ELM
121
(Manca et al., 2020; Xiang et al., 2019). As per the ELM, emotion’s impact is simple in low
involvement conditions where emotions lead to attitude changes that are consistent with the
message. Regarding the central route, however, emotions can not only act as arguments but
can also trigger biased thinking (Petty & Briñol, 2015). Therefore, emotion has an impact
through both central and peripheral routes, albeit via a more complicated process in the central
route. This has been confirmed in the ELM literature (Morris et al., 2005). Considering these
dual effects of emotion, we propose that:
H3a. Emotional content will have a positive impact on high-involvement SMBE.
H3b. Emotional content will have a positive impact on low-involvement SMBE.
6.4.4 Valence
Substantial research is devoted to the study of the valence of WOM, eWOM, online reviews,
and social media posts (De Pelsmacker et al., 2018; Hayes et al., 2018; Kumar et al., 2016;
Floh et al., 2013; Kwak et al., 2010; de Matos & Rossie, 2008). A significant amount of
literature supports the positivity bias, i.e., positive information is more rewarded on social
media (Reinecke & Trepte, 2014). For example, friends are more likely to comment on status
updates that are of positive valence (Ziegele & Reinecke, 2017). Similarly, posts with positive
and negative valences have positive and negative effects on relationship development
respectively (Orben & Dunbar, 2017). Recent studies show that content virality is also driven
by positive valence (Tellis et al., 2019). However, these researchers typically studied low
involvement situations such as friends’ life updates or influencers selling products. Negative
campaigning is a serious issue that has been on the rise (Borah et al., 2018), which poses a
question as to whether positive valence is becoming less effective in the current climate of
polarization. This concern is pertinent because the audiences of negative information are
normally highly involved.
ELM literature shows that positive valence is processed using the peripheral route, whereas
messages that are negative in valence are processed via the central route (Morris et al., 2005).
This is due to negative messages being perceived as diagnostic and credible, and therefore,
requiring greater effort and elaboration to process. There is substantial evidence for this
negativity bias (Teng et al., 2017; Rim & Song, 2016; Skowronski & Carlston, 1989). For
instance, negative valence triggers greater cognitive elaboration of online reviews (De Maeyer,
122
2012). Similarly, negative valence magnifies the impact of logos (Amos et al., 2019). In the
political context, voters with high involvement are more likely to engage with political ads that
have a negative valence by devoting more cognitive resources towards the message and
carefully processing it (Faber et al., 1993). Since valence has an effect on both high and lowinvolvement audiences, we posit that:
H4a. Negative valence will have a stronger positive impact on high-involvement SMBE than
positive valence.
H4b. Negative valence will have a stronger negative impact on low-involvement SMBE than
positive valence.
6.4.5 Visual symbolism
Visuals are an integral part of social media and lead to greater engagement (Tafesse, 2015).
Visual symbolism leads to a sense of identification, distinction, and prestige on social media
(Fujita et al., 2019). Integrating visual symbolism is an important aspect of political marketing
also (Ormrod et al., 2013; O’Shaughnessy, 2003; Hart, 1995). The current study analyses
patriotic symbolism since it is frequently utilized in American politics and allows for objective
and reliable coding. Other forms of symbolism like religious symbolism or protest symbolism
are not included. As per the ELM, visual symbols like religious or sacred symbols are
peripheral cues (Lumpkins, 2010; Dotson & Hyatt, 2000).
Posting symbolic images on social media, which are built upon a brand’s and its online
followers’ shared identity, artifacts, rituals, or values, increases the likelihood of favorable
responses (Fujita et al., 2019; Tafesse & Wien, 2018). The use of visual symbolism in
presidential campaigns has been analyzed in the literature. The integration of the American
flag, statue of liberty, bald eagle, and red-white-blue into campaign posters and images goes
back two hundred years (Benoit, 2019). Patriotic symbols increase user responses on Instagram
and were used to excellent effect by Donald J. Trump during the last presidential campaign in
2016 (Muñoz & Towner, 2017). These symbols included the Whitehouse, military, police, state
flags, and firefighters (Muñoz & Towner, 2017). Other symbols like the US Bill of Rights and
the Capitol Building are also considered in this study.
123
Like the peripheral cues discussed earlier, studies embedded in the ELM demonstrate that
sacred or religious symbols act as both peripheral cues and the central part of a persuasive
message (Lumpkins, 2010; Dotson & Hyatt, 2000). For instance, Lumpkins (2010) concluded
that high-involvement participants were equally or more likely to process sacred symbols
(Christian Cross). This conclusion is consistent with prior work that demonstrated the positive
impact of religious symbols on high-involvement subjects (Dotson & Hyatt, 2000). Lumpkins
(2010) explains that this is because of possible dual processing of central and peripheral routes
(MacKenzie et al., 1986), whereas Dotson and Hyatt (2000) assert that peripheral cues should
not be viewed in a deterministic way. Based on these studies, we put forth the following
hypotheses:
H5a. Visual symbolism will have a positive impact on high-involvement SMBE.
H5b. Visual symbolism will have a positive impact on low-involvement SMBE.
6.4.6 Social media behavioral engagement and online relationship quality
Engagement with brands on social media has a positive impact on relationship quality (Clark
et al., 2017; Achen, 2016). For instance, followers who engaged with the official pages of the
National Basketball Association (NBA) teams enjoyed higher overall relationship quality.
Similarly, students who engaged with their university’s social media channels also exhibited
high levels of relationship quality (Clark et al., 2017). Other studies demonstrate that SMBE
(conceptualized as post popularity) leads to stronger relationships on social media (Abid et al.,
2019) and that SMBE (conceptualized as content receptivity or post popularity) leads to various
favorable outcomes from a marketing perspective (Kumar et al., 2016; Chang et al., 2015).
Since social media engagement with a brand’s content or page is linked to stronger
relationships, we predict that both low and high-involvement SMBE will lead to an increase in
online relationship quality.
H6a. High-involvement SMBE will have a positive impact on online relationship quality.
H6b. Low-involvement SMBE will have a positive impact on online relationship quality.
6.5 Research methodology
To understand the impact of political MGC’s aforementioned cues, the study conducted a
content analysis of the official Facebook pages of the two primary political parties of the United
124
States of America. Online content analysis has grown substantially since the advent of social
media (Hannigan et al., 2019). The study utilized a mixed-methods approach that consisted of
qualitative coding of content and comments, and subsequent quantitative analysis. The
approach is frequently utilized to explore the impact of MGC (Tellis et al., 2019; Swani et al.,
2017; Ashley & Tuten, 2015). Online content analysis has been utilized to understand the
impact of political MGC on SMBE (Samuel-Azran et al., 2015; Bronstein, 2013). Importantly,
a content analysis of social media pages depicts actual behavior rather than that which is
investigated in a controlled setting or survey.
6.5.1 Data
Seven out of ten American adults use Facebook. Barring YouTube, it remains more popular
than any other social media platform in the USA. It is more representative of the American
population given its reach encompasses rural areas, older citizens, less-educated, and various
ethnic groups also (Pew Research Centre, 2019a; Hootsuite, 2019). Therefore, Facebook
provides a more reliable and representative audience to test the proposed hypotheses.
The data constituted a sample of 200 posts and 43,500 associated comments which were
sourced from the Facebook pages of the Republican and Democratic parties. A hundred posts
preceding a fixed date were captured from each parties’ official Facebook page. Posts about
seasonal greetings (Thanksgiving, Hanukah, Diwali, Christmas, New Year) and shopping
offers (T-shirts, Caps, Hoods, Trump merchandise) were excluded from the analysis. These
posts had exceptionally low SMBE. Similarly, outliers were removed from the data. The final
sample that was used in the statistical analysis included 169 posts and associated comments.
6.5.2 Coding of content
In the first stage of the coding process, the content was coded. A coding manual was developed
to illustrate the variables (see Table 6.1). Code descriptions were aligned with prior research
(e.g. Abid et al., 2019; Muñoz & Towner, 2017). Argument quality, emotion, source credibility,
valence, and visual patriotic symbolism were coded using dichotomous, binary codes. This is
a standard practice that is frequently employed in similar studies (Ashley & Tuten, 2015;
Sabate et al., 2014; Ertimur & Gilly, 2012).
125
Table 6.1 Content coding manual
Variable
Codes
Description
Emotion
0 = Emotion absent
1 = Emotion present
Argument
Quality
0 = Argument quality
absent
1 = Argument quality
present
0 = Symbolism absent
1 = Symbolism present
A post will be coded as 1 if the post evokes emotions like fear,
anger, hope, joy, empathy, etc., or relies on a sense of identity,
shared values, or celebrates a triumph (Samuel-Azran et al., 2015;
Bronstein, 2013).
A post will be coded as 1 if it relies on accurate and timely facts,
figures, reason, argument’s strength, historical accounts, surveys
or polls, statistics, or experts to support the claim (Abid et al.,
2019).
A post will be coded as 1 if it employs images that integrate visual
symbols of American patriotism like the flag, Whitehouse, bald
eagle, military officers, red-white-blue color scheme, etc. (Muñoz
& Towner, 2017).
A post will be coded as 0 if it is a political attack, uses negative
news or a negative appeal, makes an explicit or implicit reference
to the competing party based on political values or political issues.
A post will be coded as 1 if it has a reliability score of over 32 as
per the Media Bias Chart V. 6. A few examples of these sources
are CNN, MSNBC, ABC News, NBC News, Washington Post,
New York Times, and NPR.
Visual
symbolism
Valence
0 = Negative valence
1 = Neutral or positive
Content’s
source
credibility
0 = High credibility
source absent
1 = High credibility
source present
The primary coder, a final-year doctoral candidate, coded the content for valence and patriotic
symbolism in an objective manner. The few pieces of content that were of a neutral valence
were included in the category of positive valence. Content’s source credibility was coded using
an industry tool known as Media Bias Chart (Version 6), which was developed by a non-profit,
media watchdog organization Ad Fontes Media. The news organizations that had a score of 32
or higher (out of 64) were coded as credible sources, whereas content shared from sources that
had scores below 32 was coded as having low credibility. We did make one exception, Fox
News. It was coded as credible despite having a score below 32. It was frequently used as a
source of content for the Republican party. Considering Fox News is widely perceived as the
most credible source of news among Republicans and conservatives (Gramlich, 2020), we
coded it as having high credibility. Content shared from sources that did not feature in the
Media Bias Chart database and the remaining content were coded as low credibility.
The coding of argument quality and emotion involved subjectivity. Therefore, a second coder,
an expert in the field of digital media and communications, was engaged and the complete set
of posts was shared. Ten percent of the posts were randomly selected by the expert and coded
for argument quality and emotions. The ICR values for the two variables were above .8,
indicating a good inter-coder agreement (MacPhail et al., 2016). Finally, high-involvement
SMBE was operationalized using the number of shares, whereas low-involvement SMBE was
operationalized using the number of likes received by the content.
126
6.5.3 Coding of comments
The study involved the coding of comments to operationalize the dependent variable, online
relationship quality. Online text like social media comments are a great source of insight for
marketers and are increasingly being used to extract linguistic and psychological constructs in
various disciplines including marketing (Berger et al., 2019; Humphreys & Wang, 2018;
Coussement et al., 2017).
Most studies in the area operationalize constructs with the aid of computer-aided analysis (e.g.
LIWC), using either inbuilt dictionaries or custom dictionaries that rely on keywords (e.g.
Dolan et al., 2019). This study relies on human or manual content analysis. This is because
comments responding to political content are complex, have latent meanings, are highly
contextual, include sarcasm, and discuss a wide variety of trending issues, people, and topics.
Consequently, computer-aided text analysis is not ideal in this scenario (Humphreys & Wang,
2018). Manual content analysis is often ignored due to its low efficiency; however, it has higher
validity than computer-aided and AI-aided text analysis tools. This is because humans are far
superior in interpreting and detecting contextual, manifest, and latent meanings (Lee et al.,
2019).
To operationalize online relationship quality, comments for each post were captured via
NCapture and transferred to Evernote for subsequent analysis. Certain data procedures, which
have been highlighted in the literature, were omitted due to the employment of manual coding
(Berger et al., 2019). For instance, data cleaning, spell checks, and removal of common words
were not needed. The comments were read by the primary coder to evaluate whether the
commenter expressed trust, commitment, or satisfaction with the political party posting the
content. For each post, the online relationship quality was calculated as the percentage of total
comments that denoted favorable expressions of either relationship satisfaction, trust, or
commitment. This is in line with prior literature that operationalizes relationship quality by
using social media comments (Abid et al., 2019). The methodology resembles that of LIWC,
which employs frequencies and percentages of words for analysis. However, unlike LIWC,
this study’s unit of analysis was not a single word but the entire comment. A total of 15,048
comments out of the total 43,500 comments were classified as favorable, relational comments.
127
Reliability is usually lower for manual analysis (Lee et al., 2019). To ensure reliability, a
second coder was trained and allocated five percent of the posts. The primary coder was not
involved in the training and allocation process. The average inter-coder agreement for
relationship trust, commitment, and satisfaction was 0.77. No individual value was below 0.7.
The values for Cohen’s kappa were above 0.6. These values, although lower than desired, still
indicate a substantial inter-coder agreement and are considered acceptable (MacPhail et al.,
2016; Hallgren, 2012). The primary reason for a lower ICR was comments that conveyed more
than one of the three constructs. This increased the subjectivity of coding, leading coders to
allocate the same comment to different constructs. Overall this should have little impact since
a comment displaying any or all of the three variables was accounted for only once.
6.6 Results
6.6.1 Descriptive statistics
Before testing the hypotheses, descriptive analyses were conducted to understand the features
of the post content and the comments (see Table 6.2). For the post content, the majority of the
posts did not contain strong arguments (83%). Roughly half of the posts showed high
credibility. More than two-thirds of the posts presented emotions (68%). Fewer posts showed
positive valence (39%). Only 12% of the posts showed patriotic symbols. For the comments,
the high-involvement SMBE (i.e., the number of shares) ranged from 13 to 1350 with an
average of 316 (SD = 288.845). The low-involvement SMBE (i.e., number of likes) ranged
from 94 to 5240 with an average of 1661 (SD = 1015.298). Lastly, the percentage of comments
denoting online relationship quality ranged from 3% to 77% with an average of 32% (SD =
15.226).
128
Table 6.2 Descriptive analysis
Variables
Variable operationalization
Content cues
Argument Quality
Source credibility
Emotion
Valence
Visual Symbolism
Descriptive Analysis*
Percentage
0 = Argument quality absent
83%
1 = Argument quality present
17%
0 = Low source credibility
51%
1 = High source credibility
49%
0 = Emotion absent
32%
1 = Emotion present
68%
0 = Negative valence
61%
1 = Positive valence
39%
0 = Visual Symbolism absent
88%
1 = Visual Symbolism present
12%
Number of shares
Range (13, 1350)
Outcome variables
High-involvement SMBE
Mean (316)
Low-involvement SMBE
Number of likes
Range (94, 6943)
Mean (1816)
Online relationship quality
Percentage
Range (3%, 77%)
Mean (33%)
* Percentages are used to describe binary variables for the Content cues. Range and
Mean are used to describe continuous variables for the SMBE and online relationship
quality.
6.6.2 Hypotheses testing
A multivariate analysis of variance (MANOVA) model was conducted to test the relationships
between MGC cues and SMBE (i.e., H1-H5, see Table 6.3) using SPSS. Novak (1995) argued
that MANOVA is appropriate when hypotheses include multiple dependent variables and
categorical independent variables. Critical assumptions in MANOVA include equal covariance
matrices between groups and normality; and researchers have used natural log transformation
on the variables to normalize the dependent variables when these assumptions are not satisfied
(Steinhorst & Williams, 1985; Helgesen, 2006). High-involvement SMBE and lowinvolvement SMBE were not normally distributed (p-value of the Kolmogorov-Smirnov test <
0.05), and the skewness indicators were 1.668 (standard error of skewness = 0.187) and 1.150
(standard error of skewness = 0.187) respectively, which were greater than 1. Therefore, they
were transformed using the natural log approach for the subsequent analyses. After the
129
transformation, Ln(High Involvement SMBE) ranged from 2.56 to 7.21 (Mean = 5.375, SD = 0.906),
and Ln(Low Involvement SMBE) ranged from 4.54 to 8.56 (Mean = 7.221, SD = 0.666). Following
previous researchers (e.g., Peng & Wang, 2006), the homogeneity assumption of the
MANOVA was checked. The assumption of homogeneity of variance-covariance matrices was
satisfied because the Box’s M test was not significant (p>0.05) and the Levene’s test of equality
of error variances was not significant for the high involvement SMBE (p>0.05). However, the
Levene’s test was significant for the low involvement SMBE (p<0.05) and it is a violation of
the assumption of the equality of variances. Therefore, the interpretation of the univariate Ftest will be stricter and using a lower alpha (0.025), and Pillai’s trace will be used to interpret
the multivariate test results as suggested by Tabavhnick et al. (2019).
Table 6.3 MANOVA Results – Testing H1- H5
Argument
Quality
Source
Credibility
Emotion
Valence
Visual
Symbolism
Ln(High Involvement SMBE)– estimated
marginal means
Not present Present
F-value
5.321
5.638
2.237*
Ln(Low Involvement SMBE)– estimated
marginal means
Not present Present
F-value
7.352
7.208
1.335ns
Low
5.375
High
5.584
Low
7.175
High
7.384
Not present
5.238
Negative
5.956
Not present
5.232
Present
5.721
Positive
5.003
Present
5.726
Not present
7.042
Negative
7.458
Not present
6.977
Present
7.517
Positive
7.101
Present
7.583
2.998ns
14.378***
52.049***
6.095*
Supported
Hypothesis
H1
5.023*
H2b
23.257***
H3a, H3b
12.219**
H4a
15.304***
H5a, H5b
Note: ns means not significant; * means significant at 0.05 level; ** means
significant at 0.01 level; *** means significant at 0.001 level.
