CONTENT AND COMMUNITY:
DECODING ENGAGEMENT BY EXPLORING EMPIRICAL LINKS WITH
SOCIAL MEDIA ENGAGEMENT, BRAND EQUITY, PURCHASE INTENT, AND
ENGAGEMENT INTENT
By
Brigitte Arianna McKay

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Information and Media Studies – Doctor of Philosophy
2017

ABSTRACT
CONTENT AND COMMUNITY:
DECODING ENGAGEMENT BY EXPLORING EMPIRICAL LINKS WITH
SOCIAL MEDIA ENGAGEMENT, BRAND EQUITY, PURCHASE INTENT, AND
ENGAGEMENT INTENT
By
Brigitte Arianna McKay
This dissertation looks at the value proposition of social media to firms. Specifically, it
looks at what factors account for the empirical link between brand equity, purchase intention,
and brand social media engagement intention. This study also looks into brand equity
antecedents, specifically cognitive absorption and cognitive flow. In addition, this study looks at
the effects of community engagement metrics on social media posts, to see to what extent brand
equity, purchase intent, and/or brand social media engagement intent are the result of actual
social media content or previous/other consumers engaging with the post.
Using a simulated Facebook environment and survey format, 858 Mechanical Turk
participants were randomly assigned to one of three conditions, no community metrics shown,
true community engagement metrics shown, and switched community engagement metrics
shown (e.g. posts with established high engagement metrics were shown with low engagement
metrics and vice versa). They were then randomly assigned once more into a high or low
engagement post group. Participants were then shown three Facebook posts for McDonald’s,
and three posts for Delta Airlines, with engagement scores and presence based on their random
assignments. Brands were selected to provide the maximum variety of purchase decision
involvement, and Brand Awareness, Promotional, and Engagement post categories because they
corresponded with the Buying Behavior Model (Coursaris, Van Osch, Balogh, 2013; Wind,
1978). Following each post, participants were presented a survey about their social media
engagement intentions, purchase intentions, brand equity, as well as cognitive absorption and
flow experiences when interacting with the post.

Based on these findings, this dissertation offers the following contributions. First, it
empirically tests the relationship between social media-based brand engagement and purchase
intention and therewith attempts to expose whether or not investments by companies in social
media marketing communications are associated with a potential return on investment. Second,
this study serves as a starting point for understanding antecedents of brand equity, and
establishes cognitive flow as a highly likely predictor. Third, this study tests the new construct of
brand social media engagement intention on a public sample. Fourth, this study draws on actual
brand Facebook Page Posts that are classified according to their messaging content and
objectively evaluated in terms of their impact (i.e. the evoked level of Likes, Comments, and/or
Shares) before exposing them to participants to test the impact they may have on consumers’
perceptions and behavioral intentions towards the brand in a natural, simulated exposure
environment of Facebook. Brand communities are then actively tested to observe consumer
engagement effects on a live, simulated platform, to see if consumers interact with posts based
on the content, or the presence of community engagement metrics. Such results could demystify
and assist in some of brand investments into practices such as purchasing engagement. If it is
purely content that causes people to engage with brand posts, then companies can increase their
organic value without investing in paid resources, thus increasing their ROI. Finally, active
interaction was not established as an equivocal reflection of self-report metrics for engagement
intention. Further understanding of the Facebook environment is still needed.

Copyright by
BRIGITTE ARIANNA MCKAY
2017

For Sam.
Who put up with everything,
And never let me quit.
And taught me love.

v

ACKNOWLEDGEMENTS

A Dissertation Completion Fellowship from the College of Communication Arts & Sciences and
the Graduate School of Michigan State University funded this research.

vi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................................... ix
LIST OF FIGURES ......................................................................................................................... xi
CHAPTER 1 ...................................................................................................................................... 1
INTRODUCTION............................................................................................................................. 1
CHAPTER 2 .................................................................................................................................... 6
LITERATURE REVIEW ................................................................................................................. 6
BRAND EQUITY ..........................................................................................................................7
COGNITIVE ABSORPTION ................................................................................................. 9
COGNITIVE FLOW ............................................................................................................. 13
PURCHASE INTENT ................................................................................................................. 16
SOCIAL MEDIA ENGAGEMENT INTENTION ........................................................................18
POPULARITY & COMMUNITY ENGAGEMENT .................................................................... 20
PROPOSED THEORETICAL MODEL ..................................................................................... 23
CHAPTER 3 .................................................................................................................................. 24
METHODOLOGY ......................................................................................................................... 24
MANIPULATION ..................................................................................................................... 24
ENVIRONMENT ...................................................................................................................... 26
ENGAGEMENT METRICS PRETEST & MANIPULATION .................................................... 26
PARTICIPANTS ........................................................................................................................ 28
INSTRUMENT VALIDATION ................................................................................................... 31
CHAPTER 4 .................................................................................................................................. 37
RESULTS ...................................................................................................................................... 37
DEMOGRAPHICS .................................................................................................................... 37
SOCIAL NETWORKING SITES USAGE .................................................................................. 39
HYPOTHESIS TESTING .......................................................................................................... 43
POST-HOC TEST FOR BRAND EQUITY, PURCHASE INTENTION, AND
BRAND SOCIAL MEDIA ENGAGEMENT INTENTION EFFECTS ......................................... 51
FURTHER EXPLORATORY INQUIRES ................................................................................... 51
SIMPLIFIED MODEL............................................................................................................... 58
SUMMARY................................................................................................................................ 62
CHAPTER 5 .................................................................................................................................. 63
DISCUSSION ................................................................................................................................ 63
LIMITATIONS .......................................................................................................................... 65
DEMOGRAPHICS & PARTICIPANTS.................................................................................. 66
COGNITIVE ABSORPTION & BRAND EQUITY ................................................................. 67
SELECTED BRAND RELEVANCE & SIMULATED ENVIRONMENT ................................ 68
THEORETICAL IMPACT ......................................................................................................... 72
SELF-REPORT VS SIMULATION ........................................................................................ 73
POST POPULARITY/COMMUNITY METRICS................................................................... 74
vii

COGNITIVE ABSORPTION ................................................................................................. 79
PRACTICAL IMPACT ................................................................................................................81
CONCLUSIONS ........................................................................................................................ 82
APPENDICES ............................................................................................................................... 85
APPENDIX A. MANIPULATION - BRAND FACEBOOK
PAGE POST SELECTIONS’ ENGAGEMENT METRICS ......................................................... 86
APPENDIX B. SIMULATED FACEBOOK ENVIRONMENT................................................... 87
APPENDIX C. FACTOR LOADINGS ....................................................................................... 88
APPENDIX D. CONSTRUCT STATISTICS .............................................................................. 90
APPENDIX E. LATENT VARIABLE CORRELATIONS ............................................................ 91
APPENDIX F. MATRIX OF LOADINGS AND CROSS-LOADINGS......................................... 91
APPENDIX G. DEMOGRAPHICS ............................................................................................ 96
APPENDIX H. MODEL FIT COMPARISONS ....................................................................... 100
APPENDIX I. SURVEY ........................................................................................................... 101
BIBLIOGRAPHY .......................................................................................................................... 134

viii

LIST OF TABLES

Table 1. Constructs and Definitions ............................................................................................. 23
Table 2. Tradeoffs of Different Recruiting Methods .................................................................... 29
Table 3. Participant Count ........................................................................................................... 30
Table 4. Factor Loadings .............................................................................................................. 33
Table 5. Construct Statistics ........................................................................................................ 34
Table 6. Latent Variable Correlations ...........................................................................................35
Table 7. Matrix of Loadings and Cross-Loadings ........................................................................ 36
Table 8. Participant Social Media Activities ................................................................................ 39
Table 9. Participant Types of Liked Brand Pages ......................................................................... 41
Table 10. Participant Facebook Brand Page Visits ...................................................................... 42
Table 11. Hypothesis Testing ........................................................................................................ 44
Table 12. Stepwise Linear Regression for R2 Partitioning ............................................................47
Table 13. ANOVAs for Post Popularity Metric Exposure
Differences Compared with True Metrics Being Shown .............................................................. 48
Table 14. All Posts, Post Popularity Groups, MGA Community Metric R2 Means ........................53
Table 15. All Posts, Engagement Groups, MGA R2 Means ............................................................54
Table 16. McDonald’s Brand Awareness Post, Culture Groups, MGA R2 Means ..........................56
Table 17. McDonald’s Brand Awareness Post, Engagement Groups, MGA R2 Means .................. 57
Table 18. McDonald’s Posts, Post Popularity Groups, MGA
Community Metric R2 Means for Simplified Model .....................................................................59
Table 19. McDonald’s Posts, Post Popularity Groups, MGA
Community Metric R2 Means for Brand Equity Antecedents ...................................................... 60
Table 20. Manipulation – Brand Facebook Page Post Selections’ Engagement Metrics ............. 86
Table 21. Factor Loadings ............................................................................................................. 88

ix

Table 22. Construct Statistics: No Manipulation ......................................................................... 90
Table 23. Construct Statistics: No Post Popularity Metrics Shown.............................................. 90
Table 24. Construct Statistics: Metrics Switched ......................................................................... 90
Table 25. Latent Variable Correlations: No Manipulation ............................................................ 91
Table 26. Latent Variable Correlations: No Post Popularity Metrics Shown ................................ 91
Table 27. Latent Variable Correlations: Metrics Switched ........................................................... 92
Table 28. Matrix of Loadings and Cross-Loadings: No Manipulation ......................................... 93
Table 29. Matrix of Loadings and Cross-Loadings: No Post Popularity Metrics Shown ............. 94
Table 30. Matrix of Loadings and Cross-Loadings: Metrics Switched ..........................................95
Table 31. ANOVAs for Post Popularity Metric Exposure Differences
Compared with True Post Popularity Metrics Being Shown ........................................................ 99
Table 32. Standardized Root Mean Square Residuals (SRMR)
for Different Models Considered ................................................................................................ 100