Hypothesis 1
The influence of argument quality on the SMBE variables was significant according to the
Pillai’s trace statistic (F(2)=30.685, p<0.001, Partial Eta Squared = 0.275). As shown in Table
6.3, the estimated marginal mean of Ln(High
Involvement SMBE)
was significantly higher when
argument quality was present (mean = 5.638) compared to when it was absent (mean = 5.321,
F(1)=2.237, p<0.05). The result was reflected in the significant relationship between argument
quality and Ln(High Involvement SMBE) (B = 0.317, p=0.05), and the presence of argument quality
was associated with a roughly 32% increase in high-involvement SMBE. However, there was
no relationship between argument quality and Ln( Low Involvement SMBE) (p>0.05). Therefore, H1
was supported.
130
Hypothesis 2
The influence of source credibility on the two SMBE variables was very weak according to the
Pillai’s trace statistic (F(2)=2.497, p=0.086, Partial Eta Squared = 0.030). Source credibility
did not influence Ln(High
Involvement SMBE )
significantly (p>0.05). Therefore, H2a was not
supported. The impact of source credibility on Ln(Low Involvement SMBE) was marginal considering
the equality of variance issue on Ln(High Involvement SMBE) (F(1)= 5.023, p=0.026), but it is worth
reporting. The estimated marginal mean of Ln(Low Involvement SMBE) was higher when source
credibility was high (mean = 7.384) compared to when source credibility was low (mean =
7.175). The result was reflected in the significant relationship between source credibility and
Ln(Low
Involvement SMBE)
(B = 0.209, p=0.026), and a higher level of source credibility was
associated with a roughly 21% increase in low involvement SMBE. Although the interpretation
of H2b’s result was influenced by the equality of variance issue on Ln( High Involvement SMBE), we
believe that the predicted relationship, which is suggested in the literature, exists.
Hypothesis 3
The influence of emotion on the two SMBE variables was significant according to the Pillai’s
trace statistic (F(2)=11.572, p<0.001, Partial Eta Squared = 0.125). The estimated marginal
mean of Ln(High Involvement SMBE) was significantly higher when emotion was present (mean =
5.721) compared to when emotion was absent (mean = 5.238, F(1)=14.378, p<0.001). The
result was reflected in the significant relationship between emotion and Ln(High Involvement SMBE)
(B = 0.483, p<0.001), and the presence of emotion was associated with a roughly 48% increase
in high involvement SMBE. Consequently, H3a was supported. Similarly, the estimated
marginal mean of Ln(Low Involvement SMBE) was significantly higher when emotion was present
(mean = 7.517) compared to when emotion was absent (mean = 7.042, F(1)=23.257, p<0.001).
The result was reflected in the significant relationship between emotion and Ln( Low Involvement
SMBE)
(B = 0.475, p<0.001), and the presence of emotion was associated with a 47.5% increase
in low involvement SMBE. Therefore, H3b was supported.
To further understand the difference of emotion’s impact on high and low involvement SMBE,
Cumming (2009) method was used. Following Cumming (2009), we used Z-scores in the
model and then compared the Confidence Intervals (CIs) between the two coefficients. The
same method was applied in the following comparisons of coefficients. The CIs of the
coefficient between the Z-scores of emotion and Ln(High Involvement SMBE) (B=0.533) were from
0.255 to 0.810; and the CIs of the coefficient between the Z-scores of emotion and Ln(Low
131
Involvement SMBE)
(B=0.712) were from 0.421 to 1.004. Half of the average of the overlapping
confidence intervals was calculated (0.142) and added to the low involvement SMBE lower
bond (0.421), which yielded 0.563. The difference between the two coefficients was not
significant (p>0.05) because the high involvement SMBE upper bond (0.810) exceeded the
value of 0.563.
Hypothesis 4
The influence of valence on the two SMBE variables was significant according to the Pillai’s
trace statistic (F(2)=30.685, p<0.001, Partial Eta Squared = 0.275). The estimated marginal
mean of Ln(High Involvement SMBE) was significantly higher when the valence was negative (mean
= 5.956) compared to when the valence was positive (mean = 5.003, F(1)=52.049, p<0.001).
The result was reflected in the significant relationship between valence and Ln( High Involvement
SMBE)
(B = -0.953, p<0.001), and the negative valence was associated with a 95% increase in
high involvement SMBE. Therefore, H4a was supported. The estimated marginal mean of
Ln(Low Involvement SMBE) was significantly higher when the valence was negative (mean = 7.458)
compared to when the valence was positive (mean = 7.101, F(1)=12.219, p<0.01). The result
was reflected in the significant relationship between valence and Ln( Low Involvement SMBE) (B = 0.357, p<0.001), and the negative valence was associated with a roughly 36% increase in low
involvement SMBE. This result was contradictory to the prediction, therefore, H4b was not
supported.
The CIs of the coefficient between the Z-scores of valence and Ln(High Involvement SMBE) (B=1.052) were from -1.340 to -0.764; and CIs of the coefficient between the Z-scores of valence
and Ln(Low Involvement SMBE) (B=-0.536) were from -0.838 to -0.233. The difference between the
two coefficients was significant and the influence of negative valence on low involvement
SMBE was weaker (p<0.05).
Hypothesis 5
The influence of visual symbolism on the two SMBE variables was significant according to
the Pillai’s trace statistic (F(2)=7.924, p<0.01, Partial Eta Squared = 0.275). The estimated
marginal mean of Ln(High Involvement SMBE) was significantly higher when the visual symbolism
was present (mean = 5.726) compared to when the visual symbolism was absent (mean = 5.232,
F(1)=6.095, p<0.05). There was a positive relationship between visual symbolism and Ln( High
Involvement SMBE)
(B = 0.494, p<0.001), and the presence of the visual symbolism was associated
132
with a roughly 49% increase in high involvement SMBE. Therefore, H5a was supported. The
estimated marginal mean of Ln(Low Involvement SMBE) was significantly higher when the visual
symbolism was present (mean = 7.583) compared to when the visual symbolism was absent
(mean = 6.977, F(1)=15.304, p<0.001). The result was reflected in the significant relationship
between visual symbolism and Ln(Low Involvement SMBE ) (B = 0.606, p<0.001), and the presence
of the visual symbolism was associated with a roughly 61% increase in low involvement
SMBE. Therefore, H5b was supported.
The CIs of the coefficient between the Z-scores of between visual symbolism and Ln(High
Involvement SMBE)
(B=0.545) were from 0.109 to 0.982; and the CIs of the coefficient between the
Z-scores of visual symbolism and Ln( Low Involvement SMBE) (B=0.909) were from 0.450 to 1.368.
The difference between the two coefficients was not significant (p>0.05).
Hypothesis 6
Linear regression was used to test the relationships between SMBE and online relationship
quality (H6a and H6b, see Table 6.4). Relationship quality was not normally distributed (pvalue of the Kolmogorov-Smirnov test < 0.05) and it was transformed using the natural log
function. After the transformation, Ln(Relationship Quality) was ranged from 1.07 to 4.35 (mean =
3.336; SD = 0.536). There were no collinearity issues because the values of Variance Inflation
Factors (VIF) were lower than 5; and no normality issue according to the P-P Plot of regression
standardized residual (Pallant, 2016; Hair et al., 2010). The model was good (F(2,166)=39.711,
p<0.001), and more than 32% of the variance was explained by the model. Ln( High Involvement
SMBE)
was positively and significantly related to Ln( Relationship Quality) (B=0.114, p<0.05), and a
one percent increase in high involvement SMBE was associated with about 11% increase in
relationship quality. Therefore, H6a was supported. Ln(Low Involvement SMBE) was positively and
significantly related to Ln( Relationship Quality) (B=0.331, p<0.001), and a one percent increase in
low involvement SMBE was associated with a 33% increase in relationship quality. Therefore,
H6b was supported. The CIs of the coefficient between the Z-scores of Ln(High Involvement SMBE)
and Ln(Relationship Quality) (B=0.193) were from 0.007 to 0.379; and CIs of the coefficient between
the Z-scores of Ln(Low Involvement SMBE) and Ln(Relationship Quality) (B=0.412) were from 0.226 to
0.597. The difference between the two coefficients was not significant (p>0.05).
133
Table 6.4 Regression Results – Testing H6
DV: Ln(Relationship Quality)
B
Constant
Ln(High Involvement SMBE)
Ln(Low Involvement SMBE)
0.333
0.114
0.331
Std.
Error
0.386
0.056
0.076
Beta
t
0.193
0.412
0.863
2.054
4.378
pvalue
0.389
0.042
0.000
95% CI for B
Lower
Upper
Bound Bound
-0.429
1.095
0.004
0.224
0.182
0.480
Note: DV = dependent variable; B = unstandardized coefficient; Beta = standardized
coefficient; CI = Confidence Interval.
To confirm the results and to further explore the potential mediation effect of the SMBE
variables, Hayes’ Process Model 4 was used in SPSS (Hayes, 2013). PROCESS was used due
to its advantages in analyzing dichotomous independent variables and its ability to test multiple
mediators simultaneously (Hayes, 2012). The recommended bias-corrected (BC) confidence
intervals (CI) and 5000 bootstrap samples were used. PROCESS only allows the investigation
of a single independent variable. However, this limitation of the software is resolvable as
PROCESS allows the inclusion of covariates. Mathematically, PROCESS treats covariates as
exactly the same as independent variables (Hayes, 2012).
In the PROCESS analysis, all of the above results and tested relationships were confirmed (see
Table 6.5). The variance explained for Ln(High
Ln(Relationship
Quality)
Involvement SMBE),
Ln(Low
Involvement SMBE),
and
were 31.1% (F(5,163)=14.738, p<0.001), 23.9% (F(5,163)=10.219,
p<0.001), and 35.7% (F(7,161)=12.746, p<0.001) respectively. A set of mediation tests were
conducted for each of the independent variables (i.e., MGC cues).
134
Table 6.5 PROCESS Results
95% CI for B
B
DV: Ln(High
Std. Error
Beta
t
p-
Lower
Upper
value
Bound
Bound
22.943
0.000
4.567
5.426
Involvement
SMBE)
Constant
4.996
0.218
Argument Quality
0.317
0.162
0.350
1.960
0.050
0.002
0.636
Source Credibility
0.209
0.121
0.116
1.732
0.085
-0.029
0.447
Emotion
0.483
0.127
0.250
3.792
0.000
0.231
0.734
Valence
-0.953
0.132
-0.515
-7.215
0.000
-1.214
-0.692
Visual Symbolism
0.494
0.200
0.181
2.469
0.015
0.099
0.889
Constant
6.677
0.168
39.649
0.000
6.344
7.009
Argument Quality
-0.144
0.125
-0.217
-1.155
0.250
-0.391
0.102
Source Credibility
0.209
0.093
0.157
2.241
0.026
0.025
0.393
Emotion
0.475
0.098
0.335
4.823
0.000
0.280
0.669
Valence
-0.357
0.102
-0.262
-2.908
0.000
-0.559
-0.155
Visual Symbolism
0.606
0.155
0.301
3.912
0.000
0.299
0.911
Constant
0.234
0.423
0.554
0.581
-0.602
1.070
Ln(High Involvement SMBE)
0.201
0.070
0.340
2.886
0.004
0.064
0.339
Ln(Low Involvement SMBE)
0.247
0.090
0.307
2.736
0.007
0.069
0.425
Argument Quality
0.002
0.099
0.003
0.018
0.985
-0.193
0.197
Source Credibility
0.097
0.070
0.091
1.374
0.171
-0.042
0.236
Emotion
0.017
0.078
0.015
0.218
0.828
-0.137
0.172
Valence
0.208
0.089
0.190
2.331
0.021
0.032
0.384
Visual Symbolism
0.021
0.121
0.013
0.170
0.865
-0.218
0.259
DV:
Ln(Low
Involvement
SMBE)
DV: Ln(Relationship Quality)
Note: DV = dependant variable; B = unstandardized coefficient; Beta = standardized
coefficient; CI = Confidence Interval. To avoid confusion, the direct effects of the
independent variables on the dependent variable are reported in the manuscript.
Neither Ln(High Involvement SMBE) nor Ln(Low Involvement SMBE) mediated the relationship between
argument quality and Ln( Relationship Quality). Ln(Low Involvement SMBE) (effect =0.052, Bootstrap SE
=0.034, Bootstrap CI [0.002,0.130]) mediated the relationship between source credibility and
Ln(Relationship
Quality).
It was a partial mediation because the direct effect between source
credibility and Ln(Relationship
Quality)
was significant (B=0.190, p<0.05), and the other
135
relationships in the mediation (i.e., from independent variable to the mediator, and from the
mediator to the dependent variable) were significant as reported before.
Both Ln(High Involvement SMBE) (effect =0.097, Bootstrap SE =0.041, Bootstrap CI [0.025,0.186])
and Ln(Low Involvement SMBE) (effect =0.117, Bootstrap SE =0.061, Bootstrap CI [0.016,0.255])
mediated the relationship between emotion and Ln(Relationship
Quality).
They were partial
mediations because the direct effect between emotion and Ln(Relationship Quality) was significant
(B=0.231, p<0.01), and the other relationships in the mediation (i.e., from independent variable
to the mediator, and from the mediator to the dependent variable) were significant as reported
before.
Similarly, both Ln(High Involvement SMBE) (effect =-0.192, Bootstrap SE =0.068, Bootstrap CI [0.335,-0.066]) and Ln(Low Involvement SMBE) (effect =-0.088, Bootstrap SE =0.051, Bootstrap CI [0.203,-0.010]) mediated the relationship between valence and Ln(Relationship Quality). They were
full mediations because the direct effect between valence and Ln( Relationship
Quality)
was not
significant (B=-0.072, p>0.05), but the other relationships in the mediation were significant.
Both Ln(High Involvement SMBE) (effect =0.099, Bootstrap SE =0.051, Bootstrap CI [0.017,0.216])
and Ln(Low Involvement SMBE) (effect =0.149, Bootstrap SE =0.078, Bootstrap CI [0.025,0.325])
mediated the relationship between visual symbolism and Ln(Relationship Quality). They were partial
mediations because the direct effect between visual symbolism and Ln( Relationship Quality) was
significant (B=0.269, p<0.05), and the other relationships in the mediation were significant.
6.7 Discussion and implications
Underpinned in the ELM, our study attempted to develop a holistic framework that offered
insights into how MGC leads to behavioral engagement, which consequently leads to stronger
online relationships. We classified likes and shares as low and high-involvement SMBE
respectively and validated this classification to a certain degree. A secondary aim was to offer
a nuanced understanding of the effects of argument quality and various peripheral cues (source
credibility, emotion, valence, and visual symbolism) in the online political context. We
proposed that argument quality will only have an effect on high-involvement SMBE, whereas
peripheral cues will have an impact via both low and high-involvement SMBE in manners
consistent with the literature. All but two hypotheses were accepted. Firstly, the positive effect
136
of source credibility on high-involvement SMBE was not significant. Secondly, the effect of
negative valence on low-involvement SMBE was contrary to the direction hypothesized.
6.7.1 Marketer-generated content and social media behavioral engagement
Argument quality is a significant predictor of high-involvement SMBE but has no effect on
low-involvement SMBE. This is in line with the central tenet of ELM, that argument quality
impacts through the central route, which is activated when receivers have high involvement
with the message (Petty & Cacioppo, 1986). Further, our findings mirror earlier studies that
explore SMBE with political brands. These studies show that logic (logos) does not impact
likes (Gerodimos & Justinussen, 2014) but it is a predictor of comments and shares (SamuelAzran et al., 2015; Bronstein, 2013). Combined this body of knowledge demonstrates that an
argument’s place in politics is not diminished yet, albeit, the impact of an argument’s strength
is limited to followers who have high involvement. The analysis shows that the content’s
source credibility had a significant but weak relationship with low-involvement SMBE. The
relationship is in line with the literature and the relatively small effect on SMBE is explicable.
In a polarized and post-truth era, the content’s source credibility is less meaningful. This is
particularly true in politics because political followers are driven by motivated reasoning
(Slothuus & Vreese, 2010). Yet another explanation is that followers assess the credibility of
the content sharer and not the content’s source credibility (Sterrett et al., 2019; Turcotte et al.,
2015). Moreover, a possible reason might be banner or source-blindness, behaviors exhibited
by online followers of political entities (Veenstra, 2017; Boerman & Kruikemeier, 2016).
Traditionally, researchers utilizing the ELM have refrained from including the emotional
element of a message in their investigations. Our study is among the few ELM studies that
integrate the emotional component of the content. It confirms the scant literature that highlights
the dual effects of emotion in high and low-involvement conditions (Morris et al., 2005). The
study demonstrates that emotions have a strong effect on high-involvement SMBE also. This
happens through the central route when emotions act as an argument or trigger biased thinking
in social media followers (Petty & Briñol, 2015; Morris et al., 2005). Furthermore, emotion
was the strongest predictor of low-involvement SMBE. This validates past research which
shows that pathos (emotion) is the most effective rhetorical appeal in driving likes in the
political context (Samuel-Azran et al., 2015; Gerodimos & Justinussen, 2014; Bronstein,
2013). The findings highlight that emotions are an integral part of politics. They are highly
137
effective in engaging followers having varying levels of involvement. Like non-political
content (Tellis et al., 2019; Berger & Milkman, 2012), emotions are an important driver of
SMBE for political content also.
Valence is a cue that has distinct impacts in high and low-involvement conditions (Hayes et
al., 2018; Faber et al., 1993). In line with our hypothesis, we found that high-involvement
SMBE was driven by negative valence. This is logical because negative information is
perceived as more diagnostic in nature and more demanding of careful elaboration, and
therefore, it is processed through the central route (Teng et al., 2017). However, contrary to
our hypothesis, negative valence had a significant, positive impact on low-involvement SMBE
also. This demonstrates that in the current climate of political partisanship, polarization, and
animosity (Pew Research Centre, 2019b), negative valence is highly effective in generating
engagement on social media, which is why negative campaigning has been on the rise (Borah
et al., 2018). Although the findings are contradictory to the hypothesis, it is worth noting that
the influence of negative valence on low-involvement SMBE was significantly lower than its
impact on high-involvement SMBE, which is not only in line with the ELM but also indicative
of shares being an activity that depicts higher involvement than likes.
Our study corroborates ELM studies devoted to visual symbolism. Like religious and sacred
symbols (Lumpkins, 2010; Dotson & Hyatt, 2000), patriotic symbols act as peripheral cues
that trigger the peripheral processing in low-involvement followers. The ability of visual
patriotic symbolism in garnering responses on social media is validated in the political
marketing literature (Muñoz & Towner, 2016). The effects of visual symbols on high and lowinvolvement SMBE are a confirmation of earlier studies, which assert that visual symbols have
a positive impact on information processing in high and low-involvement audiences alike
(Lumpkins, 2010; Dotson & Hyatt, 2000).
6.7.2 Social media behavioral engagement and online relationship quality
The results demonstrate that SMBE, whether high or low-involvement, is a driver of online
relationship quality. This is consistent with prior literature that establishes the significance of
engagement with official social media pages in driving followers’ relationship quality (Clark
et al., 2017; Achen, 2016). Other studies show that social media engagement generates trust,
commitment, and satisfaction (Agnihotri et al., 2016; Habibi et al., 2014; Turri et al., 2013),
138
along with the development of emotional bonds (Sashi, 2012). Therefore, it is unsurprising that
SMBE has a strong effect on online relationship quality. The mediation results show that
SMBE, both high and low, plays an important mediating role between peripheral cues and
online relationship quality. Source credibility, emotion, and visual symbolism directly
influenced SMBE and online relationship quality, and therefore, these peripheral cues are most
suited to fostering online relationships with voters. Negative valence, however, without a
positive impact on SMBE, did not impact online relationship quality. This is consistent with
prior literature since posts with negative valence are less effective at cultivating relationships
(Orben & Dunbar, 2017). Therefore, negative valence might generate engagement but are less
effective at building online relationships with followers.
6.7.3 Theoretical implications
The study makes contributions to three areas of literature: social media, ELM, and political
marketing. Firstly, we offer a conceptual framework that links MGC, SMBE, and online
relationship quality. Numerous studies focus on MGC’s effects on SMBE and online
relationships, as well as SMBE’s impact on online relationships (e.g. Dolan et al., 2019; Abid
et al., 2019; Tafesse & Wien, 2018; Clark et al., 2017; Ashley & Tuten, 2015), but studies
integrating MGC, SMBE, and relationship variables remain scant in literature. Secondly, our
classification of social media responses as high-involvement and low-involvement SMBE is
an uncommon approach (e.g. Dolan et al., 2019) which has implications for online content
analyses of social media pages. This is because dual-process theories like the ELM, message
appeal (rational/emotional), and heuristic-systematic model are widely used to investigate
MGC and assume the same two routes to persuasion. By lumping the social media responses
(likes, shares, comments) together, researchers miss out on a deeper understanding.
We extend the ELM to integrate and understand the effect of emotions in MGC and study
patriotic symbols as peripheral cues. Few studies embedded in the ELM integrate emotions
(Manca et al., 2020; Xiang et al., 2019), and fewer validate its dual effects (Morris et al., 2005).
The study further contributes to the ELM literature by applying it in the political marketing
context, where its assimilation remains limited despite being a widely utilized model of
persuasion (e.g Iyer et al., 2017). Finally, the study adds to political marketing literature by
identifying MGC cues that can be utilized to generate engagement among followers and
strengthen online relationships.
139
6.7.4 Managerial implications
Our findings have implications for political brands and their social media managers. Political
brands should take a holistic approach that prompts SMBE from both high and low
involvement followers. Based on our findings, we recommend the use of argument quality,
negative valence, and emotion to engage highly involved followers. For instance, they should
be utilized when the mobilization of core supporters is required or when volunteers and
donations are needed. It is recommended that the reception of MGC employing arguments
should be gauged through high-involvement SMBE rather than low-involvement SMBE. If the
aim is to reach a larger and more general audience, which usually has low involvement in
politics, we recommend the use of emotions, credible sources, and visual symbolism. Political
brands are encouraged to focus on designing MGC that strengthens online relationships, in
addition to generating SMBE. Emotions, credible sources, and visual symbolism are ideally
suited to achieve this aim.
6.8 Limitations and future research
The research has several limitations. Firstly, it focuses on only one platform. Other platforms
like Twitter and Instagram are equally relevant to political marketing and these findings may
not explain user engagement on these social media platforms. Secondly, a larger sample would
have been ideal. However, the manual coding of comments justifies the sample size. Thirdly,
the political context of the study limits its generalizability to commercial contexts. Fourthly,
the human coding of the content and comments is a limitation of the study. Additionally, the
two parties were not explored individually. Another limitation of the study is that the analysis
does not control for content variables like content type (text, image, video), posting time,
content interactivity, and content vividness, among others (De Vries et al., 2012). Future
researchers are advised to conduct similar research using Twitter to verify whether retweets
and favorites fit the classification highlighted in this study. Similarly, other dual-process
theories can be employed to verify the findings of this research on Facebook. Finally, future
researchers can investigate the efficacy of other politically relevant cues like religious
symbolism or rhetorical devices.
140
CHAPTER 7: GENERAL DISCUSSION AND CONCLUSION
The final chapter comprises a holistic discussion of the insights attained from the systematic
literature review and the four papers. Further, the theoretical and managerial implications are
also considered. Before a general discussion, a summary of the articles and research questions
are presented to aid the readers.
7.1 Summary and key findings
7.1.1 The current state of research on political social media marketing (Paper 1, RQ1)
The first paper explored political social media marketing in general (RQ1). It provided an
overview of the current state of the literature using the systematic review methodology. It
identified various themes dominating the literature over the last decade, devised a conceptual
framework, and proposed a research agenda for the future. The themes highlighted in the
literature covered various marketing domains like voter behavior, predictive capabilities,
political relationship marketing, campaigning, branding, political marketer-generated content,
and political user-generated content. In addition to themes, chronological and contextual
analyses were also conducted. The themes were augmented by a conceptual framework that
provided a snapshot of the variables studied in the sixty-six articles, offering researchers a
visualization of the current state of literature. The research agenda considered the gaps
highlighted in the review and current marketing emphases. Paper 1 suggested topics for further
exploration, stressing upon the induction of up-to-date concepts like customer engagement,
influencer marketing, social media advertising, ethical marketing, and value co-creation,
among others.
7.1.2 Deconstructing social media-enabled voter relationships (Paper 2, RQ2)
The second paper attempted to understand the nature of social media-enabled voter
relationships. Specifically, it explored the gratifications, drivers, and interactions that underpin
social media-enabled voter relationships. The study relied on the Uses and Gratification
Theory, psychological contract, and service-dominant orientation. The study relied on focus
groups with young voters who followed political entities on social media. It revealed that voter
relationships were primarily driven by informational, social, and entertainment gratifications.
Instrumental and civic gratifications also played a role. A social and emotional approach,
141
mutual benefit, and trust were the predominant drivers of this relationship. Finally,
individuated, developmental, ethical, and relational social media interactions supported the
relationship. The study discovered that following political candidates was a passive activity
and following does not necessarily connote a preference. Finally, the study found that different
psychological contracts existed between voters and different types of political entities.
7.1.3 Marketing orientation of Australian political brands on social media (Paper 3, RQ3)
The third paper aimed to study the marketing orientation adopted by Australian political brands
on social media. The study explored young voters’ perceptions of the marketing orientation of
political brands. Relationship Marketing Orientation served as the conceptual framework of
the study. Focus groups with young voters revealed that Australian political brands do not
implement a relationship marketing orientation. This was not the case for minor political brands
and local politicians who enjoyed a favorable perception. They excelled in developing longterm bonds, educating young voters, and engaging via shared values. Major political brands,
on the contrary, were perceived as self-promoters. The findings were consistent with prior
literature.
7.1.4 Political marketer-generated content, behavioral engagement, and online
relationships (Papers 4 and 5, RQs 4 and 5)
Considering the significance of marketer-generated content, Papers 4 and 5 examined the direct
and indirect influence of marketer-generate content cues on online relationship quality. Papers
4 and 5 relied on Elaboration-Likelihood Model to characterize the content. The outcome
variable, online relationship quality, was operationalized through manual analysis of
comments. Paper 4 explored content cues like argument quality, source credibility,
interactivity, visuals, length, popularity, and volume of comments. The latter four had a
statistically significant relationship with online relationship quality. The analysis also revealed
that three of these relationships were moderated by content curation (length (-), interactivity (), visual (+)).
Paper 5 built on Paper 4 and addressed its shortcomings like sample size and a simplistic
conceptual model. Paper 5 provided a holistic framework that considered the role of both
marketer-generated content and behavioral engagement in driving online relationship quality.
It integrated and validated two generally accepted conclusions regarding social media
142
marketing, i.e., (1) marketer-generated content leads to engagement and (2) engagement leads
to stronger relationships. The independent variables were argument quality, source credibility,
emotions, visual symbolism, and valence. The findings were in line with the ElaborationLikelihood Model and demonstrated the effects of argument quality and peripheral cues on
engagement. Additionally, engagement led to online relationship quality. The study offered
some support to the proposition that likes and shares reflect low-involvement and highinvolvement behavioral engagement respectively.
7.2 An integrative discussion
The thesis explored political social media marketing from the relationship marketing
perspective. It explained voter relationships from the viewpoint of young Australian voters and
the role of political marketer-generated content in driving behavioral engagement and voter
relationships in the American context. Although the foci, methodologies, and contexts of the
papers varied, certain conclusions can be drawn at a holistic level.
The research confirms the existence of online voter relationships (Ancu & Cozma, 2009).
Findings from quantitative and qualitative studies show that these parasocial relationships are
primarily a social and emotional affair rather than a rational one. This is true for relationships
since stronger relationships are based on a high level of social exchange and an average level
of economic exchange (Guo et al., 2017). Various findings in this thesis support this
conclusion. For instance, Paper 5 showed that cues like emotions and visual symbolism drove
engagement and online relationships, whereas Paper 2 suggested that young voters desire
personal and social content from politicians. Similarly, Paper 3 highlighted the role of shared
values and bonding. Indeed, emotions are an integral part of politics and voter behavior (Bruter
& Harrison, 2017; Redlawsk, 2006; Marcus, 2002). The project demonstrates that despite some
rationality (e.g. argument quality’s effect (Paper 5), informational and instrumental
gratifications (Paper 2)), voters are emotional beings. Moreover, the findings reflect the rise of
‘political fandom’, an alternative way of doing politics on social media that sees politicians
develop affective relationships with voters rather than logical or interest-driven allegiances.
The approach has origins in popular culture (Bronstein, 2013; Erikson, 2008). The qualitative
studies also highlight the trend of personalization of politics, where the focus is drifting from
issues and topics to people (Colliander et al., 2017; Adam & Maier, 2010).
143
Consistent with the literature, the project reveals that major political brands do not adopt a
relational approach to political social media marketing. They use them for broadcasting and
electioneering (Parsons & Rowling, 2018; Harris & Harrigan, 2015). Politicians avoid
engaging in dialogues and offer limited interactivity on social media (Grusell & Nord, 2020;
Ryoo & Bendle, 2017). The Republican Party, however, demonstrated some understanding of
a relational approach, focusing on values (over issues), employing a positive tone, and using
calls-to-action.
Yet another finding that is consistent throughout the project is citizen’s low level of
involvement in politics, which can be inferred from the effectiveness of peripheral cues (Papers
4 and 5), as these cues become salient in low-involvement conditions (Petty & Cacioppo,
1986). Similarly, the passive, non-reciprocal, and emotional nature of online voter relationships
indicates a lower degree of involvement in politics (Papers 2 and 3). The findings affirm the
low level of political involvement among the population, particularly young voters, in Western
democracies (Martin, 2012). This explains why argument quality has a limited effect on voters
(Petty & Cacioppo, 1986).
The project demonstrates that politics is an appropriate context to validate and extend
marketing theories and concepts. Political marketing adds diversity to marketing research and
has the potential to advance theories and frameworks that are indigenous to marketing (LeesMarshment, 2019). However, contextual differences do exist. For instance, positive emotions
are more viral in non-political contexts (Tellis et al., 2019; Berger & Milkman, 2012).
Similarly, argument quality and source credibility are more effective in other contexts (Hur et
al., 2017). Interestingly, these deviations shed further light on politics. The viral nature of
negative valence might be due to political polarization, whereas motivated reasoning and
source-blindness may explain the small effects of argument quality and source credibility.
7.3 Overall theoretical implications
In addition to the contributions made by each study, the thesis contributes to the fields of
relationship marketing, social media marketing, and political marketing as a whole.
The research project extends the relationship marketing paradigm to the political context, an
area where research is infrequent (e.g. Hultman et al., 2019; Parsons & Rowling, 2018). Thus,
144
the project offers a novel relational perspective to political marketing literature by explicating
the development of online relationships from quantitative and qualitative perspectives. Further,
prior literature had not studied online relationships in relation to marketer-generated content.
The thesis offers in-depth insights into the type and characteristics of marketer-generated
content that facilitates online relationships. Marketer-generated content is the primary vehicle
of communication and engagement with customers on social media. Therefore, an emphasis
on content is mandated in the study of online relationships.
Social media marketing has a significant role to play in reviving relationship marketing (Sheth,
2017; Verma et al., 2016). The thesis furthers our understanding of social media marketing
from a relational viewpoint. It affirms that social media work best as relationship-building tools
(Achen, 2017). A sale and issue-dominated orientation conflicts with the spirit of social media.
Moreover, the project contributes to the literature by elaborating on the activity of following,
perception of followers, and their content preferences. It also confirms the pervasive influence
of emotional content on social media (e.g. Tellis et al., 2019; Berger & Milkman, 2012). Both
qualitative and quantitative studies highlight the significance of emotions. The ElaborationLikelihood Model is commonly used to study social media (e.g. Colicev et al., 2018; Teng et
al., 2017; Chang et al., 2015). The present study extends this model in Papers 4 and 5 by
investigating unexplored content cues.
Finally, the research project contributes to political marketing by integrating new marketing
concepts. It is a significant contribution given the stagnancy in the field, where calls for “a
second wave of research fueled by the adoption of new marketing theory perspectives” have
been resonating for some time (Ormrod et al., 2013, p. 115; Henneberg & O’Shaughnessy,
2009). Therefore, in line with some recent research (e.g. Hultman et al., 2019; Marder et al.,
2018; Pich et al., 2018), this project endeavors to trigger the induction of currently relevant
marketing concepts into political marketing. Concepts like the psychological contract, servicedominant orientation, relationship marketing orientation, and relationship quality had no
incidence in political marketing. Despite its widespread presence in commercial marketing and
compatibility with politics, the Elaboration-Likelihood Model is rarely used in political
marketing (e.g. Iyer et al., 2017). Thus, the project integrates the dual-process theory
perspective into political marketing. In conjunction with the systematic literature review (Paper
1), the project provides structure and direction to political marketing, which is a nascent and
fragmented domain (Perannagari & Chakrabarti, 2020). Furthermore, it adds depth to our
145
understanding of an important voter segment, young voters (Papers 2 and 3), while offering an
Australian perspective to political social media marketing, which was missing in the literature.
7.4 Managerial implications
The thesis offers political brands and political social media marketers an elaborate
understanding of the political marketing approach and orientation, marketer-generated content,
and voter behavior from a relationship marketing perspective. To build relationships with
voters, political brands should adopt a strategy that highlights the social element of the voterpolitical brand exchange. A social media marketing strategy that relies on arguments and
focuses on issues is not an ideal approach. Such a strategy has limitations on social media
where the average time a user spends per post is 2.5s (desktop) or 1.7s (smartphone) (Fiber,
2020). Political candidates are advised to post social and personal updates, in addition to
professional updates. Recent research shows that even professional updates should include a
personal touch (Colliander et al., 2017). A positive valence is recommended for politicians. A
similar conclusion is drawn in recent literature, which shows that politicians adopt a neutral or
positive valence on social media (Peres et al., 2020; Paul & Sui, 2019). The content should
look to humanize politicians by portraying them as relatable, everyday citizens. For instance,
a picture at a family barbeque or a football game. Similarly, content highlighting shared values,
educational content, and bipartisan content is also recommended. Finally, politicians should
post consistently and frequently.
Regarding political parties’ official pages, visual and emotional marketer-generated content is
ideal. The project recommends content embedded with emotional appeals, particularly those
that are negative in valence (fear, anger, shame, disgust), for increasing engagement on social
media. The values that the political party and its followers share should feature in the content.
The incorporation of patriotic symbols like the US flag and uniformed officers also engages
followers. For instance, content posted by the Republican party was laden with visual
symbolism around share values like patriotism, the right to bear arms, and stronger law
enforcement. Original content is effective in generating engagement. The content shared by the
official party pages should be from credible sources as this will increase engagement. Much of
the preceding discussion focuses on peripheral cues. Argument quality has its place in social
media marketing, although this is limited. Arguments should be used strategically to engage
the highly involved followers like core supporters, possible volunteers, and potential donors.
146
7.5 Societal implications
Modern voters have a low level of involvement in politics and are emotionally driven (Bruter
& Harrison, 2017; Redlawsk, 2006). This is per the modern culture of political fandom on
social media, which sees politicians create emotional and affective allegiances by getting
people to like them (Erikson, 2008). In the future, campaign activities like Pete Buttigieg’s
supporters’ flash mob dances, Bernie Sanders’ #MyBernieStory, or Donald Trump’s MAGA
Challenge might be the norm. These peripheral methods of engaging the voters will reinforce
an enduring criticism of political marketing that political marketing trivializes the political
process (Ormrod et al., 2013; Savigny, 2008).