x

LIST OF FIGURES

Figure 1. Companies Have Some Misperceptions Regarding
Why Consumers Interact With Them Via Social Sites ................................................................. 19
Figure 2. Proposed Research Model ............................................................................................ 23
Figure 3. Cultural Scale Distribution of Participants ................................................................... 37
Figure 4. Personality Scale Distribution of Participants ............................................................. 38
Figure 5. Participant Social Networking Site Activities ............................................................... 40
Figure 6. Participant Facebook Brand Page Likes ........................................................................ 41
Figure 7. Participant Facebook Brand Page Visits ....................................................................... 43
Figure 8. PLS Model .................................................................................................................... 44
Figure 9. PLS Model with No Post Popularity Metrics Shown ......................................................45
Figure 10. PLS Model with Post Popularity Metrics Matching Brand Content .............................45
Figure 11. PLS Model with Post Popularity Metrics Opposite Brand Content ............................. 46
Figure 12. PLS Model with Post Popularity as a Moderator ......................................................... 49
Figure 13. PLS Model, Brand Awareness McDonald’s Post.......................................................... 50
Figure 14. PLS Model, Brand Awareness Delta Airlines Post ....................................................... 50
Figure 15. All Posts, MGA Post Popularity Metric Path Differences .............................................53
Figure 16. All Posts, Engagement Groups, MGA Path Differences ............................................... 55
Figure 17. McDonald’s Brand Awareness Post,
Culture Groups, MGA Path Differences ........................................................................................56
Figure 18. McDonald’s Brand Awareness Post,
Engagement Groups, MGA Path Differences ................................................................................ 57
Figure 19. McDonald’s Posts, MGA Post Popularity
Metric Path Differences for Simplified Model ...............................................................................59
Figure 20. McDonald’s Posts, MGA Post Popularity Metric
Path Differences for Brand Equity Antecedents ............................................................................ 61

xi

Figure 21. Funding and Governance Combinations of Online Brand Communities..................... 77
Figure 22. Example of Simulated Facebook Environment ........................................................... 87
Figure 23. Age Demographics: No Post Popularity Metrics Shown ............................................. 96
Figure 24. Age Demographics: No Manipulation ......................................................................... 96
Figure 25. Age Demographics: Metrics Switched ......................................................................... 96
Figure 26. Age Demographics: All Manipulation Groups ............................................................ 96
Figure 27. Culture Demographics: No Post Popularity Metrics Shown ........................................97
Figure 28. Culture Demographics: No Manipulation ....................................................................97
Figure 29. Culture Demographics: Metrics Switched ....................................................................97
Figure 30. Culture Demographics: All Manipulation Groups .......................................................97
Figure 31. Personality Demographics: No Post Popularity metrics Shown ................................. 98
Figure 32. Personality Demographics: No Manipulation ............................................................. 98
Figure 33. Personality Demographics: Metrics Switched............................................................. 98
Figure 34. Personality Demographics: All Manipulation Groups ................................................ 98

xii

CHAPTER 1
INTRODUCTION

On March 21, 2005, the PEW Research Center started collecting social media data (Pew
Research Center, 2017). At the time, only five percent of American adults were accessing social
networking sites (Pew Research Center, 2017). By November 2016, PEW estimated that seven out
of ten adults used social media for varying activities (Pew Research Center, 2017). As of 2016,
Facebook held the majority share of social media activity (MarketingCharts). Within in a one
month period, it is estimated that 42% of users’ social media activity was conducted on Facebook
(MarketingCharts).
And the numbers continued to grow. During the first quarter of 2017, Facebook averaged
1.28 billion users each day, and had 1.94 billion unique users monthly (Facebook, 2017). By 2020,
it was projected that 2.95 billion people will have access to social networks - and Facebook is
assisting with this (eMarketer, 2016). Facebook Lite, a less than one megabyte version of
Facebook, was developed to enter markets with slower internet connections, and has already
launched in India and the Philippines (Shankar, 2015; Statista, 2017).
Facebook allows for effective use of time while still generating revenue for a company. By
dedicating as little as six hours a week towards social media planning and preparation, marketers
are able to see increases in site traffic, lead generation, and brand exposure (Kusinitz, 2014). In
terms of revenue, in the first quarter of 2017 alone, Facebook earned $8.03 billion with 97.82%
($7.86B) of that coming from advertising revenue (Facebook, 2017). Prior studies report on the
growing budgets associated with social media marketing (Coursaris et al., 2013; Moorman, 2013).
In 2009, the total United States market for virtual goods was over a billion dollars according to
Inside Network (Huang, 2011). By 2010, that number had almost doubled. The sales for
FarmVille, a Facebook social game, alone in 2009 were $0.15 billion. From the brand side, in 2011
advertisement spending was $3.8 billion worldwide (Pan, 2012). In the United States alone, $840
1

million was spent on social media advertisements in 2011 (Pan, 2012). In 2012, the Pan article
projected advertisement spending in the United States alone would reach $3.1 billion by the end
of 2016. Facebook was able to show more than doubled that in ad spending by the third quarter
of 2016, having brought in $6.8 billion in advertising revenue (LePage, 2016). Globally, social
media advertisement spending is expected to increase 26.3% and is likely to exceed $35 billion in
2017, and this represents only 16% of all digital advertisement spending projected for that year
(LePage, 2016).
Facebook highlights their personalized and targeted advertisements by emphasizing their
superior reach, relevance, engagement, and social context (Pan, 2012). As a platform that people
join faster than they can incorporate ads into the news feed (Pan, 2012), this makes sense. Users
are provided a communication platform integrated with indirect product experiences designed to
facilitate brand communication in an unobtrusive way (Huang, 2011).
However, few studies attempted to solidify the value proposition of brand social media
marketing communications. While Coursaris et al. (2014a) worked toward understanding the
value of social media, further research was still needed to solidify these findings for the general
population. Hence, this research set out to build on current understandings and address the
overarching research question:
RQ: What is the value proposition of social media to brand marketing communications?
While traditional marketing initiatives and their benefits for boosting brand equity and
purchase intention had been studied extensively, the virtual relationship between brand and
consumer still had yet to receive thorough inquiry. This study attempted to further link social
media communications with consumer brand relations, by drawing on literature linking the
likelihood of engagement to actual purchase intentions. From a company’s standpoint, it must be
demonstrated that there is financial benefit to social media marketing in order to justify increased
marketing spending in digital domains (Aaker and Jackson, 2001). Agozzino’s 2012 study showed
that public relations professionals believed that social media marketing was effective for
2

maintaining brand reputation, awareness, search rankings, and web traffic. However, increasing
the number of social media platforms a company used did not help user satisfaction with the
consumer-brand relationship (Agozzino, 2012; Aaker and Jackson, 2001). When it was also
considered that that as of 2012, Facebook held 90% of the market share of active monthly social
media usage, this became an interesting perspective (Lunden, 2012). This lead to and reinforced
the importance of answering the question of whether or not social media could offer a brand
monetary value, and questioned what factored into a consumer’s brand equity for that particular
brand (Aaker and Jackson, 2001). By understanding brand equity contributors, marketers could
better manage messaging content to increase brand equity and ultimately increasing consumer
purchase intent. This lead to the next research question to be addressed in this project:
RQ1: What mechanisms account for the link between brand equity and engaging
content?
With one in four people using a social network as of 2013, social media can produce a huge
audience for brands (Davis, Piven, and Breazeale, 2014; Hosea, 2011). And by April 2016, that
number had increased to 68% of adults using Facebook as their social media platform of choice
(PEW Research Center, 2017). Consumers interact with Facebook posts on a daily, sometimes
hourly, basis, making social media an ideal platform for brands to foster close relationships with
their consumers. Because social media functions in assisting at the input and output ends of the
sales funnel – attract and delight stages – it facilitates customer attraction as well as retention
and thus community involvement. And because 71% of consumers who have a good interaction
with a brand on social media are likely to recommend the brand to others (Hainla, 2017), it is all
the more important to understand the community aspect of social media interactions between
consumers and brands.
Therefore, traditional offline marketing theories may not be appropriate for studying
online word of mouth communications due to the friend role a brand assumes in digital and social
media (Brown, Broderick, and Lee, 2007). In particular, as social media provides a platform to
3

bring like-minded consumers together to exchange ideas and opinions about their favorite
brands, the otherwise simple relationship between brand and consumer becomes more
multifaceted in this social space. The importance of the community and understanding these
facets requires further understanding the following research question:
RQ2: What is the role of community in driving brand equity, purchase intent, and brand
social media engagement intent?
Previous empirical research suggested that through increased interactions with a brand on social
media, consumers did not merely build a relationship with the company but also with other
consumers who are connected with that brand on social media (Wang, Yu, & Wei, 2012; Wirtz, et
al., 2013; Zaglia, 2013). Suddenly, it was not just the initial consumer who “likes” something but
a sense of community – the interactions with a group of people who shared a similar interest in
the brand – that may have resulted in increased positive attitudes toward the brand (Coursaris et
al., 2013, 2014a).
Therefore, this study set out to bring insight to four objectives. First, it looks to establish
that exposure to high engagement messages did have an effect on consumers in the general public
by using an external, volunteer sample pool as opposed to the student sample used to establish
the baseline in Coursaris, et al., 2014a. Second, this research looked to explain the overall
modeled relationship in Coursaris, et al., 2014a by looking at flow and cognitive absorption as
possible contributors toward consumer brand equity. Third, it looked to establish or refute the
significance of communication within a community by manipulating the exposure of community
input to the brand message. And fourth, using an encapsulated environment, this research looked
at the variance between actual social media engagement and self-report of engagement. Such
findings were interesting to see how consumers actually interact with the brand versus how they
perceived their interactions to be, and whether or not the presence of a social community had any
effect on this.

4

Based on these findings, this dissertation offered the following contributions. First, it
empirically tested the relationship between social media-based brand engagement and purchase
intention and exposed whether or not investments by companies in social media marketing
communications were associated with a potential return on investment. Second, this research
tested the Coursaris, et al., 2014a new construct of Engagement Intention on a public sample. This
metric aimed to measure an important behavioral outcome in the context of building brand equity
and loyalty in the social space. Third, this study drew on actual brand Facebook Page Posts that
were classified according to their messaging content and objectively evaluated in terms of their
impact (i.e. the evoked level of Likes, Comments, and/or Shares) before exposing the impact they
may have on consumers’ perceptions and behavioral intentions towards the brand in its natural
exposure environment of Facebook (Coursaris, Van Osch, & Balogh, 2013; Coursaris, et al.,
2014b). Fourth, this study actively tested brand communities and consumer engagement effects
on a live, simulated platform.
The remainder of this paper is organized as follows. First, a review of the theoretical
constructs that compose the conceptual foundation of this study is presented. Then, the research
design of this study, including the experimental manipulation, participants, instrument
validation, as well as data collection and data analyses is discussed. And finally, a discussion of
possible findings in relation to industry value added to practice and social media communications
design as well as marketing scholarship and future research ideas will be presented.

5

CHAPTER 2
LITERATURE REVIEW
When interactivity is incorporated into social media posts, control of the content is given
to the consumer who in turn is able to pay attention to the brand responsiveness of the content as
well as other consumers in that community (Davis, Piven, & Brazeale, 2014). Social media
provides the ideal platform for brands to build social experience and symbolic value for the
consumer, thereby potentially contributing to an improved brand image in the consumer’s mind,
thus enhancing the consumer’s attitude toward the brand, or brand attitude.
The extent to which a social media message is engaging – i.e., resulting in consumer
engagement in the form of Likes, Comments, or Shares – can affect brand image in two ways.
First, an engaging message can facilitate the formation of positive brand attitude through the
manipulation of textual and/or visual persuasion cues. For instance, a brand can manipulate
content themes (e.g., for a messaging typology, see Coursaris, et al., 2013) as well as the richness
of a message (e.g. by including a URL, photo, or video) that increase the level of engagement
associated with the message. Second, an engaging message can positively affect brand equity by
triggering a positive feeling state (Coursaris, et al., 2014b). Therefore, a brand can manipulate the
appeal used in a message (transformational as opposed to informational) to elicit stronger
emotional experiences and in turn more positive brand attitudes thus enhancing brand equity
(Coursaris, et al., 2013).
Consumers are able to provide a persuasive effect over other consumers 22 times stronger
than marketers (Goh, Heng, and Lin, 2013). Brands are able to provide a space for consumers to
engage in thought provoking conversations and inspire resonant stories within this trusted
community of the individual’s social network (Briggs, 2010; Gensler, et al., 2013; Davis, Piven, &
Brazaeale, 2014). Consumption within this community is based on the consumer’s need for
problem solving, information search, feedback, and evaluation of series offered in an environment

6

where real-time accessibility and exchange of these types of information is considered the social
norm (Davis, Piven, & Brazeale, 2014; Hennig-Thurau, et al., 2010). By looking at social media
posts that have previously attracted high engagement, this study capitalizes on the social
interaction of, essentially, strangers within this brand community and gauge if consumers do in
fact place a higher value on these posted based on their community support to accompany the
initial marketer contributions (Davis, Piven, and Brazeale, 2014; Goh, Heng, and Lin, 2013). From
here, a relationship with brand equity should first be established.

BRAND EQUITY
The incremental utility or value added by a brand name which contributes to the
company’s long-term profitability is brand equity (Chen & Chang, 2008). Brand equity can
therefore be discussed in relation to the investor, the manufacturer, the retailer, or the consumer,
and one must consider the perspective to be used for analysis (Cobb-Walgren, Ruble, & Donthu,
1995). This study looks at brand equity from the point of view of the consumer.
Consumer brand equity can be further divided into two parts: brand image and brand
attitude, and any measurement attempts at brand equity must recognize these multidimensional
aspects for proper consideration of all factors involved (Faircloth, Capella, and Alford, 2001;
Chang & Liu, 2009). Drawn from the connections a consumer makes, brand image is defined and
evaluated based on a cluster of attributes and associations to the brand name based on consumer
perceptions of the brand’s tangible and intangible associations (Chang & Liu, 2009; Faircloth,
Capella, and Alford, 2001). Built by anything that allows consumers to experience the brand,
brand awareness can work as a driver for brand image by influencing consumer choice among the
purchase decisions set, even if there is no other knowledge or associations for the brand (Hutter,
et al., 2013). Yet few marketing studies have examined the connections between positive brand
image and brand equity as they drive purchase intention, much less in the social media realm,
thus increasing the value of this study (Faircloth, Capella, & Alford, 2001).
7

One of the most widely examined constructs in consumer behavior and an important
concept in marketing research for the past twenty years, brand attitude is defined as an
individual’s internal evaluation of an object such as a branded product (Faircloth, Capella, and
Alford, 2001; Mitchell and Olson, 1981). Brand attitude is a function of salient beliefs, those from
memory and considered by the consumer in a given situation (Mitchell and Olson, 1981). By better
understanding causal dynamics of attitude formation and image interpretation, we can aid
marketing researchers to understand the attitudinal and impression impact while assisting
managers in developing more effective marketing strategies (Mitchelle & Olson, 1981).
However, effects on brand equity can be industry specific. A 2012 study by Bruhn,
Schoenmueller, & Schafer found that brand created social media communications had a greater
indirect effect on consumer brand attitude and purchase intention than user-generated social
media communications or traditional media communications. Looking at brand image, the study
found that brand generated content has a greater influence on telecommunication and
pharmaceutical industry communications, while user-generated social media communications
have a greater influence in the tourist industry, specifically on the areas of brand awareness and
brand image (Bruhn, Schoenmueller, & Schafer, 2012). They also saw significant differences in
effect sizes for the varying industries, and indicated that social media campaign implementation
should be part of the marketing mix. However, consideration of audience, source, and credibility
direction should be carefully considered when putting together content messages in order to
enhance consumer-based brand equity (Bruhn, Schoenmueller, & Schafer, 2012).
Brand equity provides the insulation needed to protect brands from competitors (CobbWalgren, Ruble, & Donthu, 1995). High equity has been associated with consumer satisfaction,
brand preference, premium price, and high profit values (Chang & Liu, 2009). Definitions have
ranged from Keller’s 1993 definition that brand equity is the effect of brand knowledge on
consumer response to brand marketing to simply the customer’s perceived value (Chen & Chang,
2008). This study uses Chang & Liu’s 2009 definition of brand equity as the brand name’s added
8

attractiveness to the consumer for the product or service (see Table 1). Therefore the following
hypothesis is established to provide additional insights toward linking consumer brand equity and
social media activities:
H1: The higher the elicited engagement level of a brand’s post, the more positive the
consumer brand equity.
However, while previous studies have shown that there is a relationship between
consumer engagement and consumer brand equity (Coursaris, et al., 2014a; Bruhn,
Schoenmueller, & Schafer, 2012), research has not been done to look at what factors into brand
equity in a virtual environment. Understanding of these predecessors would allow a brand to
perhaps influence consumer acceptance of the brand even earlier in the buying process thus
building higher brand equity and leading to higher purchase intentions later on. By considering
the cognitive absorption and flow dimensions, further understanding can be gained of what
thoughts are being processed to make it into a consumer’s memory and, in turn, creating
consumer brand attitude and brand image to form brand equity.

COGNITIVE ABSORPTION
Cognitive absorption is considered in this study due to its unification of control, curiosity,
and focused attention attributes which drive salient beliefs, a key representation of brand attitude,
which is a driver of brand equity (Agarwal & Karahanna, 2000). Cognitive absorption can also be
described as the culmination of three areas of research: the cognitive absorption, flow, and
cognitive engagement dimensions (Agarwal & Karahanna, 2000). This study defines cognitive
absorption as the extent in which a user is absorbed when using the system (see Table 1; BurtonJones & Straub, 2006). That is to say, how immersed a person is while using technology.
Rooted in psychology, cognitive absorption is defined by a person’s deep involvement/
attention in a task, specifically, a job involving a computer (Agarwal & Karahanna, 2000;

9

Chandra, Srivastava, & Theng, 2012; Elmezni & Gharbi, 2010). It is defined by the individual
losing the concept of time and becoming so intensely involved in an activity that nothing else
matters (Elmezni & Gharbi, 2010). The experience is characterized by being so enjoyable that
people will continue to do the activity at hand, no matter the cost, for the sake of doing it (Elmezni
& Gharbi, 2010). At this point, the person is able to subconsciously process task objectives, while
still experiencing enjoyment in the activity (Elmenzni & Gharbi, 2010).
A psychological state, cognitive absorption represents an experience in which the
participant is immersed in a task of pure enjoyment, specifically with technology/software
(Wakefield & Whitten, 2006; Leger, et al., 2014; Elmezni & Gharbi, 2010; Chandra, Srivastava, &
Theng, 2012). There are two ways to look at this construct. The first looks at the factors that
influence the construct and organizes them into three dimensions of cognitive absorption, flow,
and cognitive engagement (Agarwal & Karahanna, 2000; Leger, et al., 2014). While cognitive
absorption looks at the deep attention being given to something, the theory of flow is meant to
explain how involved people get in an activity that nothing else matters (Agarwal & Karahanna,
2000; Chandra, Srivastava, & Theng, 2012). Accompanying these, cognitive engagement
considers the playfulness and inherent interest someone has in the activity at hand (Chandra,
Srivastava, & Theng, 2012). Chandra, Srivastava, & Theng (2012) acknowledge that these are very
similar concepts, but together work as a triangulation of three streams of research used to describe
the behavioral state of someone being cognitively absorbed in an activity (Chandra, Srivastava, &
Theng, 2012).
The second way to look at cognitive absorption is that it is comprised of five dimensions:
temporal dissociation, focused immersion, heightened enjoyment, control, and curiosity (Agarwal
& Karahanna, 2000; Elmezni & Gharbi, 2010). Temporal dissociation is the inability for the
participant to register the passage of time while in focused immersion in which all other stimuli
are ignored (Agarwal & Karahanna, 2000; Elmezni & Gharbi, 2010). Heightened enjoyment is
what captures all of the pleasure aspects of the activity. The participant’s control allows them to
10