Moreover, a focus on feel-good moments and a socio-emotional relationship might result in an
ill-informed electorate and undermine the core elements of politics like argument, logic, policy,
and issues. Complex political problems cannot be condensed into 280 characters and social
media are not conducive to a political debate, especially if the goal is to maintain a positive
and likable image (Peres et al., 2020; Paul & Sui, 2019). Notably, social media might reward
candidates who possess traits like attractiveness, social lifestyle, and personal appeal.
On a positive note, social media are egalitarian. Minor parties, often ignored by the mainstream
media, generally perform well on social media and among young voters (Lees-Marshment,
2014; Maier & Tenscher, 2009). The last presidential elections in the US confirm this. Despite
spending big on digital and social media marketing, Michael Bloomberg’s campaign hardly
gained any traction online (Leskin, 2020). Lastly, young voters’ low level of trust in
mainstream political brands, preference for minor parties, and dedication to a single issue have
implications for the future of democracy. For instance, an increase in the number of political
parties or the rise of third parties are plausible in the future.
7.6 Limitations and further research
The individual limitations of the papers have been discussed in the relevant chapters. However,
there are limitations at the holistic level also. Firstly, sample sizes are a shortcoming in the
studies comprising the thesis. Therefore, researchers are encouraged to validate these findings
using larger samples. The second limitation of the project is the inherent subjectivity in the
qualitative coding of data. Future researchers could adopt an automated keyword analysis (e.g.
Dolan et al., 2019) or a text-analysis software (e.g. LIWC) to code the comments. Also, the
147
predominantly deductive approach could have limited the revelation of new insights.
Therefore, a purely inductive approach may be able to generate further insights (Thomas,
2006). Thirdly, the qualitative findings cannot be generalized to other geographic contexts.
Ideally, future researchers should validate these qualitative findings via quantitative research.
Also, the results from the quantitative studies are specific to Facebook. Therefore, the findings
cannot be generalized to other social media platforms having different audiences or uses.
However, there are significant similarities between the results of Papers 4 and 5 and recent
Twitter-based political marketing studies (e.g. Walker et al., 2017). Nevertheless, future
researchers may want to validate the findings using another social media platform.
Fourth, the quantitative analyses do not control for variables like posting time, day, media type,
and political party. Some prior studies control for such variables (e.g. Dolan et al., 2019; De
Vries et al., 2012). Future researchers may want to rectify this shortcoming. Fifth, an
inconsistency in the quantitative studies is the coding of source credibility, which varied in
Papers 4 and 5, the latter being objective and based on an external measure. Lastly, some issues
are inherent in the online content analysis methodology. For example, the actual reach of the
marketer-generated content is not available to researchers unless the account owner agrees to
provide it. Additionally, the adoption of the methodology does not allow the study of the effects
of user- or situation-based variables on voter relationships. Future researchers can incorporate
such variables using experiment and survey methodologies.
Besides these limitations, the political context limits the generalizability of the research project.
The findings need to be replicated in the commercial context to validate their universal
applicability because divergences exist between how social media users behave towards
political and commercial brands (Boerman & Kruikemeier, 2016). Therefore, future research
can look at the commercial applicability of the thesis’ findings. It is also worth mentioning that
since the outcome variable was derived using comments (Papers 4 and 5), it demonstrates a
general relationship between content, engagement, and relationship quality. Future researchers
are encouraged to see if these findings hold for individual voters.
Beyond rectifying the shortcomings of the thesis, future researchers should engage with this
domain as it will remain relevant to the political and social media landscapes (Appel et al.,
2020). Researchers should explore if social media-enabled voter relationships predict offline
attitudes and voter outcomes. In other words, is a relationship approach more effective in
148
increasing voting likelihood? Similarly, it needs to be confirmed whether political brands that
adopt a relational approach are more engaging. Researchers need to demonstrate that
relationship marketing on social media is more effective than other approaches or orientations.
Researchers can explore inter-platform differences in relationship building. For instance, is
Facebook better suited to relationship marketing than Twitter or Instagram?
The thesis portrays a future where voters with low political involvement will use peripheral
processing to make political decisions. There is an urgent need to research the implications of
such behavior on political marketing, democracy, and society. Will such an approach erode the
quality of the political process or will this approach reinvigorate it by engaging young voters,
an alienated political segment. Young voters are used to personal and social strategies from
their favorite celebrities, authors, athletes, and influencers on social media. However, the effect
this approach will have on older voters needs urgent exploration.
149
REFERENCES
Abeza, G., O’Reilly, N., Finch, D., Séguin, B., & Nadeau, J. (2020). The role of social media
in the co-creation of value in relationship marketing: a multi-domain study. Journal of
Strategic Marketing, 28(6), 472-493.
Abid, A., & Harrigan, P. (2020). An exploration of social media-enabled voter relationships
through uses and gratifications theory, psychological contract and service-dominant
orientation. Australasian Marketing Journal, 28(2), 71–82.
Abid, A., Harrigan, P., & Roy, S. (2019). Online relationship marketing through content
creation and curation. Marketing Intelligence & Planning. 28(6), 699-712.
Abid, A., Harrigan, P., & Roy, S. (2020). A relationship marketing orientation in politics:
Young voters’ perceptions of political brands’ use of social media. Journal of Strategic
Marketing, 29(4), 1–16.
Achen, R. M. (2016), The influence of Facebook engagement on relationship quality and
consumer behavior in the National Basketball Association. Journal of Relationship
Marketing, 15(4), 247-268.
Agnihotri, R., Dingus, R., Hu, M. Y., & Krush, M. T. (2016). Social media: Influencing
customer satisfaction in B2B sales. Industrial Marketing Management, 53, 172-180.
Ajiboye, T., Harvey, J., & Resnick, S. (2019). Customer engagement behaviour on social media
platforms: A systematic literature review. Journal of Customer Behaviour, 18(3), 239–256.
Alalwan, A., Rana, N., Dwivedi, Y., & Algharabat, R. (2017). Social media in marketing: A
review and analysis of the existing literature. Telematics and Informatics, 34(7), 1177–1190.
Alhabash, S., Almutairi, N., Lou, C., & Kim, W. (2019). Pathways to virality:
Psychophysiological responses preceding likes, shares, comments, and status updates on
Facebook. Media Psychology, 22(2), 196-216.
Alrubaiee, L., & Al-Nazer, N. (2010). Investigate the impact of relationship marketing
orientation on customer loyalty: The customer's perspective. International Journal of
Marketing Studies, 2(1), 155.
American Institutional Confidence Poll. (2018). The health of American democracy in an era
of hyper polarization. https://bakercenter.georgetown.edu/wp-content/uploads/2018/10/2018American-Institutional-Confidence-Poll-1.pdf [Accessed 20 August 2019]
150
Amoncar, N. (2020). Entrepreneurial marketing and digital political communication – a
citizen-led perspective on the role of social media in political discourse. Journal of Research
in Marketing and Entrepreneurship, 22(2), 145–159.
Amos, C. King, S., & Allred, A. (2019). An Investigation into How the Modes of Persuasion
and Valence affect Perceptions of Online Reviews. Association of Marketing Theory and
Practice Proceedings 2019. Available at: https://digitalcommons.georgiasouthern.edu/amtpproceedings_2019/1
Ancu, M., & Cozma, R. (2009). MySpace politics: Uses and gratifications of befriending
candidates. Journal of Broadcasting & Electronic Media, 53(4), 567-583.
Andrews, J. C., Durvasula, S., & Akhter, S. H. (1990). A framework for conceptualizing and
measuring the involvement construct in advertising research. Journal of Advertising, 19(4), 2740.
Anim, P., Asiedu, F., Adams, M., Acheampong, G., & Boakye, E. (2019). “Mind the gap”: to
succeed in marketing politics, think of social media innovation. The Journal of Consumer
Marketing, 36(6), 806–817.
Antoniades, N. (2020). Political Marketing Communications in Today’s Era: Putting People at
the Center. Society, 57, 646–656.
Antoniades, N., & Mohr, I. (2019). Presidential Candidates’ Popularity and Voter Loyalty in
the Age of Social Media. Society, 56(5), 445–452.
Antoniades, N., & Mohr, I. (2020). Strengthening U.S. Politicians’ Reputation. Society, 57(1),
41–52.
Appel, G., Grewal, L., Hadi, R., & Stephen, A. (2020). The future of social media in
marketing. Journal of the Academy of Marketing Science, 48(1), 79–95.
Arrigo, E. (2018). Social media marketing in luxury brands: A systematic literature review and
implications for management research. Management Research Review, 41(6), 657–679.
Ashley, C., & Tuten, T. (2015). Creative strategies in social media marketing: An exploratory
study of branded social content and consumer engagement. Psychology & Marketing, 32(1),
15-27.
151
Bal, P. M., Chiaburu, D. S., & Jansen, P. G. (2010). Psychological contract breach and work
performance: is social exchange a buffer or an intensifier?. Journal of Managerial Psychology,
25(3): 252-273.
Bannon, D. P. (2005). Relationship marketing and the political process. Journal of Political
Marketing, 4(2-3), 73-90.
Barger, V., Peltier, J., & Schultz, D. (2016). Social media and consumer engagement: a review
and research agenda. Journal of Research in Interactive Marketing, 10(4), 268–287.
Barnes, J. G., & Cumby, J. A. (2002). Establishing customer relationships on the Internet
requires more than technology. Australasian Marketing Journal, 10(1), 36-46.
Baxter, J., & Eyles, J. (1997). Evaluating qualitative research in social geography: establishing
‘rigour’ in interview analysis. Transactions of the Institute of British geographers, 22(4), 505525.
Becker, A. (2020). Trump Trumps Baldwin? How Trump’s Tweets Transform SNL into
Trump’s Strategic Advantage. Journal of Political Marketing, 19(4), 386–404.
Bell, S. J., Auh, S., & Eisingerich, A. B. (2017). Unraveling the customer education paradox:
When, and how, should firms educate their customers?. Journal of Service Research, 20(3),
306-321.
Bennett, C. (2016). Voter databases, micro-targeting, and data protection law: can political
parties campaign in Europe as they do in North America?. International Data Privacy
Law, 6(4), 261–275.
Benoit, W. L. (2019). A functional analysis of visual and verbal symbols in presidential
campaign posters, 1828–2012. Presidential Studies Quarterly, 49(1), 4-22.
Berger, J. (2011). Arousal increases social transmission of information. Psychological
Science, 22(7), 891-893.
Berger, J., & Milkman, K. L. (2012). What makes online content viral?. Journal of Marketing
Research, 49(2), 192-205.
Berger, J., Humphreys, A., Ludwig, S., Moe, W. W., Netzer, O., & Schweidel, D. A. (2019).
Uniting the tribes: Using text for marketing insight. Journal of Marketing, 84(1), 1-25.
152
Berman, R., Melumad, S., Humphrey, C., & Meyer, R. (2019). A Tale of Two Twitterspheres:
Political Microblogging During and After the 2016 Primary and Presidential Debates. Journal
of Marketing Research, 56(6), 895–917.
Berry, L. L. (1983). Relationship marketing. In L. L. Berry, G. L. Shostack, & G, Upah, (Eds.),
Emerging Perspectives on Services Marketing, (p. 25-28). Chicago, IL: American Marketing
Association.
Bhattacherjee, & Sanford. (2006). Influence Processes for Information Technology
Acceptance: An Elaboration Likelihood Model. MIS Quarterly, 30(4), 805–825.
Billard, T., & Moran, R. (2020). Networked political brands: consumption, community and
political expression in contemporary brand culture. Media, Culture & Society, 42(4), 588–604.
Blumenberg, S., Beimborn, D., & Koenig, W. (2008). Determinants of IT outsourcing
relationships: a conceptual model. In Hawaii International Conference on System Sciences,
Proceedings of the 41st Annual (p. 12-12). IEEE.
Blumler, J. G. (1979). The role of theory in uses and gratifications studies. Communication
Research, 6(1): 9-36.
Blumler, J. G., & McQuail, D. (1969). Television in Politics. Chicago: Univ. of Chicago Press.
Bode, L., & Dalrymple, K. (2016). Politics in 140 Characters or Less: Campaign
Communication, Network Interaction, and Political Participation on Twitter. Journal of
Political Marketing, 15(4), 311–332.
Boerman, S., & Kruikemeier, S. (2016). Consumer responses to promoted tweets sent by
brands and political parties. Computers in Human Behavior, 65, 285–294.
Bojei, J., Julian, C. C., Wel, C. A. B. C., & Ahmed, Z. U. (2013). The empirical link between
relationship marketing tools and consumer retention in retail marketing. Journal of Consumer
Behaviour, 12(3), 171-181.
Borah, P., Fowler, E., & Ridout, T. N. (2018). Television vs. YouTube: political advertising in
the 2012 presidential election. Journal of Information Technology & Politics, 15(3), 230-244.
Bossetta, M. (2018). The Digital Architectures of Social Media: Comparing Political
Campaigning on Facebook, Twitter, Instagram, and Snapchat in the 2016 U.S.
Election. Journalism & Mass Communication Quarterly, 95(2), 471–496.
153
Boulianne, S. (2020). Twenty Years of Digital Media Effects on Civic and Political
Participation. Communication Research, 47(7), 947–966.
Bradbury‐Jones, C., Sambrook, S., & Irvine, F. (2009). The phenomenological focus group: an
oxymoron?. Journal of Advanced Nursing, 65(3), 663-671.
Brader, T., & Marcus, G. E. (2013). Emotion and political psychology. In L. Huddy, D. O.
Sears, & J. S. Levy (Eds.), The Oxford handbook of political psychology (p. 165–204). Oxford,
UK: Oxford University Press.
Braekkan, K. F. (2013). "Union Yes": The Moderating Effects of Union Instrumentality on
Psychological Contract Violations. Journal of Workplace Rights, 17(2), 219-237.
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research
in Psychology, 3(2), 77-101.
Brennan, R., & Henneberg, S. C. (2008). Does political marketing need the concept of
customer value?. Marketing Intelligence & Planning, 26(6), 559-572.
Brod, M., Tesler, L. E., & Christensen, T. L. (2009). Qualitative research and content validity:
developing best practices based on science and experience. Quality of Life Research, 18(9),
1263-1278.
Brodie, R., Hollebeek, L., Jurić, B., & Ilić, A. (2011). Customer Engagement: Conceptual
Domain, Fundamental Propositions, and Implications for Research. Journal of Service
Research, 14(3), 252–271.
Bronstein, J. (2013). Like me! Analyzing the 2012 presidential candidates’ Facebook
pages. Online Information Review, 37(2), 173–192.
Bruter, M., & Harrison, S. (2017). Understanding the emotional act of voting. Nature Human
Behaviour, 1(1), 1-3.
Buccoliero, L., Bellio, E., Crestini, G., & Arkoudas, A. (2020). Twitter and politics: Evidence
from the US presidential elections 2016. Journal of Marketing Communications, 26(1), 88–
114.
Butler, P., & Harris, P. (2009). Considerations on the evolution of political marketing
theory. Marketing Theory, 9(2), 149–164.
Cacciotto, M. (2017). Is Political Consulting Going Digital? Journal of Political
Marketing, 16(1), 50–69.
154
Calder, B. J. (1977). Focus groups and the nature of qualitative marketing research. Journal of
Marketing Research, 14(3) 353-364.
Calderón-Monge, E. (2017). Twitter to Manage Emotions in Political Marketing. Journal of
Promotion Management, 23(3), 359–371.
Callaghan, M., McPhail, J., & Yau, O. H. (1995). Dimensions of a relationship marketing
orientation: an empirical exposition. In Proceedings of the Seventh Biannual World Marketing
Congress, 7(2), 10-65.
Cameron, R. (2009). A sequential mixed model research design: Design, analytical and display
issues. International Journal of Multiple Research Approaches, 3(2), 140-152.
Cameron, M., Barrett, P., & Stewardson, B. (2016). Can Social Media Predict Election Results?
Evidence From New Zealand. Journal of Political Marketing, 15(4), 416–432.
Carey, M. A., & Asbury J. E. (2016). Focus group research. Oxon: Routledge.
Carr, C. T., Wohn, D. Y., & Hayes, R. A. (2016). As social support: Relational closeness,
automaticity, and interpreting social support from paralinguistic digital affordances in social
media. Computers in Human Behavior, 62, 385-393.
Chadwick, A., & Vaccari, C. (2019). News sharing on UK social media: Misinformation,
disinformation, and correction. Survey Report.
https://repository.lboro.ac.uk/articles/News_sharing_on_UK_social_media_misinf
ormation_disinformation_and_correction/9471269 [Accessed 12 June, 2020]
Chang, H. H., Lu, Y. Y., & Lin, S. C. (2020). An elaboration likelihood model of consumer
respond
action
to
Facebook
second-hand
marketplace:
Impulsiveness
as
a
moderator. Information & Management, 57(2), 103171.
Chang, Y. T., Yu, H., & Lu, H. P. (2015). Persuasive messages, popularity cohesion, and
message diffusion in social media marketing. Journal of Business Research, 68(4), 777-782.
Chattananon, A., & Trimetsoontorn, J. (2009). Relationship marketing: a Thai case.
International Journal of Emerging Markets, 4(3), 252-274.
Chatzidakis, A., Hibbert, S., Mittusis, D., & Smith, A. (2004). Virtue in consumption?. Journal
of Marketing Management, 20(5-6), 526-543.
Chebat, J. C., Filiatrault, P., & Perrien, J. (1990). Limits of credibility: The case of political
persuasion. The Journal of Social Psychology, 130(2), 157-167.
155
Chen, C. D., & Ku, E. C. (2013). Bridging indistinct relationships and online loyalty: evidence
from online interest-based communities. Online Information Review, 37(5), 731–751.
Choudhury, M. M., & Harrigan, P. (2014). CRM to social CRM: the integration of new
technologies into customer relationship management. Journal of Strategic Marketing, 22(2),
149-176.
Chung, S., & Cho, H. (2017). Fostering parasocial relationships with celebrities on social
media: Implications for celebrity endorsement. Psychology & Marketing, 34(4), 481-495.
Clark, M., Fine, M. B., & Scheuer, C. L. (2017). Relationship quality in higher education
marketing: the role of social media engagement. Journal of Marketing for Higher Education,
27(1), 40-58.
Cogburn, D., & Espinoza-Vasquez, F. (2011). From Networked Nominee to Networked
Nation: Examining the Impact of Web 2.0 and Social Media on Political Participation and Civic
Engagement in the 2008 Obama Campaign. Journal of Political Marketing, 10(1-2), 189–213.
Colicev, A., Malshe, A., Pauwels, K., & O'Connor, P. (2018). Improving consumer mindset
metrics and shareholder value through social media: The different roles of owned and earned
media. Journal of Marketing, 82(1), 37-56.
Cornfield, M. (2017). Empowering the Party-Crasher: Donald J. Trump, the First 2016 GOP
Presidential Debate, and the Twitter Marketplace for Political Campaigns. Journal of Political
Marketing, 16(3-4), 212–243.
Coussement, K., Debaere, S., & De Ruyck, T. (2017). Inferior member participation
identification in innovation communities: The signaling role of linguistic style use. Journal of
Product Innovation Management, 34(5), 565-579.
Cvijikj, I.P., & Michahelles, F. (2011). A case study of the effects of moderator posts within a
Facebook brand page. Social Informatics, Springer Berlin Heidelberg, 161-170.
Cyr, J. (2016). The pitfalls and promise of focus groups as a data collection method.
Sociological Methods & Research, 45(2) 231-259.
Dalton, R. J. (2013). Citizen politics: Public opinion and political parties in advanced
industrial democracies CQ Press, Thousand Oaks, California: CQ Press.
156
De Maeyer, P. (2012). Impact of online consumer reviews on sales and price strategies: a
review and directions for future research. Journal of Product & Brand Management, 21(2),
132–139.
De Pelsmacker, P., Dens, N., & Kolomiiets, A. (2018). The impact of text valence, star rating
and rated usefulness in online reviews. International Journal of Advertising, 37(3), 340-359.
De Vries, L., Gensler, S., & Leeflang, P. S. (2012). Popularity of brand posts on brand fan
pages: An investigation of the effects of social media marketing. Journal of Interactive
Marketing, 26(2), 83-91.
Dean, D., Croft, R., & Pich, C. (2015). Toward a conceptual framework of emotional
relationship marketing: An examination of two UK political parties. Journal of Political
Marketing, 14(1-2), 19-34.
Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of consumer
participation in network-and small-group-based virtual communities. International Journal of
Research in Marketing, 21(3), 241-263.
Dick, A. S., & Basu, K. (1994). Customer loyalty: toward an integrated conceptual
framework. Journal of the Academy of Marketing Science, 22(2), 99-113.
Digital Marketing Institute, (n.d.). How social media can enhance political campaigns,
https://digitalmarketinginstitute.com/en-au/blog/how-social-media-can-enhance-politicalcampaigns [Accessed 14th March, 2020]
Dimitrova, D., & Bystrom, D. (2017). Role of Social Media in the 2016 Iowa
Caucuses. Journal of Political Marketing, 16(3-4), 386–406.
Dolan, R., Conduit, J., Frethey-Bentham, C., Fahy, J., & Goodman, S. (2019). Social media
engagement behavior. European Journal of Marketing, 53(10), 2213–2243.
Dotson, M. J., & Hyatt, E. M. (2000). Religious symbols as peripheral cues in advertising: A
replication of the elaboration likelihood model. Journal of Business Research, 48(1), 63-68.
Doyle, L., Brady, A. M., & Byrne, G. (2009). An overview of mixed methods research. Journal
of Research in Nursing, 14(2), 175-185.
Dwivedi, Y., Ismagilova, E., Hughes, D., Carlson, J., Filieri, R., Jacobson, J., Jain, V.,
Karjaluoto, H., Kefi, H., Krishen, A., Kumar, V., Rahman, M., Raman, R., Rauschnabel, P.,
Rowley, J., Salo, J., Tran, G., & Wang, Y. (2020). Setting the future of digital and social media
157
marketing research: Perspectives and research propositions. International Journal of
Information Management, 102168.
Effing, R., & Spil, T. (2016). The social strategy cone: Towards a framework for evaluating
social media strategies. International Journal of Information Management, 36(1), 1–8.
Egan, J. (1999). Political marketing: lessons from the mainstream. Journal of Marketing
Management, 15(6), 495-503.
Eisingerich, A. B., & Bell, S. J. (2006). Relationship marketing in the financial services
industry: The importance of customer education, participation and problem management for
customer loyalty. Journal of Financial Services Marketing, 10(4), 86-97.
Elder, E., & Phillips, J. (2017). Appeals to the Hispanic Demographic: Targeting through
Facebook Autoplay Videos by the Clinton Campaign during the 2015/2016 Presidential
Primaries. Journal of Political Marketing, 16(3-4), 319–342.
Elo, S., & Kyngäs, H. (2008). The qualitative content analysis process. Journal of Advanced
Nursing, 62(1), 107-115.
Ertimur, B., & Gilly, M. C. (2012). So whaddya think? Consumers create ads and other
consumers critique them. Journal of Interactive Marketing, 26(3), 115-130.
Evans, M., Stoker, G., & Halupka, M. (2018). Australians’ trust in politicians and democracy
hits an all-time low: new research. The Conversation. https://theconversation.com/australianstrust-in-politicians-and-democracy-hits-an-all-time-low-new-research-108161
Faber, R. J., Tims, A. R., & Schmitt, K. G. (1993). Negative political advertising and voting
intent: The role of involvement and alternative information sources. Journal of
Advertising, 22(4), 67-76.
Facebook
Media,
(n.d).
Tips
for
politicians
and
campaigns.
https://www.facebook.com/facebookmedia/best-practices/tips-for-politicians-and-campaigns
[Accessed 1st March, 2018]
Feinberg, M., & Willer, R. (2015). From gulf to bridge: when do moral arguments facilitate
political influence?. Personality and Social Psychology Bulletin, 41(12), 1665-1681.
Fitzgerald, J., & Wolak, J. (2016). The roots of trust in local government in Western Europe.
International Political Science Review, 37(1), 130-146.
158
Flanagin, A., & Metzger, M. J. (2017). Digital media and perceptions of source credibility in
political communication. The Oxford handbook of political communication, 417.
Flanagin, A., & Metzger, M. J. (2001). Internet use in the contemporary media environment.
Human Computer Research, 27, 153-181.
Floh, A., Koller, M., & Zauner, A. (2013). Taking a deeper look at online reviews: The
asymmetric effect of valence intensity on shopping behaviour. Journal of Marketing
Management, 29(5-6), 646-670.
Fujita, M. (2018). Enhancing student-university relationships through social media brand
communities: An identity theories perspective. [Doctoral dissertation, The University of
Western Australia].
Fujita, M., Harrigan, P., Roy, S. K., & Soutar, G. (2019). Two-way acculturation in social
media: The role of institutional efforts. Technological Forecasting and Social Change, 145,
532-542.
Galer, S. S. (2020). US election 2020: TikTok gets pulled into the campaigns. BBC News.
https://www.bbc.com/news/technology-54374710 [Accessed 18th Jan, 2021]
Garbarino, E., & Johnson, M. S. (1999). The different roles of satisfaction, trust, and
commitment in customer relationships. Journal of Marketing, 63(2), 70-87.
Garcia-Castañon, M., Rank, A., & Barreto, M. (2011). Plugged In or Tuned Out? Youth, Race,
and Internet Usage in the 2008 Election. Journal of Political Marketing, 10(1-2), 115–138.
Garramone, G. M., Harris, A. C., & Anderson, R. (1986). Uses of political computer bulletin
boards. Journal of Broadcasting & Electronic Media, 30(3), 325-339.
Gassenheimer, J. B., Houston, F. S., & Davis, J. C. (1998). The role of economic value, social
value,
and
perceptions
of
fairness
in
interorganizational
relationship
retention
decisions. Journal of the Academy of Marketing Science, 26(4), 322-337.
Gerodimos, R., & Justinussen, J. (2014). Obama’s 2012 Facebook Campaign: Political
Communication in the Age of the Like Button. Journal of Information Technology &
Politics, 12(2), 113–132.
Gil de Zuniga, H., Barnidge, M., & Diehl, T. (2018). Political persuasion on social media: A
moderated moderation model of political discussion disagreement and civil reasoning. The
Information Society, 34(5), 302-315. Add for ELM in politics
159
Goh, K. Y., Heng, C. S., & Lin, Z. (2013). Social media brand community and consumer
behavior: Quantifying the relative impact of user-and marketer-generated content. Information
Systems Research, 24(1), 88-107.
Goodrich, K., Schiller, S. Z., & Galletta, D. (2015). Consumer reactions to intrusiveness of
online-video advertisements: do length, informativeness, and humor help (or hinder) marketing
outcomes?. Journal of Advertising Research, 55(1), 37-50.
Gordon, M. E., McKeage, K., & Fox, M. A. (1998). Relationship marketing effectiveness: the
role of involvement. Psychology & Marketing, 15(5), 443-459.
Graduate Research School. (2016). Thesis as a series of papers. The University of Western
Australia
https://www.postgraduate.uwa.edu.au/__data/assets/pdf_file/0006/2930217/ThesisByPapers_
160916.pdf [Accessed 7th June, 2021]
Gramlich,
J.
(2020).
5
facts
about
Fox
News.
Pew
Research
Center.
https://www.pewresearch.org/fact-tank/2020/04/08/five-facts-about-fox-news/ [Accessed 15th
August, 2020].
Grewal, D., Levy, M., Mathews, S., Harrigan, P., Bucic, T., & Kopanidis, F. (2017). Marketing
2e. Australia: McGraw-Hill Education.
Grönroos, C. (1994). From marketing mix to relationship marketing: towards a paradigm shift
in marketing. Asia-Australia Marketing Journal, 2(1), 9-29.
Grönroos, C. (1997). Keynote paper: From marketing mix to relationship marketing-towards a
paradigm shift in marketing. Management Decision, 35(4), 322-339.
Grönroos, C. (2004). The relationship marketing process: communication, interaction,
dialogue, value. Journal of Business & Industrial marketing, 19(2), 99-113.
Grönroos, C., & Gummerus. J. (2014). The service revolution and its marketing implications:
service logic vs service-dominant logic. Managing Service Quality, 24(3), 206-229.
Gruen, T. W. (1995). The outcome set of relationship marketing in consumer
markets. International Business Review, 4(4), 447-469.
Grusell, M., & Nord, L. (2020a). Not so Intimate Instagram: Images of Swedish Political Party
Leaders in the 2018 National Election Campaign. Journal of Political Marketing, 1-16.
160
Grusell, M., & Nord, L. (2020b). Setting the Trend or Changing the Game? Professionalization
and Digitalization of Election Campaigns in Sweden. Journal of Political Marketing, 19(3),
258–278.
Guest, G., Namey, E., & McKenna, K. (2017). How many focus groups are enough? Building
an evidence base for nonprobability sample sizes. Field Methods, 29(1), 3-22.
Gummerus, J., von Koskull, C., & Kowalkowski, C. (2017). Guest editorial: relationship
marketing – past, present and future. Journal of Services Marketing, 32(1), 1-5.
Gummesson, E. (1998). Implementation requires a relationship marketing paradigm. Journal
of the Academy of Marketing Science, 26(3), 242-249.
Gummesson, E. (1994). Broadening and specifying relationship marketing. Asia-Australia
Marketing Journal, 2(1), 31-43.
Gummesson, E. (2017). From relationship marketing to total relationship marketing and
beyond. Journal of Services Marketing, 31(1), 16-19.
Gunarathne, P., Rui, H., & Seidmann, A. (2017). Whose and what social media complaints
have happier resolutions? Evidence from Twitter. Journal of Management Information
Systems, 34(2), 314-340.
Guo, L., Gruen, T., & Tang, C. (2017). Seeing relationships through the lens of psychological
contracts: the structure of consumer service relationships. Journal of the Academy of Marketing
Science, 45(3), 357–376.
Guzmán, F., & Sierra, V. (2009). A political candidate's brand image scale: Are political
candidates brands?. Journal of Brand Management, 17(3), 207-217.
Habibi, M. R., Laroche, M., & Richard, M. O. (2014). Brand communities based in social
media: How unique are they? Evidence from two exemplary brand communities. International
Journal of Information Management, 34(2), 123-132.
Habibi, M. R., Laroche, M., & Richard, M. O. (2014). The roles of brand community and
community engagement in building brand trust on social media. Computers in Human
Behavior, 37, 152-161.
Haenschen, K., & Jennings, J. (2019). Mobilizing millennial voters with targeted internet
advertisements: a field experiment. Political Communication, 36(3), 357-375.
161
Hair, J., Black, W., Babin, B., Anderson, R. and Tathum, R. (2010), Multivariate data analysis
(7th ed.). Saddle River, NJ: Prentice Hall, Inc.
Hajli, M. N. (2014). The role of social support on relationship quality and social
commerce. Technological Forecasting and Social Change, 87, 17-27.
Hajli, N., Shanmugam, M., Papagiannidis, S., Zahay, D., & Richard, M. (2017). Branding cocreation with members of online brand communities. Journal of Business Research, 70, 136144.
Hallgren, K.A. (2012). Computing inter-rater reliability for observational data: an overview
and tutorial. Tutorials in Quantitative Methods for Psychology, 8(1), pp. 23-34.
Hannah, D., Treen, E., Pitt, L., & Berthon, P. (2016). But you promised! Managing consumers’
psychological contracts. Business Horizons, 59(4), 363-368.
Hannan-Morrow, S., & Roden, M. (2014). Gender, Age and Generational Effects on Turnout
in Australian Federal Elections. Australian Election Commission.
https://www.aec.gov.au/About_AEC/research/files/apsa-2014-gender-age-and-generationaleffects-on-turnout-in-australian-federal-elections.pdf
Hannigan, T. R., Haans, R. F., Vakili, K., Tchalian, H., Glaser, V. L., Wang, M. S., ... &
Jennings, P. D. (2019). Topic modeling in management research: Rendering new theory from
textual data. Academy of Management Annals, 13(2), 586-632.
Harmer, E., & Wring, D. (2013). Julie and the Cybermums: Marketing and Women Voters in
the UK 2010 General Election. Journal of Political Marketing, 12(2-3), 262–273.
Harris, L., & Harrigan, P. (2015). Social Media in Politics: The Ultimate Voter Engagement
Tool or Simply an Echo Chamber? Journal of Political Marketing, 14(3), 251–283.
Hart, J. (1995). President Clinton and the politics of symbolism: Cutting the White House
staff. Political Science Quarterly, 110(3), 385-403.
Hasham, N. (2016). Election 2016: Voter turnout lowest since compulsory voting began in
1925.
Sydney
Morning
Herald.
http://www.smh.com.au/federal-politics/political-
news/election-2016-voter-turnout-lowest-since-compulsory-voting-began-in-1925-20160808gqnij2.html
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable
mediation, moderation, and conditional process modeling [White paper].
162
Hayes, A. F. (2013). Model templates for PROCESS for SPSS and SAS.
Hayes, J. L., Shan, Y., & King, K. W. (2018). The interconnected role of strength of brand and
interpersonal relationships
and user comment
valence on brand video sharing
behaviour. International Journal of Advertising, 37(1), 142-164.
Hayes, R. A., Carr, C. T., & Wohn, D. Y. (2016). One click, many meanings: Interpreting
paralinguistic digital affordances in social media. Journal of Broadcasting & Electronic
Media, 60(1), 171-187.
Helgesen, Ø. (2006). Are loyal customers profitable? Customer satisfaction, customer (action)
loyalty and customer profitability at the individual level. Journal of Marketing Management
22,(3-4), 245-266.
Henneberg, S. C. M. (1996). Conference report: second conference on political marketing.
Journal of Marketing Management, 12(8), 777–783.
Henneberg, S. C. (2002). Understanding political marketing. The idea of political marketing,
93-170.
Henneberg, S., & O’Shaughnessy, N. (2009). Political Relationship Marketing: some
macro/micro thoughts. Journal of Marketing Management, 25(1-2), 5–29.
Henneberg, S. C., Scammell, M., & O'Shaughnessy, N. J. (2009). Political marketing
management and theories of democracy. Marketing Theory, 9(2), 165-188.
Highschool Australia. n.d. The Australian Education System. https://www.high-schoolaustralia.com/australian-education-system/
Hollebeek, L. D., Glynn, M. S., & Brodie, R. J. (2014). Consumer brand engagement in social
media: Conceptualization, scale development and validation. Journal of Interactive
Marketing, 28(2), 149-165.
Hollebeek, L., Srivastava, R., & Chen, T. (2019). S-D logic–informed customer engagement:
integrative framework, revised fundamental propositions, and application to CRM. Journal of
the Academy of Marketing Science, 47(1), 161–185.
Homonoff, H. (2020). 2020 Political Ad Spending Exploded: Did It Work? Forbes.
https://www.forbes.com/sites/howardhomonoff/2020/12/08/2020-political-ad-spendingexploded-did-it-work/?sh=382fe51e3ce0 [Accessed 24th Feb, 2021]
163
Hootsuite.
(2019).
What
is
content
creation
vs.
curation?
https://hootsuite.com/education/courses/social-marketing/content/creation-vs-curation
[Accessed 29 August 2019]
Hoppner, J. J., Griffith, D. A., & White, R. C. (2015). Reciprocity in relationship marketing:
A cross-cultural examination of the effects of equivalence and immediacy on relationship
quality and satisfaction with performance. Journal of International Marketing, 23(4), 64-83.