focus on the task at hand while curiosity is what allows the interaction to tap into their sensory
and cognitive functions (Agarwal & Karahanna, 2000; Elmezni & Gharbi, 2010).
It makes sense that these two structures exist, in that they possibly overlap each other with
similar concepts. To parallel the first structure of looking at cognitive absorption as a construct,
one could consider the similarities of the dimension of cognitive absorption to align with the
control and part of the curiosity aspect of the second structure, while the dimension of flow could
align with the temporal dissociation and focused immersion aspects, and cognitive engagement
could align with heightened enjoyment and part of the curiosity dimensions (Agarwal &
Karahanna, 2000). While Agarwal & Karahanna (2000) utilize both models, it is not clear why
two different structures have been proposed and one not selected as the accepted practice for
analyzing cognitive absorption states in technology usage. Chandra, Srivastava, & Theng (2012)
utilize the smaller model of three dimensions as a prime second-order construct being evaluated
as a single collapsed indicator, a reflected construct, while acknowledging this option can cause
measurement and validity problems. They simply indicate that this is common practice to subconstruct larger dimensions; perhaps for initial investigation. Therefore, for the purposes of this
research, the first model, considering three dimensions that make up the cognitive absorption
construct will be further analyzed for an overarching analysis, to investigate if cognitive
absorption and flow, as dimensions, possibly impact brand equity in a consumer-brand purchase
and interaction involvement model.
Cognitive Absorption is an interesting construct to consider when looking at behavioral
intention. Agarwal & Karahanna’s 2000 study proposed a theoretical model in which cognitive
absorption was an antecedent to perceived usefulness and perceived ease-of-use. In turn, these
dimensions could lead to intent of a specified behavior. By intrinsically motivating consumers
with increased cognitive absorption, cognitive burden is lowered, thus increasing the perceived
ease of use and usefulness of working with a technology (Shang, Chen, & Shen, 2005; Agarwal &
Karahanna, 2000). Agarwal & Karahanna (2000) argue that that increased perceived usefulness
11

shows significant influence on utilizing a technology system because user performance is
enhanced by the increase in job/activity performance due to the decrease in cognitive burden to
perform required job functions. The focused immersion aspect of cognitive absorption also works
to facilitate dedication to a task thereby reducing the cognitive burden associated with performing
the task (Argarwal & Karahanna, 2000; Burton-Jones & Straub, 2006). Decreased cognitive
burden and increased cognitive absorption are therefore important factors for a user to form
beliefs and use intentions (Wakefield & Whitten, 2006).
While there has been research linking intrinsic motivations and cognitive absorption, the
application of these motivations to social media is relatively new with the new medium (Leger, et
al., 2014). Lin’s 2009 study promotes user friendly design in virtual worlds as a way to promote
cognitive absorption and facilitate usage intentions through increased pleasure with the
experience. This suggests that Facebook’s design is created to facilitate further use of the
community and their media environment. Just as it is important for cognitive absorption to be an
intrinsic motivator for user acceptance of a virtual environment, it is also important for the
community provider to try and keep members online as long as possible (Lin, 2009).
By building longevity and loyalty with a technology, companies are able to keep users
engaged with the brand longer and potentially increase the consumer’s trust and loyalty in the
brand, ultimately increasing brand equity (Chandra, Srivastava, & Theng, 2012). Cognitive
absorption becomes a relevant dimension both to increase the longevity and time spent by users
in virtual worlds as well as building trust with the promoted/hosting brand of the environment,
therefore decreasing any perceived risks associated with virtual worlds and adaptive use
intentions (Chandra, Srivastava, & Theng, 2012; Kang & Hur, 2012). With the increasing
popularity of relationship based marketing in social platforms, trust becomes of the utmost
importance. Multiple studies – Laroche, Habibi, & Richard, 2013; Manara & Roquilly, 2011; Sashi,
2012; Kang & Hur, 2012; to name a few – have established trust as an important predecessor to
relationship marketing and increasing brand equity. Therefore, it makes sense to consider the
12

impact of a participant’s cognitive absorption with a social platform (Kang & Hur, 2012), to see
how such deep involvement impacts a consumer’s brand equity and therefore devotion to the
brand. Saade & Bahli’s 2005 study even goes so far as to suggest that such brand equity can even
be linked to other return on investment benefits such as reduced marketing costs, trade leverage,
price premiums, and increased market share. And Leger, et al.’s 2014 study found that
participants with higher cognitive absorption have better training outcomes. This further
strengthens the reasoning to investigate cognitive absorption as a factor of brand equity and
ultimately, social media brand engagement.

COGNITIVE FLOW
An important element to understanding human-technology interactions (Agarwal &
Karahanna, 2000), cognitive processing flow represents an affective state when individuals are so
involved with computer usage activities thus resulting in playful and exploratory behavior with
their usage intentions (Jiang & Benbasat, 2004). First proposed in 1975, Csikszentmihalyi
originally intended for flow to represent a psychological state of immersion in an activity or
behavior (Cheng, Chieng, M.H., & Chieng, W.H., 2014). According to Gao & Bai (2014), flow is
when one’s skills and challenges overpower the baseline experience that a user will experience a
state of flow. For this study, flow is defined as the perceived sense of intrinsic enjoyment form
interacting with the virtual environment (see Table 1; Animesh, et al., 2011).
Similar to cognitive absorption, there is dispute over the number of dimensions that feed
into flow as its own construct. Two dimensions seem to be constant across several studies: intense
concentration and a sense of control (Agarwal & Karahanna, 2000; Leger, et al., 2014;
Csikszentmihalyi, 2014). Intense concentration refers to when an individual pays such strong
attention to something that their focus is on a very narrow stimulus represented by the technology
(Trevino & Webster, 1993; Agarwal & Karahanna, 2000). Sense of control refers to the individual’s

13

perception that they control their interaction with the technology (Agarwal & Karahanna, 2000).
Trevino & Webster’s 1993 study utilizes curiosity and intrinsic interest as two additional
dimensions to define the flow construct. However, in another study by the same group of
researchers, they were unable to distinguish between the curiosity and intrinsic interest
dimensions and therefore recommended combining the two for the third dimension leading to
the construct of flow (Webster, et al., 1993). The curiosity dimension is meant to indicate that
during the flow experience, there is an increased arousal of sensory and cognitive inquisitiveness,
while intrinsic interest indicates an individual’s interaction with technology extends beyond
pleasure and enjoyable in itself (Trevino & Webster, 1993; Agarwal & Karahanna, 2000).
Essentially, it makes sense that they might have trouble distinguishing between the two groups,
when curiosity can work hand in hand with pure enjoyment thus making them concepts that
essentially fuel each other. Finally, Csikszentmihalvi in 1990 proposed in addition to intense
concentration and a sense of being control, that the dimensions of loss of self-consciousness and
transformation of time be included as dimensions that build to the flow construct. Jiang &
Benbasat (2004) probably categorize it the best when they kept the control and
focus/concentration dimensions, but considered the other disputed areas as cognitive enjoyment.
Hoffman and Novak (1996) state that flow is accompanied by a loss of self-consciousness
and is therefore self-reinforcing. Additionally, when the level of difficulty of a task matches a
individual’s skills, the individual will intrinsically motivated to dedicate attention to a task and
while entering a state of cognitive flow in which they experience a relaxed state of calm and control
while completing the task (Leger, et al., 2014). In their 2011 study, Animesh, Yang & Oh note that
too many avatars in a virtual world increases the possibility of sensory overload which forces
participants to engage in undesired actions, thus weakening involvement and engagement, and
therefore flow, in the virtual world. However, they also argue that if a community has a stable
population, then perhaps the consumer will enjoy their time and participation in the environment
(Animesh, Yang & Oh, 2011; Gao & Bai, 2014). Individuals who have experienced flow reported a
14

more positive experience, and a higher likelihood of satisfaction and loyalty (Animesh, Yang &
Oh, 2011). Finally, they showed that individuals who experience flow have a significant impact on
purchase intention, while Cheng, Chieng, M.H. & Chieng, W.H.’s 2014 study showed that flow is
positively related to loyalty. So perhaps there is a relationship between flow and loyalty through
the brand equity area thus leading to the idea that consumers who frequently visit a brand’s page
have greater loyalty to the brand and therefore higher brand equity than compared to consumers
who do not frequently visit brand pages (Cheng, Chieng, M.H. & Chieng, W.H., 2014).
Social interactions are a basic need of humans (Animesh, Yang, & Oh, 2011). Multiple
studies have shown that increased social interactions correlates to experiences of flow (Kim, et al.,
2005; Animesh, Yang, & Oh, 2011; Huang, 2014). As a basic human need, social interactions have
been shown create enjoyable experience, which technology only enhances and facilitates the
enjoyment (Park, et al., 2012; Animesh, Yang, & Oh, 2011). A sense of psychological closeness
created by social interactions has been shown to lead to flow experiences, and therefore increasing
the longevity of engagement with social platforms and virtual worlds (Animesh, Yang, & Oh, 2011;
Drengner, Gaus, & Jahn, 2008). Drenger, Gaus, & Jahn’s 2008 study recommends managers
provide an environment and the opportunities for facilitating communication and interaction as
the flow created will assist in the social interactions and a positive impact on brand image can
even be expected.
Cognitive flow is an important component of understanding technology interactions as
and therefore an antecedent to understanding attitudes towards technologies and, in the digital
age, brands (Argarwal & Karahanna, 2000; Drenger, Gaus, & Jahn, 2008). The immersion created
by cognitive flow allows for positive experiences which influence learning and attitudes (Na,
Eschenbrenner, Y DeWester, 2014). In their 2014 study, Na, Eschenbrenner, & DeWester found
that telepresence positively influences both enjoyment and brand equity. Nah, et al. (2010) also
found that the flow experience has a direct influence on perceptions of brand equity in a branding
virtual world site. This in turn increased participant perceptions of the brand, and increased the
15

likelihood that they would use the brand associated with the site (Nah, et al., 2010). This is
interesting to consider when looking at the proposed theoretical model, because while Huang’s
2014 study found that virtual businesses can expect to see increased purchase behaviors when
cognitive involvement and flow experiences increase and enhanced the better the social features
are, Nah et al.’s 2010 study found that an increase in flow, while it does show an increase in brand
equity, it does not show an increase in purchase intention. These two contradictory positions
warrant further study which is provided in the proposed research model.

PURCHASE INTENT
Abundant evidence exists in the consumer research literature that the more positive the
brand attitude, the higher the consumers’ purchase intention (Chang & Liu, 2009; Park, et al.,
2010). Purchase intention here refers to the behavioral inclination of consumers to plan to
purchase a certain product or service in the future (See Table 1; Dodds et al., 1991; Schiffman and
Kanuk, 2007). Positive purchase intention, in turn, is viewed as an important antecedent to actual
purchase action (Fishbein and Ajzen, 1975; Schiffman and Kanuk, 2007).
Attitudes examine a person’s evaluations while intentions look at a person’s motivations
to carry out an effort or a behavior, thus making sense that intentions and attitudes are two
distinct measures (Irshad, 2012). Fishbein and Ajzen’s Theory of Reasoned Action (1975) uses
attitudes and personal behavior norms to determine intentions. Because intention has wider
implications and often a positive impact on an individual’s actions (Hung, et al., 2011), this study
focuses on purchase intentions and therefore a customer’s willingness to buy a brand (Irshad,
2012).
However, asking purchase intent questions has been shown to prime purchase processing
decisions in favorable ways toward the brand. In a 1993 study by Morwitz, Johnson, and
Schmittlein, purchasing rates increased by 84% for armature personal computer shoppers just by
measuring their purchase intent. A 1996 study by Fitsimons and Morwitz found that simply asking
16

intent questions caused purchase rates to increase for brands that already had high market shares.
Consumers in the late states of the decision-making process are more likely to have already
formed attitudes and purchase intention thus prompting retrieval of preexisting cognitions
toward the brand when asked about purchase intent (Fitzsimons and Morwitz, 1993). To combat
this, Dodds, Monroe, & Grewal’s 1991 study shows that as brand name or store name information
is provided to a participant, the negative effect on price perception is decreased. Thus a brand
with increasing brand equity has a higher ability to control the perception of the brand and
possibly moderate any purchase intention decrease caused by question priming.
Positive, or high, brand equity is a precursor for customer activities such as recommending
the brand to others as well as making repeat purchases (Chung, Lee, and Heath, 2013). This is
consistent with brand equity being an asset to the business and thus leading to higher future-term
financial performances (Chang & Liu, 2009; Aaker and Jacobson, 2001; Faircloth, Capella, and
Alford, 2001). Therefore, the following hypothesis is proposed:
H2: The more positive the consumer brand equity, the greater the consumer’s brand
purchase intention.
However, another factor in the proposed model and in today’s social shopping society is
the online social influence and how this affects purchase intentions. Huang’s 2011 study showed
that when participants use a social networking site, their cognitive involvement increases as they
are exposed to site features such as the timeline, updates, and other activities. This increased
involvement by the online platform could affect the response of participates to adopt products
that their network has also adopted (Hung, et al., 2011; Huang, 2011). Huang’s 2011 study also
showed that online involvement has a moderating effect on purchase intent. In fact, their study
found that cognitive involvement and flow experiences are improved and even enhanced when
participants were exposed to high quality social and interactive online features (Hung, et al.,
2011). So now the question becomes how purchase intent relates to social media engagement
intention.
17

SOCIAL MEDIA ENGAGEMENT INTENTION
Consumers do not forward or participate in content unless there is something for them to
build on their own profile or strengthen their personal brand image (Wasko and Faraj, 2005;
Hosea, 2011). In addition to the link between brand attitude and purchase intention, this research
proposes that brand equity will positively affect consumers’ intentions to engage with a brand’s
social touch points. Such engagement may take the form of Liking, Commenting, or Sharing a
brand’s profile (e.g., Facebook Page). If a brand creates engaging content resulting in positive
impressions about and attitudes toward the brand, it is anticipated that consumers will be more
likely to continue visiting the brands’ social media platforms and continue to interact with posts
generated by the brand. For the purposes of this study, the variable of Brand Social Media
Engagement Intent is defined as the degree to which a consumer considers engaging with a
brand’s social media touch point (see Table 1; Coursaris, et al., 2014a).
As Figure 1 below demonstrates, sometimes there are disconnects between consumer
engagement needs and what the brand/company thinks the consumer needs (Baird & Parasnis,
2011). When asked why they visit social media sites, only 23% of those surveyed indicated visiting
social platforms was to interact with a brand, versus 70% choosing to connect with their personal
network (Baird & Parasnis, 2011). This is interesting when it is considered that before Web 2.0
existed, the networking, social component of now being specified as so important to social
network users, was not (Gummerus, et al., 2013). Prior to the dawn of Web 2.0, consumers were
found to primarily experience trust and time saving benefits from their online interactions
(Gummerus, et al., 2013). Perhaps this helps to explain some of the disconnect between the
consumer need to visit brands on online and brand understanding of these consumer needs. The
same study by Baird & Parasnis (2011) found that 33% of their participants did not think
interacting with a business through social media would make them more loyal to the business,
and only 49% of participants believed engaging with the brand would lead to future purchases. As
a result, it is advised that before even considering entering a social media space, a firm should
18

make sure that it has a strong grasp on its branding and that it has the capability to deliever
branding and meet consumer needs through all social touchpoints (Gensler, et al., 2013;
Gummerus, et al., 2013).

Figure 1. Companies Have Some Misperceptions Regarding
Why Consumers Interact With Them Via Social Sites
Building engagement intention requires a shift from marketer led brand messaging to
include the consumer in the messaging creation strategy and establishing a co-creation of
meaning (Davis, Piven, and Breazeale, 2014; Hollebeek, Glynn, and Brodie, 2014). Using the
symbolic aspect of brands and communities built by social media, consumers are able to build
their social identities and increase their self-confidence from other consumer “Likes” and
interactions with their posts (Davis, Piven, and Breazeale, 2014). Such interaction increases brand
tangibility and could result in repeat consumption of the brand as well as interaction with the
social media community established by the brand (Davis, Piven, and Breazeale, 2014; Hollebeek,
Gylnn, and Brodie, 2014). Still, understanding of both approaches to measurement and
19

assessments is still lacking as to whether or not social media message consumption actually
increases purchase intention and/or further social media-based engagement intention. Thus, the
following hypothesis is proposed:
H3: The more positive the brand equity, the greater the brand social media engagement
intention.

POPULARITY & COMMUNITY ENGAGEMENT
Brand communities are not a new concept. Online brand communities, on the other hand,
are working through the initial stages of development and understanding.
As mentioned earlier, brands take on more of a peer role in brand communities (Wang,
Yu, & Wei, 2012). Online brand communities are composed of people who have some level of
loyalty to the entity that has organized the page (Laroche, Habibi & Richard, 2013; Erdogmus &
Cicek, 2012). Because of this new and unique relationship among strangers uniting for a common
interest, consumers and brands are able to create a dialog back and forth. Suddenly we’re not just
looking at e-word of mouth, but brands actively facilitating an electronic exchange in a controlled
marketing environment of the social media world (Erdogmus & Cicek, 2012; Alhabash, et al.,
2013). Therefore, it makes sense that perhaps we could borrow from peer theories such as
consumer socialization theory and that community relations would have an effect on consumer
attitudes towards the brand and the community.
On one side it can be argued that consumers who follow a brand on social media have
increased loyalty to that brand (Laroche, Habibi & Richard, 2013; Wirtz, et al., 2013). On the
other, consumers following the brand already have loyalty to that brand, and therefore have no
need to further interact on their social platform (Baird & Parasnis, 2011). An antecedent of loyalty
is trust (Laroche, Habibi & Richard, 2013). In fact, the 2013 study by Laroche, Habibi & Richard
shows that brand trust is a mediator in assisting brand communities, which translates into brand
loyalty. Brand loyalty is an antecedent to brand image, which is an antecedent to brand equity
20

(Bondesson, 2012; Chang & Liu, 2009). Because current customers can add so much value to a
purchase decision discussion (Sashi, 2012), it therefore makes sense to try and connect the brand
equity and engagement with community effects.
Social media facilitates consumer socialization theory by providing the communication
tools to make the socialization and learning process easy for users (Wang, Yu & Wei, 2012). The
consumer socialization theory is the process in which consumers learn skills and behaviors by
communicating with others, which in turn helps them function as a consumer (Wang, Yu, & Wei,
2012). While conventional socialization occurs among people who generally know one another,
online social media enables the learning pool of educators for socialization purposes to open up
to include strangers who also follow a brand (Wang, Yu & Wei, 2012). This idea of strangers
facilitating consumer socialization in an online environment needs and deserves further study.
Consumers have become increasingly more active participants in interactive feedback
loops which provide mostly immediate communication response with other members (Brodie, et
al, 2013). For this study, brand communities are defined as a specialized, non-geographically
bound community based people created around brand admirers. These brand communities foster
social relations (Laroche, et al., 2012), as indicated by the number of Likes, Comments, and Shares
a post receives, which for the purposes of this study, shall be referred to as Post Popularity from
here on out (see Table 1). Consumers interact in brand communities when they can identify
themselves with the brand and believe that the value of contributing outweighs any perceived
risks of contributing (Brodie, et al., 2013; Laroche, et al., 2012). Such interaction builds the
popularity of the post within the post exposure to the brand follower’s as determined by
EdgeRank, the algorithm used by Facebook to determine what should appear in a person’s news
feed. Even Facebook, itself, is now experimenting with showing users the number of views a post
has next to its engagement levels – in addition to showing the active interaction engagement with
the post, so as to show how popular the video is among the followers in the group. Brodie, et al.
(2013) also showed that consumer engagement increases when consumers perceive that the
21

brand’s governing authority and value added to the conversation exceeds the level of effort put
forward by the consumer. Laroche, et al (2012) proposes that brand community practices may not
have evolved enough to obtain conclusive results of their effects. This adds to the need for more
research in the area, and thus the following hypotheses were proposed:
H4a: The higher the post popularity, the greater the consumer brand equity.
H4b: The higher the post popularity, the greater the consumer purchase intent.
H4c: The higher the post popularity, the greater the consumer brand social media
engagement intention.