Hossain, M. A., Dwivedi, Y. K., Chan, C., Standing, C., & Olanrewaju, A. S. (2018). Sharing
political content in online social media: A planned and unplanned behaviour
approach. Information Systems Frontiers, 20(3), 485-501.
Howe, L. C., & Krosnick, J. A. (2017). Attitude strength. Annual Review of Psychology, 68,
327-351.
Hsieh, H. F., & Shannon, S. E. (2005). Three approaches to qualitative content
analysis. Qualitative Health Research, 15(9), 1277-1288.
Hughes, A., & Dann, S. (2009). Political marketing and stakeholder engagement. Marketing
Theory, 9(2), 243–256.
Hughes, A., & Dann, S. (2006). Political marketing 2006: direct benefit, value and managing
the voter relationship. The Australian National University, 1-8.
Hultman, M., Ulusoy, S., & Oghazi, P. (2019). Drivers and outcomes of political candidate
image creation: The role of social media marketing. Psychology & Marketing, 36(12), 1226–
1236.
Humphreys, A., & Wang, R. J. H. (2018). Automated text analysis for consumer
research. Journal of Consumer Research, 44(6), 1274-1306.
Hur, K., Kim, T. T., Karatepe, O. M., & Lee, G. (2017). An exploration of the factors
influencing social media continuance usage and information sharing intentions among Korean
travellers. Tourism Management, 63, 170-178.
Ifinedo, P. (2016). Applying uses and gratifications theory and social influence processes to
understand students' pervasive adoption of social networking sites: Perspectives from the
Americas. International Journal of Information Management, 36(2), 192-206.
164
Iyer, P., Yazdanparast, A., & Strutton, D. (2017). Examining the effectiveness of
WOM/eWOM communications across age-based cohorts: implications for political
marketers. The Journal of Consumer Marketing, 34(7), 646–663.
Jackson, N. A., & Lilleker, D. G. (2009). Building an architecture of participation? Political
parties and Web 2.0 in Britain. Journal of Information Technology & Politics, 6(3-4), 232-250.
Jackson, N. A., & Lilleker, D. G. (2007). Seeking unmediated political information in a
mediated
environment:
The
uses
and
gratifications
of
political
parties'
e-
newsletters. Information, Community and Society, 10(2), 242-264.
Jackson, N. A., Lilleker, D. G., & Schweitzer, E. (2012). Political Marketing in an Online
Election Environment: Short Term Sales or Long-Term Relationships?. Routledge Handbook
of Political Marketing, (p. 286-300), London: Routledge.
Jain, V., & Ganesh, B. E. (2020). Understanding the Magic of Credibility for Political Leaders:
A Case of India and Narendra Modi. Journal of Political Marketing, 19(1-2), 15–33.
Jensen, M., & Bang, H. (2017). Populism and Connectivism: An Analysis of the Sanders and
Trump Nomination Campaigns. Journal of Political Marketing, 16(3-4), 343–364.
Jiang, G., Kotabe, M., Zhang, F., Hao, A. W., Paul, J., & Wang, C. L. (2020). The determinants
and performance of early internationalizing firms: A literature review and research
agenda. International Business Review, 29(4), 101662.
Jo, S. (2005). The effect of online media credibility on trust relationships. Journal of Website
Promotion, 1(2), 57-78.
Joathan, Í., & Lilleker, D. G. (2020). Permanent Campaigning: a meta-analysis and framework
for measurement. Journal of Political Marketing, 1-19.
Joiner, K., & Lusch, R. (2016). Evolving to a new service-dominant logic for health
care. Innovation and Entrepreneurship in Health, 3, 25-33.
Jungherr, A. (2016). Twitter use in election campaigns: A systematic literature review. Journal
of Information Technology & Politics, 13(1), 72–91.
Kalliny, M., Ghanem, S., & Kalliny, M. (2018). The Impact of Advertising and Media on the
Arab Culture: The Case of the Arab Spring, Public Spheres, and Social Media. Journal of
Political Marketing, 17(1), 62–89.
165
Kalogeropoulos, A., Negredo, S., Picone, I., & Nielsen, R. K. (2017). Who shares and
comments on news?: A cross-national comparative analysis of online and social media
participation. Social Media + Society, 3(4), 2056305117735754.
Kalsnes, B. (2016). The social media paradox explained: Comparing political parties’
Facebook strategy versus practice. Social Media + Society, 2(2), 2056305116644616.
Kamberelis, G., & Dimitriadis, G. (2013). Focus groups: From structured interviews to
collective conversations. Oxon: Routledge.
Kang, J. W., & Namkung, Y. (2019). The information quality and source credibility matter in
customers’ evaluation toward food O2O commerce. International Journal of Hospitality
Management, 78, 189-198.
Kantar Media. (2018). Political ad spending hits new record for 2018. Kantar Media.
https://www.axios.com/political-ad-spending-hits-new-record-for-2018-midterm-elections1541509814-28e24943-d68b-4f55-83ef-8b9d51a64fa9.html [Accessed 12th August 2019]
Kantar Media. (2019). Kantar Forecasts $6 Billion in Political Ad Spending for 2019-2020
Election Cycle.
Kantar
Media.
https://www.kantarmedia.com/global/newsroom/press-
releases/kantar-forecasts6-billion-in-political-ad-spending-for-2019-2020-election-cycle [26th
June, 2020]
Kaplan, A., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities
of Social Media. Business Horizons, 53(1), 59–68.
Kapoor, K., Tamilmani, K., Rana, N., Patil, P., Dwivedi, Y., & Nerur, S. (2018). Advances in
Social Media Research: Past, Present and Future. Information Systems Frontiers, 20(3), 531–
558.
Karpen, I. O., Bove, L. L., Lukas, B. A., & Zyphur, M. J. (2015). Service-dominant orientation:
measurement and impact on performance outcomes. Journal of Retailing, 91(1), 89-108.
Karpen, I. O., Bove, L. L., & Lukas, B. A. (2012). Linking service-dominant logic and strategic
business practice: A conceptual model of a service-dominant orientation. Journal of service
research, 15(1), 21-38.
Katz, E., Blumler, J. G., & Gurevitch, M. (1973). Uses and gratifications research. The Public
Opinion Quarterly, 37(4), 509-523.
166
Kaye, B. K., & Johnson, T. J. (2002). Online and in the know: Uses and gratifications of the
web for political information. Journal of Broadcasting & Electronic Media, 46(1), 54-71.
Kees, J., Berry, C., Burton, S., & Sheehan, K. (2017). An analysis of data quality: professional
panels, student subject pools, and Amazon's mechanical turk. Journal of Advertising, 46(1),
141-155.
Kemp, E., Bui, M., Krishen, A., Homer, P. M., & LaTour, M. S. (2017). Understanding the
power of hope and empathy in healthcare marketing. Journal of Consumer Marketing, 34(2),
85-95.
Kenski, K., Filer, C., & Conway-Silva, B. (2017). Communicating Party Labels and Names on
Twitter During the 2016 Presidential Invisible Primary and Primary Campaigns. Journal of
Political Marketing, 16(3-4), 267–289.
Khan, M. L. (2017). Social media engagement: What motivates user participation and
consumption on YouTube?. Computers in Human Behavior, 66, 236-247.
Kidd, P. S., & Parshall, M. B. (2000). Getting the focus and the group: enhancing analytical
rigor in focus group research. Qualitative Health Research, 10(3), 293-308.
Kim, D. H., Spiller, L., & Hettche, M. (2015). Analyzing media types and content orientations
in Facebook for global brands. Journal of Research in Interactive Marketing, 9(1), 4-30.
Kim, D., & Johnson. T. J. (2012). Political blog readers: Predictors of motivations for accessing
political blogs. Telematics and Informatics, 29(1), 99-109.
Kim, M. J. , Chung, N. , Lee, C.-K. , & Preis, M. W. (2016). Dual-route of persuasive
communications in mobile tourism shopping. Telematics and Informatics , 33(2), 293–308.
Kinder, D. R. (1986). Presidential character revisited. Political Cognition. Hillsdale, NJ:
Lawrence Erlbaum.
Kitzinger, J. (1994). The methodology of focus groups: the importance of interaction between
research participants. Sociology of Health & Illness, 16(1), 103-121.
Koc, E., & Ilgun, A. (2010). An investigation into the discourse of political marketing
communications in Turkey: The use of rhetorical figures in political party slogans. Journal of
Political Marketing, 9(3), 207-224.
167
Kolowich, L. (2017), How Often Should You Post on Facebook? [New Benchmark Data],
Hubspot,
https://blog.hubspot.com/marketing/facebook-post-frequency-
benchmarks#sm.00005x98lq12afhsyx41k4r79b6f9 [Accessed 12th August, 2019]
Kotler, P., & Levy, S. (1969). Broadening the Concept of Marketing. Journal of
Marketing, 33(1), 10–15.
Koziol, M. (2017). Distrustful nation: Australians lose faith in politics, media and business.
Sydney
Morning
Herald,
https://www.smh.com.au/politics/federal/distrustful-nation-
australians-lose-faith-in-politics-media-and-business-20170118-gttmpd.html [Accessed 4th
March, 2018]
Krippendorff, K. (1989). Content analysis. In Barnouw, E., Gerbner, G., Schramm, W., Worth,
T. L., & Gross, L. (Eds.), International encyclopedia of communication, 1, 403-407, New York,
NY: Oxford University Press.
Krueger, R. A., & Casey, M. A. (2014). Focus Groups: A practical guide for applied health
research. London: Sage Publications.
Kumar, A., Bezawada, R., Rishika, R., Janakiraman, R., & Kannan, P. K. (2016). From social
to sale: The effects of firm-generated content in social media on customer behavior. Journal of
Marketing, 80(1), 7-25.
Kumar, A., Paul, J., & Unnithan, A. (2020). “Masstige” marketing: A review, synthesis and
research agenda. Journal of Business Research, 113, 384–398.
Kümpel, A. S., Karnowski, V., & Keyling, T. (2015). News sharing in social media: A review
of current research on news sharing users, content, and networks. Social Media + Society, 1(2),
2056305115610141.
Kwak, D. H., Kim, Y. K., & Zimmerman, M. H. (2010). User-versus mainstream-mediagenerated content: Media source, message valence, and team identification and sport
consumers’ response. International Journal of Sport Communication, 3(4), 402-421.
Kwan, R., & Carlson, J. L. (2017). The ties that bind us: examining relationship marketing
orientation and its impact on firm performance in the information systems outsourcing services
sector. Journal of Strategic Marketing, 25(5-6), 495-510.
Landtsheer, C. L. D., De Vries, P., & Vertessen, D. (2008). Political impression management:
How metaphors, sound bites, appearance effectiveness, and personality traits can win
elections. Journal of Political Marketing, 7(3-4), 217-238.
168
Lariscy, R. W., Tinkham, S. F., & Sweetser, K. D. (2011). Kids these days: Examining
differences in political uses and gratifications, Internet political participation, political
information efficacy, and cynicism on the basis of age. American Behavioral Scientist, 55(6),
749-764.
Lee, L., Dabirian, A., McCarthy, I., & Kietzmann, J. (2020). Making sense of text: artificial
intelligence-enabled content analysis. European Journal of Marketing, 54(3), 615–644.
Lees-Marshment, J. (2019). Marketing Scholars and Political Marketing: the Pragmatic and
Principled Reasons for Why Marketing Academics Should Research the Use of Marketing in
the Political Arena. Customer Needs and Solutions, 6(3-4), 41–48.
Lees-Marshment, J. (2014). Political Marketing: Principles and Applications (2nd ed.).
Abingdon, Oxon: Routledge.
Leung, X. Y., & Bai, B. (2013). How motivation, opportunity, and ability impact travelers'
social media involvement and revisit intention. Journal of Travel & Tourism Marketing, 30(12), 58-77.
Lilleker, D. G., & Jackson, N. (2014). Brand management and relationship marketing in online
environments. Political marketing in the United States, 165-184. New York: Routledge.
Lin, H. (2017). How Political Candidates’ Use of Facebook Relates to the Election
Outcomes. International Journal of Market Research, 59(1), 77–96.
Lin, J., & Himelboim, I. (2019). Political Brand Communities as Social Network Clusters:
Winning and Trailing Candidates in the GOP 2016 Primary Elections. Journal of Political
Marketing, 18(1-2), 119–147.
Lin, T. M., Lu, K. Y., & Wu, J. J. (2012), The effects of visual information in eWOM
communication. Journal of Research in Interactive Marketing, 6(1), 7-26.
Liu, Y., & Shrum, L. J. (2002). What is interactivity and is it always such a good thing?
Implications of definition, person, and situation for the influence of interactivity on advertising
effectiveness. Journal of Advertising, 31(4), 53-64.
Liu, Y., Rui J. R., & Cui X. (2017). Are people willing to share their political opinions on
Facebook? Exploring roles of self-presentational concern in spiral of silence. Computers in
Human Behavior, 76, 294–302.
169
Loader, B. D, Vromen, A., & Xenos, M. A. (2014). The networked young citizen: social media,
political participation and civic engagement. Information, Communication & Society, 7(2),
143-150.
Lucarelli, A., Fuschillo, G., & Chytkova, Z. (2020). How cyber political brands emerge: a
socio-material analysis of the Italian Five Star movement and the Czech pirate party. European
Journal of Marketing, 55(4), 1130-1154.
Lüftenegger, E., Comuzzi, M., & Grefen, P. W. (2017). Designing a tool for service-dominant
strategies using action design research. Service Business, 11(1), 161-189.
Lumpkins, C. Y. (2010). Sacred symbols as a peripheral cue in health advertisements: An
assessment of using religion to appeal to African American women about breast cancer
screening. Journal of Media and Religion, 9(4), 181-201.
Macafee, T. (2013). Some of these things are not like the others: Examining motivations and
political predispositions among political Facebook activity. Computers in Human
Behavior, 29(6), 2766-2775.
Macafee, T., & De Simone, J. J. (2012). Killing the bill online? Pathways to young people's
protest
engagement
via
social
media. Cyberpsychology,
Behavior,
and
Social
Networking, 15(11), 579-584.
MacInnis, D. J., & Jaworski, B. J. (1989). Information processing from advertisements: Toward
an integrative framework. Journal of Marketing, 53(4), 1-23.
MacKenzie, S., Lutz, R., & Belch, G. (1986). A role of attitude toward the ad as a mediator of
advertising effectiveness: A test of competing explanations. Journal of Marketing Research,
23, 130–143.
MacPhail, C., Khoza, N., Abler, L., & Ranganathan, M. (2016). Process guidelines for
establishing intercoder reliability in qualitative studies. Qualitative Research, 16(2), 198-212.
Maecker, O., Barrot, C., & Becker, J. U. (2016). The effect of social media interactions on
customer relationship management. Business Research, 9(1), 133-155.
Maier, M., & Tenscher, J. (2009). “Struggling for Voters”: European Campaigning in
Germany. Journal of Political Marketing, 8(1), 20-34.
170
Malhotra, N., Sahadev, S., & Purani, K. (2017). Psychological contract violation and customer
intention to reuse online retailers: Exploring mediating and moderating mechanisms. Journal
of Business Research, 75, 17-28.
Manca, S., Altoè, G., Schultz, P. W., & Fornara, F. (2020). The Persuasive Route to Sustainable
Mobility: Elaboration Likelihood Model and Emotions predict Implicit Attitudes. Environment
and Behavior, 52(8), 830-860.
Marder, B., Marchant, C., Archer-Brown, C., Yau, A., & Colliander, J. (2018). Conspicuous
political brand interactions on social network sites. European Journal of Marketing, 52(3/4),
702–724.
Marland, A. (2003). Marketing political soap: A political marketing view of selling candidates
like soap, of electioneering as a ritual, and of electoral military analogies. Journal of Public
Affairs: An International Journal, 3(2), 103-115.
Marquart, F., Brosius, A., & de Vreese, C. (2019). United Feelings: The Mediating Role of
Emotions in Social Media Campaigns for EU Attitudes and Behavioral Intentions. Journal of
Political Marketing, 1–27.
Martin, A. (2012). Young People and Politics: Political Engagement in the Anglo-American
Democracies, Oxon: Routledge.
Mason, C., & Simmons, J. (2012). Are they being served? Linking consumer expectation,
evaluation and commitment. Journal of Services Marketing, 26(4), 227-237.
Mayring, P. (2000). Qualitative content analysis. Qualitative Social Research, 1(2), 20.
Mende, M., Bolton, R. N., & Bitner, M. J. (2013). Decoding customer–firm relationships: how
attachment styles help explain customers' preferences for closeness, repurchase intentions, and
changes in relationship breadth. Journal of Marketing Research, 50(1), 125-142.
Mertens, D. M. (2005). Research and evaluation in education and psychology: Integrating
diversity with quantitative, qualitative and mixed-methods (2nd ed). Thousand Oaks CA: Sage
Publications.
Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook.
Thousand Oaks, California: Sage Publications.
Milewicz, C., & Saxby, C. (2013). Leaders’ social media usage intentions for in-bound
customer communications. Management Research Review, 36(9), 849–867.
171
Miller, W. (2013). We Can’t All Be Obama: The Use of New Media in Modern Political
Campaigns. Journal of Political Marketing, 12(4), 326–347.
Mohr, J., & Spekman, R. (1994). Characteristics of partnership success: partnership attributes,
communication behavior, and conflict resolution techniques. Strategic Management Journal,
15(2), 135-152.
Morgan, D. L. (2007). Paradigms lost and pragmatism regained: Methodological implications
of combining qualitative and quantitative methods. Journal of Mixed Methods Research, 1(1),
48-76.
Montgomery, N. V., Raju, S., Desai, K. K., & Unnava, H. R. (2018). When good consumers
turn bad: Psychological contract breach in committed brand relationships. Journal of
Consumer Psychology, 28(3), 437-449.