22

Table 1. Constructs and Definitions
Constructs

Definition

Reference

Cognitive Absorption

The extent to which a user is absorbed when
using the system.

Burton-Jones & Straub,
2006

Flow

The perceived sense of intrinsic enjoyment
obtained from interacting with the virtual
environment.

Animesh, et al., 2011

Brand Equity

The added attractiveness to the customer that
a brand name confers on a product or service.

Chang & Liu, 2009

Post Popularity

The amount of Likes, Comments, Shares a post
receives on Facebook

Documented via
Facebook

Purchase Intention

The degree to which a consumer considers
purchasing a product.

Dodds et al., 1991

Brand Social Media
Engagement Intention

The degree to which a consumer considers
engaging with a brand’s social media touch
point.

Defined and developed
for this study

PROPOSED THEORETICAL MODEL
Based on the above discussion of the six constructs comprising this study, Figure 2
summarizes the proposed research model that will be tested.

Figure 2. Proposed Research Model

23

CHAPTER 3
METHODOLOGY
MANIPULATION
To explore the effect of consumer engagement with brand Facebook Page posts on Brand
Equity, Purchase Intention, and intentions to engage with a brand’s social media presence, actual
posts were selected from two brands, Delta Airlines and McDonald’s. These two brands were used
to allow for the most diverse category selection possible. These brands were chosen for three
reasons. First, they are both among the top 110 Fortune companies ranked by gross revenue, enjoy
strong Brand Equity, and maintain a considerable social media presence. Second, these brands
represent opposite levels of purchase-decision involvement, as follows: low involvement as found
for McDonald’s and other fast-moving consumer goods (FMCGs), and, high purchase-decision
involvement found in Delta Airlines and other expensive and greater risk-bearing goods and
services. Third, the purchase cycle was one additional consideration in the case selection process
resulting in a selection of brands representing both consumables and durables (Sullivan and
Sheffrin, 2003). Purchase cycle – consumables versus durables – is a key factor in brand
messaging when considering purchase-decision frequency and the need for tailored messaging.
Selection of actual Facebook Page posts by the above two brands was then performed by
referring to the messaging typology proposed by Coursaris et al. (2013) and controlling for the
content of posts. Gensler, et al.’s 2013 study emphasizes the importance of order selection as the
sequence of exposure can affect consumer evaluation of brands. Therefore, three content
categories—Brand Awareness, (Product) Promotional, and Engagement—were identified as being
most closely aligned with the four stages of the Buying Behavior Model: Consider, Search, Choose,
and Buy (Coursaris, et al., 2013; Wind, 1978); with Promotional corresponding to both Search
and Choose, and therefore used, in that sequence, for this study. For each of these categories, two
brand Facebook Page posts were selected that were significantly different in the level of elicited

24

engagement, so as to serve as the manipulation in either the low or high engagement condition.
For an overview of the brand’s post content and corresponding engagement, please see Appendix
A.
To calculate the level of elicited engagement, the raw number of Likes, Comments, and
Shares associated with each post was summed to determine an absolute engagement score for an
individual brand Facebook Page post. Hence, six actual brand posts were selected from the official
Facebook Page of Delta Airlines and of McDonald’s (n=12), three that elicited a low level of
engagement from each page and three that elicited a high level of engagement from each page.
These twelve posts were deliberately selected not only on the basis of their engagement scores,
but also to ensure internal consistency with respect to all characteristics of the post, namely the
post’s content category and subcategory as well as media type (e.g. does the post include text only
or a URL, photo, or embedded video) between the two engagement groups.
To illustrate, if a highly engaging Brand Awareness (subcategory: Operations) Facebook
Page post from Delta Airlines that included a photo was selected, another post with the exact same
message characteristics (content = Brand Awareness, Operations; richness = photo) was selected
that displayed a low engagement score. As a manipulation check, one-tailed t-test analysis of the
12 brand posts (6 in the low engagement condition vs. 6 in the high engagement condition) was
performed to determine if the brand Facebook Page posts in the two conditions were indeed
significantly different in the extent to which they generated engagement with the post in the form
of Likes, Comments, and/or Shares. The result (p=0.038) offers support for the significantly
different levels of engagement evoked between the low and high engagement conditions.
Participants visiting the survey URL were randomly assigned to one of six surveys to take.
Two of the surveys featured the six posts with no engagement metrics shown. Two of the surveys
showed six posts with the original engagement metrics as collected in Coursaris, et al. (2013) and
shown in Appendix A. Finally, two of the surveys showed the six posts with the engagement
metrics switch; for example, an established high engagement post would show low engagement
25

metrics, and vice versa. These last two surveys were established to investigate whether or not the
message content is indeed what causes community engagement statistics to begin with.

ENVIRONMENT
Posts were displayed in a simulated Facebook environment in an attempt to preserve
ecological validity with how they would normally see these posts (see example in Appendix B).
Participants were given a notice that this environment was in no way connected to their actual
Facebook profile, as well as given the option to actually Like, Comment, or Share the posts. Upon
clicking the Continue Survey button, participants were then presented with Brand Equity,
Purchase Intention, and Engagement Intention questions. Engagement Intention questions were
asked in follow-up as a self-reported measure to stand as backup to the simulated environment.

ENGAGEMENT METRICS PRETEST & MANIPULATION
The concern was raised that participants might not have any context as to whether or not
engagement metrics were actually high or low. For example, if a post is showing with 9,000 Likes,
is this actually a low engagement post? When considering the matching, established high
engagement post has 20,000 Likes (Coursaris, et al., 2013), it does; but perhaps to a participant
that does not know the full range of the brand’s normal engagement scores, 9,000 might look like
a very high number. In the event that consumers do not perceive the engagement metrics to be in
their corresponding high or low categories, altered metrics should be displayed during the study
to enhance the test of community effects by ensuring participants know they are looking at a large
population engagement.
Therefore, a pretest was run to check engagement levels before displaying them in the
official survey. Fourteen upper-division undergraduates were asked to mark ten engagement
metrics as high or low and to provide estimates on what they thought was a high or low

26

engagement metrics for personal and company brand pages. While all but one student indicated
they thought there was a difference between personal and company Facebook accounts for high
and low acceptable levels of engagement, the only two metrics marked correctly were the two that
had been inflated to look high, with five digits, and to look low, with less than three digits.
Furthermore, participants indicated that an acceptable amount of high engagement would be in
the six to seven digits for a company brand Facebook post. Therefore, it was decided that the
engagement metrics would be adapted for this study. This also served as a way to normalize the
data between industries since both McDonald’s and Delta posts use two different distribution
scales for low and high engagement.
For the purposes of this study, participants viewed inflated or deflated metrics depending
on the survey manipulation they were assigned to. For the surveys showing established high
engagement posts with high engagement metrics, the number of actual likes was multiplied by a
factor of ten, and Random.org was used to create a new ones place digit for the metric. For
example, the McDonald’s Brand Awareness high engagement post had an actual metric of 7,904
Likes. This was displayed in the survey as 79,048, with the eight being supplied by Random.org.
Because the pretest participants correctly marked five digit engagement metrics as high, to
preserve some of the original integrity of the metrics, only five digits were used in the
manipulation. All of the Comments and Shares metrics were three digits, and therefore were kept
the same. Therefore, the same McDonald’s Brand Awareness post which had 616 Comments and
267 Shares was displayed in this study’s survey as such.
For established low engagement posts, the first two digits from the number of Likes were
used for the deflated Like metric. Comments and Shares used only the first digit of the actual
engagement metric. If no metric existed, the post was displayed with a zero. For example, the
McDonald’s Brand Awareness post with established low engagement had 2,108 Likes, 181
Comments, and 41 Shares. This post displayed in the survey environment as having 21 Likes, 1
Comment, and 4 Shares.
27

PARTICIPANTS
One thousand, two hundred participants were recruited from Amazon’s Mechanical Turk
to participate in this study. This was to cover at least 100 participants for each of the twelve
exposure types – two engagement tracks (high vs low), for two brands (McDonalds and Delta),
for three content categories (Brand Awareness, Promotional, and Engagement), resulting in
twelve posts/exposures, or six exposures per participant – while allowing for some dropout. The
only selection criteria specified was that people be in the United States, and be at least 18 years
old, and have a Facebook account. The survey thanked people for their time but dismissed them
if they did not have a Facebook account. Originally, 1,361 people were able to start the survey.
Participant head count after the initial selection criteria was applied was 1,185. Any concerns for
using Amazon Mechanical Turk works for this study should be minimalized. Per Paolacci,
Chandler, & Ipeirotis’ 2010 study and analysis of different sample methods, using Mechanical
Turk workers can actually be less risky than tradition research collection methods (see Table 2).

28

Table 2. Tradeoffs of Different Recruiting Methods

Laboratory

Traditional
web study

Web study
with purpose
built website

Mechanical
Turk

Susceptibility to coverage
error

High

Moderate

Moderate

Low

Heterogeneity of samples
across labs

Moderate

High

High

Low

Non-response error

Low

High

High

Moderate

Subject Motivation

Moderate/High

Low

Low

Low

None

Moderate

Moderate

Low

Risk of contaminated
subject pool

Moderate

High

Moderate

Low

Risk of dishonest responses

Moderate

Low

Low

Low

Risk of experimenter effects

Low

None

None

None

Tradeoff

Risk of multiple responses
by one person

(Paolacci, Chandler, & Ipeirotis, 2010)

In general, Amazon Mechanical Turk workers are slightly younger than the primary internet user
demographic, slightly more educated, slightly lower income, and more knowledgeable about
internet technologies and algorithms (Paolacci, Chandler, & Ipeirotis, 2010; Rader & Gray, 2015).
A major advantage to using these workers though is the ability to decrease the risk of validity
issues via experimenter cross talk and reactance since the participants complete experiments
without interacting with other participants and sometimes do not even know that they are in an
experiment to begin with (Paolacci, Chandler, & Ipeirotis, 2010).
Several participants indicated at the end of the survey in the demographic questions that
they were not in the United States, therefore they were removed from the final response pool.