Morgan, D. L. (1996). Focus groups as qualitative research (Vol. 16). Thousand Oaks,
California: Sage Publications.
Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship marketing.
Journal of Marketing, 58(3), 20-38.
Morris, J. D., Woo, C., & Singh, A. J. (2005). Elaboration likelihood model: A missing intrinsic
emotional implication. Journal of Targeting, Measurement and Analysis for Marketing, 14(1),
79-98.
Morrison, E. W., & Robinson, S. L. (1997). When employees feel betrayed: A model of how
psychological contract violation develops. Academy of Management Review, 22(1), 226-256.
Morse, J. M. (2015). Critical analysis of strategies for determining rigor in qualitative
inquiry. Qualitative Health Research, 25(9), 1212-1222.
Müller, J., & Christandl, F. (2019). Content is king–But who is the king of kings? The effect
of
content
marketing,
sponsored
content
&
user-generated
content
on
brand
responses. Computers in Human Behavior, 96, 46-55.
Muñoz, C., & Towner, T. (2017). The Image is the Message: Instagram Marketing and the
2016 Presidential Primary Season. Journal of Political Marketing, 16(3-4), 290–318.
Murphy, K. (2017). Major headache for Malcolm Turnbull as minor parties rise. The Guardian,
https://www.theguardian.com/australia-news/2017/feb/06/major-headache-for-malcolmturnbullas-minor-parties-rise
172
Nadin, S., & Cassell, C. (2007). New deal for old? Exploring the psychological contract in a
small firm environment. International Small Business Journal, 25(4), 417-443.
Newell, S. J., & Henderson, K. V. (1998). Super Bowl advertising: Field testing the importance
of advertisement frequency, length and placement on recall. Journal of Marketing
Communications, 4(4), 237-248.
Newman, B. (2016). Reinforcing Lessons for Business from the Marketing Revolution in U.S.
Presidential Politics: A Strategic Triad. Psychology & Marketing, 33(10), 781–795.
Newman, B. I. (2001). An assessment of the 2000 US presidential election: A set of political
marketing guidelines. Journal of Public Affairs, 1(3), 210-216.
Newman, N., Fletcher, R., Kalogeropoulos, A., Levy, D. A., & Nielsen, R. K. (2017), Digital
News Report 2017, Reuters Institute,
https://reutersinstitute.politics.ox.ac.uk/sites/default/files/Digital%20News%20Report%2020
17%20web_0.pdf [Accessed 17 July 2019]
Ngai, E., Tao, S., & Moon, K. (2015). Social media research: Theories, constructs, and
conceptual frameworks. International Journal of Information Management, 35(1), 33–44.
Nguyen Hau, L., & Viet Ngo, L. (2012). Relationship marketing in Vietnam: An empirical
study. Asia Pacific Journal of Marketing and Logistics, 24(2), 222-235.
Noble, H., & Smith, J. (2015). Issues of validity and reliability in qualitative research.
Evidence-Based Nursing, 18(2), 34–35.
Novak, T. P. (1995). MANOVAMAP: Graphical representation of MANOVA in marketing
research. Journal of Marketing Research, 32(3), 357-374.
O'Shaughnessy, N. (2003). The symbolic state: A British experience. Journal of Public Affairs:
An International Journal, 3(4), 297-312.
O'Shaughnessy, N. J. (2004). Politics and propaganda: Weapons of mass seduction.
Manchester, UK: Manchester University Press.
O'Shaughnessy, N. J., & Henneberg, S. C. (2002). The idea of political marketing. Westport,
CT: Greenwood Publishing Group.
O’Cass, A. (2009). A resource-based view of the political party and value creation for the votercitizen: an integrated framework for political marketing. Marketing Theory, 9(2), 189–208.
173
Ok, C., Shanklin, C. W., & Back, K. J. (2008). Generalizing survey results from student
samples: implications from service recovery research. Journal of Quality Assurance in
Hospitality & Tourism, 8(4), 1-23.
Onwuegbuzie, A. J., Dickinson, W. B., Leech, N. L., & Zoran, A. G. (2009). A qualitative
framework for collecting and analyzing data in focus group research. International Journal of
Qualitative Methods, 8(3), 1-21.
Orben, A. C., & Dunbar, R. I. (2017). Social media and relationship development: The effect
of valence and intimacy of posts. Computers in Human Behavior, 73, 489-498.
Ordenes, V. F., Grewal, D., Ludwig, S., Ruyter, K. D., Mahr, D., & Wetzels, M. (2019). Cutting
through content clutter: How speech and image acts drive consumer sharing of social media
brand messages. Journal of Consumer Research, 45(5), 988-1012.
Ormrod, R. P., Henneberg, S. C., & O'Shaughnessy, N. J. (2013). Political marketing: Theory
and concepts. London: SAGE Publications.
Page, J., & Duffy, M. (2018). What Does Credibility Look like? Tweets and Walls in U.S.
Presidential Candidates’ Visual Storytelling. Journal of Political Marketing, 17(1), 3–31.
Pallant, J. (2016), SPSS Survival Guide Manual (6th edition), NSW, Australia: Allen and
Unwin.
Palmatier, R. W., Dant, R. P., Grewal, D., & Evans, K. R. (2006). Factors influencing the
effectiveness of relationship marketing: A meta-analysis. Journal of Marketing, 70(4), 136153.
Palmatier, R. W. (2008). Relationship marketing. Cambridge, MA: Marketing Science
Institute.
Palomba, C. A., & Banta. T. W. (1999). Assessment Essentials: Planning, Implementing, and
Improving Assessment in Higher Education. Higher and Adult Education Series. San
Francisco: Jossey-Bass, Inc., Publishers.
Papagiannidis, S., Coursaris, C. K., & Bourlakis, M. (2012). Do websites influence the nature
of voting intentions? The case of two national elections in Greece. Computers in Human
Behavior, 28(2), 300-307.
Park, C. S. (2013). Does Twitter motivate involvement in politics? Tweeting, opinion
leadership, and political engagement. Computers in Human Behavior, 29(4), 1641-1648.
174
Parmelee, J. H., & Roman. N. (2019). Insta-Politicos: Motivations for Following Political
Leaders on Instagram. Social Media+Society, 5(2), 1-12.
Parsons, M., & Rowling, M. (2018). Social Media and the Paradox of Fear: An Exploratory
Study of Political Relationship Marketing Within South Wales. Journal of Political
Marketing, 17(3), 235–257.
Paul, J., & Criado, A. (2020). The art of writing literature review: What do we know and what
do we need to know? International Business Review, 29(4), 101717.
Paul, J., Lim, W. M., O’Cass, A., Hao, A. W., & Bresciani, S. (2021). Scientific procedures
and rationales for systematic literature reviews (SPAR‐4‐SLR). International Journal of
Consumer Studies, 00, 1-16.
Paul, N., & Sui, M. (2019). I Can Feel What You Feel: Emotion Exchanges in Twitter
Conversations between Candidates and the Public. Journal of Political Marketing, 1–21.
Pavlou, P. A., & Gefen, D. (2005). Psychological contract violation in online marketplaces:
Antecedents, consequences, and moderating role. Information Systems Research, 16(4), 372399.
Payne, A., & Frow, P. (2017). Relationship marketing: looking backwards towards the future.
Journal of Services Marketing, 31(1), 11-15.
Pelletier, J. A., & Sharma, N. (2015). Investigating the Impact of Student Samples on Data
Quality and Research Outcomes. In: Kubacki, K. (eds) Ideas in Marketing: Finding the New
and Polishing the Old. Developments in Marketing Science: Proceedings of the Academy of
Marketing Science. Springer, Cham.
Peng, L. Y., & Wang, Q. (2006). Impact of relationship marketing tactics (RMTs) on switchers
and stayers in a competitive service industry. Journal of Marketing Management, 22(1-2), 2559.
Penney, J. (2016). Motivations for participating in “viral politics”: A qualitative case study of
Twitter users and the 2012 US presidential election. Convergence, 22(1), 71–87.
Penney, J. (2017). Social Media and Citizen Participation in “Official” and “Unofficial”
Electoral Promotion: A Structural Analysis of the 2016 Bernie Sanders Digital
Campaign. Journal of Communication, 67(3), 402–423.
175
Perannagari, K., & Chakrabarti, S. (2020). Analysis of the literature on political marketing
using a bibliometric approach. Journal of Public Affairs, 20(1), 1-13.
Peres, R., Talwar, S., Alter, L., Elhanan, M., & Friedmann, Y. (2020). Narrowband Influencers
and Global Icons: Universality and Media Compatibility in the Communication Patterns of
Political Leaders Worldwide. Journal of International Marketing, 28(1), 48–65.
Petty, R. E., & Briñol, P. (2015). Emotion and persuasion: Cognitive and meta-cognitive
processes impact attitudes. Cognition and Emotion, 29(1), 1-26.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion.
In Communication and persuasion, (p. 1-24). New York, NY: Springer.
PEW Research Centre. (2018). Party Identification Trends, 1992-2017, https://www.peoplepress.org/2018/03/20/party-identification-trends-1992-2017/party [Accessed 9th September
2019]
Pew
Research
Centre.
(2019a).
Social
media
fact
sheet.
https://www.pewresearch.org/internet/fact-sheet/social-media/ [Accessed 12th May, 2020]
Pew Research Centre. (2019b). Partisan Antipathy: More Intense, More Personal.
https://www.pewresearch.org/politics/2019/10/10/partisan-antipathy-more-intense-morepersonal/ [Accessed 13th Oct, 2020]
Pich, C., Harvey, J., Armannsdottir, G., Poorrezaei, M., Branco-Illodo, I., & Kincaid, A.
(2018). Marketing Brexit: An exploratory study of young voter engagement in relation to the
EU referendum. International Journal of Market Research, 60(6), 589–610.
Pollitt, E. (2020). The 82 million dollar election: 2019 ad spend roundup. B&T,
https://www.bandt.com.au/82-million-dollar-election-2019-ad-spend-roundup/ [Accessed 7th
May, 2021]
Prahalad, C. K., & Ramaswamy, V. (2004). The future of competition: Co-creating unique
value with customers. Cambridge, MA: Harvard Business Press.
Quintelier, E. (2007). Differences in political participation between young and old people.
Contemporary Politics, 13(2), 165-180.
Quinton, S., & Harridge-March, S. (2010). Relationships in online communities: the potential
for marketers. Journal of Research in Interactive Marketing, 4(1), 59-73.
176
Rabiee, F. (2004). Focus-group interview and data analysis. Proceedings of the Nutrition
Society, 63(4), 655-660.
Radio
Info.
(2020).
The
winner
of
the
US
Election.
https://www.radioinfo.com.au/news/winner-us-election [Accessed 6th November, 2020]
Rana, J., & Paul, J. (2017). Consumer behavior and purchase intention for organic food: A
review and research agenda. Journal of Retailing and Consumer Services, 38, 157–165.
Reed, J., & Payton. V. R. (1997). Focus groups: issues of analysis and interpretation. Journal
of Advanced Nursing, 26(4), 765-771.
Reinecke, L., & Trepte, S. (2014). Authenticity and well-being on social network sites: A twowave longitudinal study on the effects of online authenticity and the positivity bias in SNS
communication. Computers in Human Behavior, 30, 95-102.
Rhoads, J. C., Thomas, D. B., & McKeown, B. (2016). Social media in a subjective-science
mode: The “Facebook likes” study reconfigured with self-reference. The Journal of Social
Media in Society, 5(1), 111-136.
Rim, H., & Song, D. (2016). “How negative becomes less negative”: Understanding the effects
of comment valence and response sidedness in social media. Journal of Communication, 66(3),
475-495.
Robey, D., Farrow, D. L., & Franz, C. R. (1989). Group process and conflict in system
development. Management Science, 35(10), 1172-1191.
Rockler, H. (2013). Why American Voters Decide to Vote for Third Parties in Presidential
Elections. Syracuse University Honors Program Capstone Projects. Paper 52.
Rooney, T., Krolikowska, E., & Bruce, H. L. (2021). Rethinking relationship marketing as
consumer led and technology driven: propositions for research and practice. Journal of
Relationship Marketing, 20(1), 42-61.
Rosado-Serrano, A., Paul, J., & Dikova, D. (2018). International franchising: A literature
review and research agenda. Journal of Business Research, 85, 238-257.
Rousseau, D. M. 1995. Psychological contracts in organizations: Understanding written and
unwritten agreements. Thousand Oaks, California: Sage Publications.
177
Rousseau, D. M., & Tijoriwala, S. A. (1998). Assessing psychological contracts: Issues,
alternatives and measures. Journal of Organizational Behavior: The International Journal of
Industrial, Occupational and Organizational Psychology and Behavior, 19(S1), 679-695.
Rowe, I. (2015), Civility 2.0: A comparative analysis of incivility in online political
discussion. Information, Communication & Society, 18(2), 121-138.
Rowley, J., & Keegan, B. (2020). An overview of systematic literature reviews in social media
marketing. Journal of Information Science, 46(6), 725–738.
Roy, S. K., & Eshghi, A. (2013). Does relationship quality matter in service
relationships?. Journal of Strategic Marketing, 21(5), 443-458.
Ruggiero, T. E. (2000). Uses and gratifications theory in the 21st century. Mass
Communication & Society, 3(1), 3-37.
Ryoo, J., & Bendle, N. (2017). Understanding the Social Media Strategies of U.S. Primary
Candidates. Journal of Political Marketing, 16(3-4), 244–266.
Sabate, F., Berbegal-Mirabent, J., Cañabate, A., & Lebherz, P. R. (2014). Factors influencing
popularity of branded content in Facebook fan pages. European Management Journal, 32(6),
1001-1011.
Safiullah, M., Pathak, P., Singh, S., & Anshul, A. (2017). Social media as an upcoming tool
for political marketing effectiveness. Asia Pacific Management Review, 22(1), 10–15.
Salisbury, C. (2019). Minor parties perform well in federal election and reconfirm the power
of preference deals. The Conversation, https://theconversation.com/minor-parties-performwell-in-federal-election-and-reconfirm-the-power-of-preference-deals-117192 [Accessed 1st
March, 2020]
Samuel-Azran, T., Yarchi, M., & Wolfsfeld, G. (2015). Aristotelian rhetoric and Facebook
success in Israel’s 2013 election campaign. Online Information Review, 39(2), 149–162.
Sanchez-Franco, M. J., & Rondan-Cataluña, F. J. (2010). Virtual travel communities and
customer loyalty: Customer purchase involvement and web site design. Electronic Commerce
Research and Applications, 9(2), 171-182.
Sanghvi, M., & Hodges, N. (2015). Marketing the female politician: an exploration of gender
and appearance. Journal of Marketing Management, 31(15-16), 1676-1694.
178
Sashi,
C.
(2012).
Customer
engagement,
buyer‐seller
relationships,
and
social
media. Management Decision, 50(2), 253–272.
Sashi, C. M., Brynildsen, G., & Bilgihan, A. (2019). Social media, customer engagement and
advocacy. International Journal of Contemporary Hospitality Management. 31(3), 1247-1272.
Schivinski, B., & Dabrowski, D. (2016). The effect of social media communication on
consumer perceptions of brands. Journal of Marketing Communications, 22(2), 189-214.
Schneiker, A. (2019). Telling the Story of the Superhero and the Anti-Politician as President:
Donald Trump’s Branding on Twitter. Political Studies Review, 17(3), 210–223.
Sehl, K. (2019). 27 Top Facebook Demographics that Matter to Social Media Marketers.
Hootsuite. https://blog.hootsuite.com/facebook-demographics/ [Accessed 12th May, 2020]
Serazio, M. (2017). Branding politics: Emotion, authenticity, and the marketing culture of
American political communication. Journal of Consumer Culture, 17(2), 225-241.
Sheldon, P., & Bryant, K. (2016). Instagram: Motives for its use and relationship to narcissism
and contextual age. Computers in Human Behavior, 58, 89-97.
Sherman, E., Schiffman, L., & Thelen, S. (2012). Young Voters’ Trust of Information and
Media Sources: The 2008 U.S. Presidential Election. Journal of Political Marketing, 11(4),
246–264.
Sheth,
J.
(2017).
Revitalizing
relationship
marketing. The
Journal
of
Services
Marketing, 31(1), 6–10.
Shmargad, Y. (2018). Twitter Influencers in the 2016 US Congressional Races. Journal of
Political Marketing, 1–18.
Shu, M., & Scott, N. (2014). Influence of social media on Chinese students’ choice of an
overseas study destination: An information adoption model perspective. Journal of Travel &
Tourism Marketing, 31(2), 286-302.
Sin, L. Y., Alan, C. B., Yau, O. H., Chow, R. P., Lee, J. S., & Lau, L. B. (2005). Relationship
marketing orientation: scale development and cross-cultural validation. Journal of Business
Research, 58(2), 185-194.
Sin, L. Y., Tse, A. C., Chan, H., Heung, V. C., & Yim, F. H. (2006). The effects of relationship
marketing orientation on business performance in the hotel industry. Journal of Hospitality &
Tourism Research, 30(4), 407-426.
179
Sin, L. Y., Tse, A. C., Yau, O. H., Lee, J. S., & Chow, R. (2002). The effect of relationship
marketing orientation on business performance in a service-oriented economy. Journal of
Services Marketing, 16(7), 656-676.
Skowronski, J. J., & Carlston, D. E. (1989). Negativity and extremity biases in impression
formation: A review of explanations. Psychological Bulletin, 105(1), 131-142.
Slothuus, R., & De Vreese, C. H. (2010). Political parties, motivated reasoning, and issue
framing effects. The Journal of Politics, 72(3), 630-645.
Small, T. (2008). The Facebook effect? On-line campaigning in the 2008 Canadian and US
elections. Policy Options, November, 84–87.