29

Similarly, participants who did not finish the survey were also removed. A potential issue with
using Amazon Mechanical Turk workers is how feasible it is for them to not complete the Human
Interaction Task (HIT) that they signed up for, thus naturally causing higher attrition rates that
some studies (Rand, 2011). Finally, the standard deviation for all participant responses in similar
construct categories (Cognitive Absorption, Flow, Brand Equity, Purchase Intention, and Brand
Social Media Engagement Intention) was checked for variance to ensure participant engagement.
All participants with standard deviations under five percent were analyzed, and in the end only
participants with a standard deviation below two percent were removed per data screening
recommendations (Kerlinger & Lee, 2000; Gaskin, 2013). Final participant count was 858 across
all conditions and manipulations.

Table 3. Participant Count
Condition

High Engagement Posts

Low Engagement Posts

Original
Total

Final
Total

No Metrics Shown

191

148

Matching Metrics Shown

210

145

Metrics Switched
(shown Low Engagement Metrics)

187

144

No Metrics Shown

189

137

Matching Metrics Shown

199

140

Metrics Switched
(shown High Engagement Metrics)

209

144

Manipulation

Participants were asked to complete baseline survey questions to establish demographics;
social media usage in terms of purposeful interaction, such as Liking, Commenting, or Sharing a
brand Facebook Page post; as well as previous brand Facebook Page visits. In addition, cognitive
absorption was measured using Burton-Jones & Straub’s (2006) scale, flow was measured using
the Animesh, et al.’s (2011) scale, Brand Equity was measured after each post exposure using Nah,
Eschenbrenner, & DeWester’s (2001) Brand Equity scale, and purchase intention was measured
using Chandran, Sucharita, & Morwitz’s (2005) scale. Survey items can be viewed in Appendix I.
30

Participants were randomly assigned into either the low or high Facebook Page post engagement
treatment group, and then within that group, randomly assigned to the different post popularity
statistic viewing levels.
Brand Facebook Page posts were presented in order to match the Buying Behavior Model
(Wind, 1978). Brand presentation order was randomly decided to avoid response bias.
Participants were asked to interact with the posts as if they were on their own Facebook profile by
Liking, Commenting, or Sharing the post. Following this interaction, they were asked to indicate
the likelihood of their interaction on a seven point Likert scale developed for this study to measure
Brand Social Media Engagement Intention. Such a set up will allow us to also measure the
accuracy of self-report between their interaction and their engagement intention responses.
Conclusion of this survey resulted in payment to the participant.

INSTRUMENT VALIDATION
The questionnaire used in this study consists of scales measuring the constructs from the
research model, as summarized in Table 4. Three formative constructs were used – Cognitive
Absorption, Flow, and Post Popularity; while the reflective constructs – Brand Equity, Purchase
Intention, and Brand Social Media Engagement Intention – had significant factor loadings greater
than 0.5 to ensure construct validity (Shimp and Sharma 1987; Carmines and Zeller 1979; Hulland
1999) and were further validated by adequate item-to-total correlations at above the 0.35
threshold as suggested by Saxe and Weitz (1982).
With respect to the dimension of Cognitive Absorption, it was adapted from the scale used
by Burton-Jones & Straub (2006). This scale captures respondents’ immersion with the brand’s
posts using a seven-point Likert scale ranging from Strongly Disagree to Strongly Agree. For the
dimension of flow, the same seven point Likert scale was used with an adapted scale from
Animesh, et al. (2011) to measure respondents’ enjoyment of interacting with the brand’s
Facebook posts. Both dimensions used seven-point a Likert scale anchored as “Strongly
31

Disagree,” “Disagree,” Somewhat Disagree,” “Neither Disagree Nor Agree,” “Somewhat Agree,”
“Agree,” and “Strongly Agree.”
In regards to the construct of Brand Equity, measurements were adapted from the scale
used by Nah, Eschenbrenner, & DeWester (2001). This scale captures respondents’ perceptions
about the added value of a brand along seven-point Likert scale anchored as “Strongly Disagree,”
“Disagree,” Somewhat Disagree,” “Neither Disagree Nor Agree,” “Somewhat Agree,” “Agree,” and
“Strongly Agree.”
The Purchase Intention construct was adapted from Chandran, Sucharita, & Morwitz
(2005) and measured along seven-point Likert scale. The first three items used the anchors of
“Very Low,” “Low,” “Possibly Low,” “Neither Low Nor High,” “Possibly High,” “High,” and “Very
High.” The last two items were anchored as “Strongly Disagree,” “Disagree,” Somewhat Disagree,”
“Neither Disagree Nor Agree,” “Somewhat Agree,” “Agree,” and “Strongly Agree.”
Brand Social Media Engagement Intention was defined in this study by considering the
entire set of a user’s potential interaction behavior with a brand Facebook page post, and creating
a corresponding scale item, for a total of five items. These items were measured along seven-point
scales anchored as “Highly Unlikely,” “Unlikely,” “Somewhat Unlikely,” “Neither Likely Nor
Unlikely,” “Somewhat Likely,” “Likely,” and “Highly Likely.”
Tests for nonresponse and common method biases were not significant, revealing that no
such biases existed. The factor loadings for the items used in this study are summarized in Table
4. As an additional validation check, Appendix C contains the Factor Loadings for each of the
individual manipulation conditions to make sure they fall within acceptable ranges as well.

32

Table 4. Factor Loadings
Resources

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intent

Items

Loadings

When I was reading the post, I was able to block out all other
Distractions.

Formative

When I was reading the post, I felt totally immersed in the
post.

Formative

When I was reading the post, I felt completely absorbed in the
post.

Formative

When I was reading the post, my attention did not get diverted
very easily.

Formative

My imagination is aroused when I interact with the post.

Formative

I feel curious when interacting with the post.

Formative

The interaction with the post is interesting.

Formative

I am absorbed in the interaction in the post.

Formative

It’s fun to interact with the post.

Formative

Even if another [food service/airline] offers the same quality
of services as [McDonald’s/Delta Airlines], I would prefer to
use the services of [McDonald’s/Delta Airlines].

0.975

If there is another [food service/airline] as good as
[McDonald’s/Delta Airlines], I prefer to go to
[McDonald’s/Delta Airlines].

0.978

It makes sense to use the services of [McDonald’s/Delta
Airlines] instead of services of any other [food service/airline]
even if they are the same.

0.964

The likelihood that I would purchase from [McDonald’s/Delta
Airlines]

0.962

The probability that I would consider buying from
[McDonald’s/Delta Airlines]

0.961

My willingness to buy from [McDonald’s/Delta Airlines]

0.960

For this particular type of purchase, I would use
[McDonald’s/Delta Airlines]

0.947

My intention would be to purchase from
[McDonald’s/Delta Airlines]

0.961

Considering this [McDonald’s/Delta Airlines] Facebook Page
post, how likely are you to do each of the following?
Brand Social
Media
Engagement
Intention

Like this post

0.854

Comment on this post

0.851

Share this post on my wall

0.910

Share this post on a friend’s wall

0.875

Like the [McDonald’s/Delta Airlines] Facebook Page

0.884

33

The quality of measures was tested with the partial least squares (PLS) approach to structural
equation modeling via SmartPLS 3.0. The results of the tests for convergent validity (Bagozzi,
1981), discriminant validity (Bagozzi, 1981; Fornell & Larcker, 1981), construct means, and
Cronbach’s α can be found in Table 5, which shows that all constructs had adequate Cronbach’s α
(i.e. > 0.80), reliability (i.e. above 0.7 threshold according to Nunnally, 1978; Carmines & Zeller,
1979), and convergent validity (i.e. AVE above the 0.5 benchmark according to Fornell & Larcker,
1981). See Appendix D for Construct Statistics for each of the individual manipulation conditions.
Table 5. Construct Statistics
Mean

Cronbach’s
Alpha (α)

Composite
Reliability

Convergent
Validity
(AVE)

Discriminan
t Validity
(√AVE)

Cognitive Absorption

4.754

Formative

Formative

Formative

Formative

Flow

3.935

Formative

Formative

Formative

Formative

Brand Equity

3.803

0.971

0.981

0.945

0.972

Purchase Intention

4.218

0.978

0.982

0.918

0.958

Brand Social Media
Engagement Intention

2.719

0.924

0.942

0.766

0.875

Construct

As shown in the last column of the table above, discriminant validity was supported by
confirming that the square root of the variance shared between a construct and its items was
greater than the correlations between the construct and any other construct in the model (Fornell
and Larker 1981).

34

Table 6. Latent Variable Correlations
Construct

Cognitive Absorption

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intention

Brand Social
Media
Engagement
Intention

n/a
(formative)

Flow

0.688

n/a
(formative)

Brand Equity

0.404

0.562

1.00

Purchase Intention

0.419

0.535

0.812

1.00

Brand Social Media
Engagement Intention

0.478

0.688

0.580

0.528

1.00

Discriminant validity was further supported by verifying that all items loaded highly on
their corresponding factors while loading lower on other factors (see Table 6). Although the
correlation between flow and cognitive absorption was quite high (i.e., 0.688), it was not found to
be higher than the 0.85 threshold proposed by Kline (1998), offering further support for the
discriminant validity of the two constructs. All under construct relations were found under this
threshold.