Small, T. (2012). Are we friends yet? Online relationship marketing by political parties. In A.
Marland, T. Giasson, & J. Lees-Marshment (Eds.), Political marketing in Canada, (p. 193208). Vancouver: UBC.
Snyder, H., Witell, L., Gustafsson, A., Fombelle, P., & Kristensson, P. (2016). Identifying
categories of service innovation: A review and synthesis of the literature. Journal of Business
Research, 69(7), 2401–2408.
Sofaer, S. (1999). Qualitative methods: what are they and why use them? Health Services
Research, 34(5 part 2), 1101-1118.
Steinhoff, L., Arli, D., Weaven, S., & Kozlenkova, I. (2019). Online relationship
marketing. Journal of the Academy of Marketing Science, 47(3), 369–393.
Steinhorst, R. K., & Williams, R. E. (1985). Discrimination of groundwater sources using
cluster analysis, MANOVA, canonical analysis and discriminant analysis. Water Resources
Research, 21(8), 1149-1156.
Sterrett, D., Malato, D., Benz, J., Kantor, L., Tompson, T., Rosenstiel, T., ... & Loker, K.
(2019). Who Shared It?: Deciding What News to Trust on Social Media. Digital
Journalism, 7(6), 783-801.
Suh, M., Greene, H., Israilov, B., & Rho, T. (2015). The impact of customer education on
customer loyalty through service quality. Services Marketing Quarterly, 36(3), 261-280.
Sundar, S. S., & Limperos, A. M. (2013). Uses and grats 2.0: New gratifications for new
media. Journal of Broadcasting & Electronic Media, 57(4), 504-525.
180
Swani, K., Milne, G. R., Brown, B. P., Assaf, A. G., & Donthu, N. (2017). What messages to
post? Evaluating the popularity of social media communications in business versus consumer
markets. Industrial Marketing Management, 62, 77-87.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (Vol. 5). Boston, MA:
Pearson.
Tabachnick, B., Fidell, L., & Ullman, J. (2019). Using multivariate statistics (Seventh edition).
Boston, MA: Pearson.
Tafesse, W. (2015). Content strategies and audience response on Facebook brand
pages. Marketing Intelligence & Planning, 33(6), 927–943.
Tafesse, W., & Wien, A. (2017). A framework for categorizing social media posts. Cogent
Business & Management, 4(1), 1284390.
Tafesse, W., & Wien, A. (2018). Using message strategy to drive consumer behavioral
engagement on social media. Journal of Consumer Marketing, 35(3), 241–253.
Tellis, G. J., MacInnis, D. J., Tirunillai, S., & Zhang, Y. (2019). What drives virality (sharing)
of online digital content? The critical role of information, emotion, and brand
prominence. Journal of Marketing, 83(4), 1-20.
Temple, M. (2013). The Media and the Message. Journal of Political Marketing, 12(2-3), 147–
165.
Teng, S., Khong, K. W., Chong, A. Y. L., & Lin, B. (2017). Persuasive electronic word-ofmouth messages in social media. Journal of Computer Information Systems, 57(1), 76-88.
Thaichon, P., Liyanaarachchi, G., Quach, S., Weaven, S., & Bu, Y. (2019). Online relationship
marketing: evolution and theoretical insights into online relationship marketing. Marketing
Intelligence & Planning. 38(6), 676-698.
Tian, M., Deng, P., Zhang, Y., & Salmador, M. (2018). How does culture influence innovation?
A systematic literature review. Management Decision, 56(5), 1088–1107.
Tillet, A. (2019). 'Wake-up call': 75pc of voters say politicians can't be trusted. The Australian
Financial
Review,
https://www.afr.com/politics/federal/wake-up-call-75pc-of-voters-say-
politicians-can-t-be-trusted-20191208-p53hwg [Accessed 1st March, 2020]
Timmermans, E., & De Caluwé, E. (2017). Development and validation of the Tinder Motives
Scale (TMS). Computers in Human Behavior, 70, 341-350.
181
Towner, T., & Dulio, D. (2011). The Web 2.0 Election: Does the Online Medium
Matter? Journal of Political Marketing, 10(1-2), 165–188.
Towner, T., & Dulio, D. (2012). New Media and Political Marketing in the United States: 2012
and Beyond. Journal of Political Marketing, 11(1-2), 95–119.
Towner, T., & Muñoz, C. (2018). Baby Boom or Bust? the New Media Effect on Political
Participation. Journal of Political Marketing, 17(1), 32–61.
Tranfield, D., Denyer, D., & Smart, P. (2003). Towards a Methodology for Developing
Evidence-Informed Management Knowledge by Means of Systematic Review. British Journal
of Management, 14(3), 207–222.
Tse, A. C., Sin, L. Y., Yau, O. H., Lee, J. S., & Chow, R. (2004). A firm's role in the
marketplace and the relative importance of market orientation and relationship marketing
orientation. European Journal of Marketing, 38(9/10), 1158-1172.
Tseng, S., & Wang, C. (2016). Perceived risk influence on dualroute information adoption
processes on travel websites. Journal of Business Research, 69(6), 2289–2296.
Tsvetkova, N., Rushchin, D., Shiryaev, B., Yarygin, G., & Tsvetkov, I. (2020). Sprawling in
Cyberspace:
Barack
Obama’s
Legacy
in
Public
Diplomacy
and
Strategic
Communication. Journal of Political Marketing, 1–13.
Turcotte, J., York, C., Irving, J., Scholl, R. M., & Pingree, R. J. (2015). News recommendations
from social media opinion leaders: Effects on media trust and information seeking. Journal of
Computer-Mediated Communication, 20(5), 520-535.
Turnley, W. H., Bolino, M. C., Lester, S. W., & Bloodgood, J. M. (2004). The effects of
psychological contract breach on union commitment. Journal of Occupational and
Organizational Psychology, 77(3), 421-428.
Turri, A. M., Smith, K. H., & Kemp, E. (2013). Developing affective brand commitment
through social media. Journal of Electronic Commerce Research, 14(3), 201-214.
Tuten, T., & Solomon, M. (2013). Social media marketing. Upper Saddle River, NJ: Pearson.
Tuten, T.L., & Solomon, M.R. (2017). Social Media Marketing. Sage Publications.
Twenge, J. M., Campbell, W. K., & Freeman, E. C. (2012). Generational differences in young
adults' life goals, concern for others, and civic orientation, 1966–2009. Journal of Personality
and Social Psychology, 102(5), 1045.
182
Universities Australia. (2019). Data Snapshot. https://www.universitiesaustralia.edu.au/wpcontent/uploads/2019/06/Data-snapshot-2019-FINAL.pdf
Varca, P. E. (2009). Emotional empathy and front line employees: does it make sense to care
about the customer? Journal of Services Marketing, 23(1), 51-56.
Vargo, S. L., & Lusch, R. F. (2016). Institutions and axioms: an extension and update of
service-dominant logic. Journal of the Academy of Marketing Science, 44(1), 5-23.
Veenstra, A., Lyons, B., & Degim Flannagan, İ. (2017). Intraparty Hostility: Social Identity,
Subidentity, and the Hostile Media Effect in a Contested Primary. Journal of Political
Marketing, 16(3-4), 365–385.
Verma, V., Sharma, D., & Sheth, J. (2016). Does relationship marketing matter in online
retailing? A meta-analytic approach. Journal of the Academy of Marketing Science, 44(2), 206217.
Vesnic-Alujevic, L., & Van Bauwel, S. (2014). YouTube: A Political Advertising Tool? A
Case Study of the Use of YouTube in the Campaign for the European Parliament Elections
2009. Journal of Political Marketing, 13(3), 195–212.
Voorveld, H. (2019). Brand Communication in Social Media: A Research Agenda. Journal of
Advertising, 48(1), 14–26.
Vrontis, D., Makrides, A., Christofi, M., & Thrassou, A. (2021). Social media influencer
marketing:
A
systematic
review,
integrative
framework
and
future
research
agenda. International Journal of Consumer Studies, 1-28.
Walker, L., Baines, P., Dimitriu, R., & Macdonald, E. (2017). Antecedents of Retweeting in a
(Political) Marketing Context. Psychology & Marketing, 34(3), 275–293.
Wallsten, K. (2011). Many Sources, One Message: Political Blog Links to Online Videos
During the 2008 Campaign. Journal of Political Marketing, 10(1-2), 88–114.
Wang, C. Y., Lee, H. C., Wu, L. W., & Liu, C. C. (2017). Quality dimensions in online
communities influence purchase intentions, Management Decision, 55(9), 1984-1998.
Warren, A. M., Sulaiman, A., & Jaafar, N. I. (2014). Social media effects on fostering online
civic engagement and building citizen trust and trust in institutions. Government Information
Quarterly, 31(2), 291-301.
183
Whiting, A., & Williams, D. (2013). Why people use social media: a uses and gratifications
approach. Qualitative Market Research: An International Journal, 16(4), 362-369.
Wieland, H., Polese, F., Vargo, S. L., & Lusch, R. F. (2012). Toward a service (eco) systems
perspective on value creation. International Journal of Service Science, Management,
Engineering, and Technology, 3(3), 12-25.
Wilden, R., & Gudergan, S. P. (2015). The impact of dynamic capabilities on operational
marketing and technological capabilities: investigating the role of environmental
turbulence. Journal of the Academy of Marketing Science, 43(2), 181-199.
Williams, C. B. (2017). Introduction: Social Media, Political Marketing and the 2016 U.S.
Election, Journal of Political Marketing, 16(3-4), 207-211,
Williams, C. B., & Gulati, G. J. (2014). Relationship Marketing in Social Media Practice. In
Lees-Marshment, J., Conley, B., & Cosgrove, K. (eds), Political Marketing in the United
States, (p. 185 - 201), New York, NY: Routledge.
Winchester, T., Hall, J., & Binney, W. (2014). Young adult voting decision-making: Studying
the effect of usage from a consumer behaviour perspective. Australasian Marketing Journal,
22(2): 144-154.
Winchester, T., Hall, J., & Binney, W. (2015). Evaluating young adult voter decision-making
involvement within a compulsory political system. Qualitative Market Research: An
International Journal, 18(3), 252-277.
Winther-Nielsen, S. (2017). On Political Brands: A Systematic Review of the
Literature. Journal of Political Marketing, 16(2), 118–146.
Wood, D., & Daley, J. (2018). A crisis of trust: The rise of protest politics in Australia. Carlton,
VIC: Grattan Institute.
Wood, N. T., & Herbst, K. C. (2007). Political star power and political parties: Does celebrity
endorsement win first-time votes?. Journal of Political Marketing, 6(2-3), 141-158.
Wright, S., & Koslowski, M. (2019). Voter turnout at record low after young people disengage.
Sydney Morning Herald, www.smh.com.au/federal-election-2019/voter-turnout-at-recordlow-after-young-people-disengage-20190530-p51sol.html.
184
Xiang, D., Zhang, L., Tao, Q., Wang, Y., & Ma, S. (2019). Informational or emotional appeals
in crowdfunding message strategy: an empirical investigation of backers’ support
decisions. Journal of the Academy of Marketing Science, 47(6), 1046-1063.
Yau, O. H., McFetridge, P. R., Chow, R. P., Lee, J. S., Sin, L. Y., & Tse, A. C. (2000). Is
relationship marketing for everyone?. European Journal of Marketing, 34(9/10), 1111-1127.
YLI-KAITALA, K. (2014). Revolution 2.0 in Egypt: Pushing for Change, Foreign Influences
on a Popular Revolt. Journal of Political Marketing, 13(1-2), 127–151.
Yoganathan, D., Jebarajakirthy, C., & Thaichon, P. (2015). The influence of relationship
marketing orientation on brand equity in banks. Journal of Retailing and Consumer
Services, 26, 14-22.
Zadeh, A. H., & Sharda, R. (2014). Modeling brand post popularity dynamics in online social
networks. Decision Support Systems, 65, 59-68.
Zell, A. L., & Moeller, L. (2018). Are you happy for me… on Facebook? The potential
importance of “likes” and comments. Computers in Human Behavior, 78, 26-33.
Zhang, J., & Bloemer, J. M. (2008). The impact of value congruence on consumer-service
brand relationships. Journal of Service Research, 11(2), 161-178.
Zhou, Z., Zhang, Q., Su, C., & Zhou, N. (2012). How do brand communities generate brand
relationships? Intermediate mechanisms. Journal of Business Research, 65(7), 890-895.
Ziegele, M., & Reinecke, L. (2017). No place for negative emotions? The effects of message
valence, communication channel, and social distance on users’ willingness to respond to SNS
status updates. Computers in Human Behavior, 75, 704-71
185
Appendix A. Introduction of participants
Pseudo
nym
Tim
ID No.
Introduction
P1, F1
Current
Enrolment
Undergrad, 2nd year
George
P2, F1
Undergrad, 2nd year
I have an interest in politics. I kind of want to be a politician so I
follow politicians on social media.
Alex
P3, F1
Undergrad, 2nd year
My view on politics is not that good because I was in the defense force
for a little while, and we got screwed over by politicians twenty-four
seven. So, I apologize in advance.
Brian
P4, F1
Undergrad, 1st year
I am pretty interested in politics. I’d say not so much Australian but
more of what’s happening around the world. Just keep track of
Australian politics.
Patrick
P5, F1
Undergrad, 3rd year
I run a youth-led think tank that analyses economic affairs and
student relations. I have done a lot of work with the US Consul
General and some with Canadians. My grandfather and all my mom’s
family are very big in politics.
Bill
P6, F1
Undergrad, Hons
year
I am more interested in the marketing side of it. In politics, in general,
I am not like super-engaged, just enough to know what is going on.
Alisa
P7, F1
Undergrad, Hons
year
I am probably not as involved in politics as I have been in the past,
but yes, interested in general.
Lucy
P8, F1
Undergrad, 3rd year
I come from a pretty different side of political background. My family
migrated from China. I just want to know more about Australian
politics. Not much involved but interested.
Mila
P1, F2
Graduate, 1st year
Talking about social media and the political aspect, I am really active
in following politicians on Instagram. Honestly, I follow both parties,
the parties that I support or the parties that I don’t support. Just to
see that how the news is conflicted with each other.
Samuel
P2, F2
Undergrad, 1st year
I am an accounting and marketing major. I do follow a few
politicians, but I specifically only follow one person on Instagram,
just to see how he acts. I have more interest in if he uses social media
as a platform to push his ideas or does he use that platform to engage
with his followers.
Shasha
P3, F2
Graduate, 1st year
I am doing my degree in strategic communications. I follow both
sides. I lean a bit towards the left, but I do follow a few right-wing
candidates, just to see how controversial they can get. I think I follow
the lefts basically just for information and see how things are going
on.
Sara
P4, F2
Undergrad, 1st year
At the moment, I am following the Australian Christian Lobby
because of the upcoming plebiscite. I find it really interesting because
the mainstream media like tv channels are presenting, a very onesided perspective, I think. So, seeing the other side is really
interesting. I am more conservative, but I am a liberal supporter. It’s
very interesting to see different sides.
I was involved in one election campaign, and several grassroots
campaigns organized by smaller groups. I went to Coalition’s offices
and met with them to discuss certain topics. I have been involved in
local initiatives by politicians in the community. I just have interest in
politics. Hope to go into it, but that’s a long shot.
186
Daniel
P5, F2
Undergrad, 3rd year
I follow a whole range of different politicians, like a lot of state
politicians and a lot of national politicians. I am sort of more
progressive in my values, so a lot of them do come from the Labor
party. I also follow Julia Bishop because you get that good insight
into her foreign ministry work, and then obviously like Hillary
Clinton, Obama, and others.
Megan
P6, F2
Undergrad, 2nd year
I am a second-year commerce student. On social media, I follow a lot
of politicians and I am more progressive in my views, but I don’t like
to be too partisan in my views, but I am, I guess, generally more leftleaning.
Shauna
P1, F3
Undergrad, 1st year
I follow a few Greens, some of whom have resigned now, and I also
follow a couple of Labor politicians and my local members who are
both Liberals, both like state and federal. And I pay attention to what
politicians are saying. I am a bit left-leaning.
Jim
P2, F3
Undergrad, 2nd year
I wouldn’t say I am very politically active, but I do follow politicians
on social media. I follow my federal member Kim White and a few
other politicians like Julia bishop.
Alicia
P3, F3
Undergrad, 2nd year
I am a Political Science and Marketing student. Bit of a political
tragic, member of the Young Liberals and all that. I follow a pretty
wide range of politicians on Facebook from my state and federal
member, who are both Labor, through to a huge array from across
Australia.
Geoff
P4, F3
Undergrad, 3rd year
I am a third-year engineering management student; I tend to have
centrist leanings. I follow Australian and American politicians, as
well as Great Britain and Eurozone.
Mary
P5, F3
Undergrad, Hons
year
I am not engaged in politics. I do follow Australian politicians, but
not actively. I am more interested in US, European, and international
politics.
Annie
P6, F3
Undergrad, 3rd year
I am a psychology student. I think I am a bit more progressive as in I
just go for whatever is best for people. There was a time when I really
looked into politics and followed Andrea Mitchell and Roger Cook,
Minister for Mental Health, closely.
Lilly
P7, F3
Undergrad, 2nd year
I am studying politics. I follow a lot of politicians and parties on my
Instagram and Facebook, but my favorite is Malcolm Turnbull on
Snapchat because it is actually quite funny.
Chris
P8, F3
Undergrad, 3rd year
I am an engineering student. I tend to follow politics quite closely. I
follow Australian politics but also take interest in US politics.
Arnold
P9, F3
Undergrad, 1st year
In terms of political views, I am a liberal and very much into
libertarian thought and that sort of thing.
Tom
P10, F3
Undergrad, 1st year
I am a Commerce student. In terms of following, I follow a lot of
Australian politicians and obviously US politicians like Donald
Trump and Hillary Clinton, and Republican and Democratic parties.
187
Download