35

Table 7. Matrix of Loadings and Cross-Loadings
Flow
(formative)

Brand
Equity
(reflective)

Purchase
Intention
(reflective)

Brand Social
Media
Engagement
Intention
(reflective)

0.793

0.433

0.248

0.276

0.264

2

0.943

0.699

0.417

0.418

0.502

3

0.938

0.703

0.424

0.418

0.512

4

0.862

0.531

0.290

0.339

0.342

1

0.607

0.917

0.510

0.479

0.610

2

0.588

0.914

0.480

0.464

0.584

3

0.638

0.946

0.531

0.512

0.646

4

0.716

0.922

0.523

0.488

0.642

5

0.625

0.919

0.547

0.525

0.687

1

0.399

0.556

0.975

0.801

0.572

2

0.391

0.543

0.978

0.799

0.558

3

0.388

0.540

0.964

0.767

0.564

1

0.391

0.493

0.763

0.962

0.484

2

0.386

0.488

0.751

0.961

0.468

3

0.395

0.502

0.762

0.960

0.481

4

0.419

0.543

0.794

0.947

0.538

5

0.415

0.535

0.816

0.961

0.551

1

0.532

0.679

0.532

0.520

0.854

2

0.434

0.569

0.434

0.357

0.851

3

0.498

0.583

0.498

0.433

0.910

4

0.484

0.530

0.484

0.403

0.875

5

0.572

0.633

0.572

0.564

0.884

Cognitive
Absorption
(formative)
1

Construct

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intent

Brand Social
Media
Engagement
Intention

Additionally, participants were asked what they though the study was about as a validation
check. A few participants mentioned the research was about how they interacted with brands, but
they did not mention anything specific. No one was able to guess the post popularity manipulation
intent, since they did not know what manipulation group they were in or that others were being
exposed to different manipulations. Thus, we will consider the results of this survey not
compromised by participants guessing the intent of the research.
36

CHAPTER 4
RESULTS
DEMOGRAPHICS
Of the 858 participants, 480 (55.9%) were female, 378 (44.1%) male. A criterion to
complete the study, no participants were under age the age of 18. However, 118 participants
(13.8%) were between the ages of 18 and 23, 202 (23.5%) between 24-29 years old, 190 (22.1%)
between 30-35 years old, 174 (20.3%) between 36-45 years old, 105 (12.2%) between 46-55 years
old, 60 (7%) between 56-65 years old, and 9 (1%) participants were over 66 years old. Only two
participants did not have regular access to the Internet via a computer, tablet, or smartphone
throughout the day. And, participants were also asked to complete a cultural scale to gage their
individual versus collective society traits, as well as an introverted versus extroverted scale to gage
their acceptance of social situations. Distributions of these results are below in Figures 3 and 4.

Individualistic

Collective

Cultural Scale Items

Figure 3. Cultural Scale Distribution of Participants

37

Extrovert

Introvert

Personality Scale Items

Figure 4. Personality Scale Distribution of Participants
Average participant spends 63.49 minutes a day on Facebook, with answers ranging from
1 minute to 720 minutes (12 hours), and the most popular response being 60 minutes. There is a
wide, positively skewed distribution, but the wording of the question does not specify how long
one actively participates/engages on Facebook in a day, thus leading to openness for answers
creating the steady distribution toward such a strange answer as 12 hours. Perhaps several people
leave their browser open to Facebook all day and therefore answered accordingly. The top 25% of
respondents indicated that they spend an hour and a half or more a day on Facebook.
Additionally, the question “Do you currently Like the McDonald’s Facebook Page?” was
used as a pre-exposure question to consider participants prior interactions with the McDonald’s
brand. The question was recoded, with the answers “I don’t think so” and “No” were put into a
group for “No;” while the answers “I think so” and “Yes” were put into a group for “Yes.” As a
comparison, the last question from the Brand Social Media Engagement Intention questions, after
the first post exposure, “Considering this McDonald’s Facebook Page post, how likely are you to
do each of the following?... Like the McDonald’s Facebook Page” was used to see if behavior intent
had possibly changed over the course of the study. This question was recoded with “Very
38

Unlikely,” “Unlikely,” “Somewhat Unlikely” and “Neither Unlikely Nor Likely” being categorized
as “No;” while “Somewhat Likely,” “Likely,” and “Very Likely” being categorized as “Yes.” Using
an ANOVA, these two before and after question areas can be found to have significant mean
differences (F = 26.312, p < 0.001). Similar analysis was conducted on the first Delta Airlines post,
in which significant difference between initial previous brand interactions and post exposure
brand interaction can also be observed (F = 17.877, p < 0.001).

SOCIAL NETWORKING SITES USAGE
Of the 858 participants, 93.8% indicated that they had used a social network site within
the past week. An additional 12 people (1.4%) indicated they do regularly participate in social
media activities but did not during the week prior to completing the survey. Participants averaged
visiting two to three social media sites in the past week with the total application engagement
ranging between zero and six applications.

Table 8. Participant Social Media Activities
Activity

Count

Percentage

Social Network Sites

805

93.8

Content Communities (i.e. YouTube)

577

67.2

Collaborative Projects (i.e. Wikipedia)

295

34.4

Blogs

285

33.2

Virtual Game Worlds (i.e. World of Warcraft)

142

16.6

Virtual Social Worlds (i.e. Second Life)

33

3.8

I do not regularly participate in any of these activities

20

2.3

39

Social Networking Site Engagement Counts

Figure 5. Participant Social Networking Site Activities

When further prodded into their Facebook usage, 153 participants (17.8%) indicated that
they thought they had liked a brand’s Facebook page, while 525 participants (61.2%) were positive
they had. Table 9 breaks down the types of brand pages liked by participants. The average number
of liked pages by participants was just over three pages (3.02). Specific to this study, 83
participants (9.7%) thought they did currently Like the McDonald’s Facebook Page and 91
participants (10.6%) were sure that they had Liked the Page. For Delta’s Facebook Page, 58
participants (6.8%) thought that they did currently like the Delta Page and 64 participants (7.5%)
were sure they had.

40

Table 9. Participant Types of Liked Brand Pages
Page Type

Count

Percentage

Media/television actors
(i.e., Marvin Zindler, Paul Harvey, Mel Gibson, Jack Nicholson, etc)

375

43.7

Television networks/channels
(i.e., local news station)

334

38.9

Sports teams
(i.e., Greenbay Packers, Chicago Bulls, Detroit Red Wings, etc)

300

35

Big Box retailers
(i.e., Walmart, Target, etc)

284

33.1

Creation/Manufacturing
(i.e., artist, Etsy, guitar picks, etc)

279

32.5

Consumer products
(i.e., linens, kitchen supplies, bathroom supplies, etc)

238

27.7

Fast food
(i.e., McDonalds, Arby's, Taco Bell, etc)

232

27

Consumer foods
(i.e., milk, ice cream, bananas, etc)

230

26.8

Services
(i.e., insurance, car repair, contractors, etc)

204

23.8

Airlines
(i.e., Delta, Southwest, United, etc)

116

13.5

Other

81

9.4

I do not Like brand Facebook Pages

135

15.7

Facebook Brand Page Like Counts

Figure 6. Participant Facebook Brand Page Likes

41

Finally, to glimpse into participant brand Page activity, participants were asked what types
of brand Facebook Pages they had visited in the past week. The average number of pages visited
by participants was more than one (1.31). Specifically for this study, 26 participants (3%) thought
they had visited the McDonald’s Facebook Page within the last week and 55 participants (6.4%)
were sure that they had. For Delta’s Facebook Page, 33 participants (3.8%) thought that they had
visited the Facebook Page within the last week and 46 participants (5.4%) were sure that they had.
Table 10 breaks down the types of brand pages visited by participants. Additionally, 224 people
(26.1%) indicated they do regularly visit brand Facebook pages but did not this past week.

Table 10. Participant Facebook Brand Page Visits
Page Type

Count

Percentage

Television networks/channels
(i.e., local news station)

172

20

Media/television actors
(i.e., Marvin Zindler, Paul Harvey, Mel Gibson, Jack Nicholson, etc)

151

17.6

Big Box retailers
(i.e., Walmart, Target, etc)

125

14.6

Fast food
(i.e., McDonald’s, Arby's, Taco Bell, etc)

116

13.5

Sports teams
(i.e., Greenbay Packers, Chicago Bulls, Detroit Red Wings, etc)

114

13.3

Creation/Manufacturing
(i.e., artist, Etsy, guitar picks, etc)

106

12.4

Consumer products
(i.e., linens, kitchen supplies, bathroom supplies, etc)

101

11.8

Services
(i.e., insurance, car repair, contractors, etc)

89

10.4

Consumer foods
(i.e., milk, ice cream, bananas, etc)

86

10

Airlines
(i.e., Delta, Southwest, United, etc)

62

7.2

Other

49

5.7

I do not visit brand Pages on Facebook

127

14.8

42

Facebook Brand Page Visit Counts

Figure 7. Participant Facebook Brand Page Visits

HYPOTHESIS TESTING
The research model shown in Figure 2 was tested using the variance-based PLS method.
PLS allowed specifying the relationships between the various endogenous and exogenous
constructs in the model (structural model), as well as with their underlying items (measurement
model). Thus, data analysis provided support for both how well the items measured each
construct, and how well the hypothesized relationships between constructs supported the
proposed theory.
The bootstrapping re-sampling method with 5,125 cases and 500 re-samples was used for
structural model estimation. Based on the results of the model estimation, only hypotheses 2, 3,
and 4a were supported (see Table 11). The results of the structural model estimations, including
R2 values, are presented in Figure 8.

43

Table 11. Hypothesis Testing
Path
Coefficient

t-Value

Status

Brand Equity

-0.017

1.448

Rejected

Brand Equity

Purchase Intention

0.812

118.759*

Supported

3

Brand Equity

Brand Social Media
Engagement Intention

0.580

64.312**

Supported

4a

Post Popularity

Brand Equity

-0.024

2.157*

Supported

4b

Post Popularity

Purchase Intention

0.000

0.013

Rejected

4c

Post Popularity

Brand Social Media
Engagement Intention

0.010

0.884

Rejected

Hypothesis

From

To

1

Engaging Brand
Content

2

* significant at 0.01 level; ** significant at 0.001 level

* significant at 0.01 level; ** significant at 0.001 level

Figure 8. PLS Model
To further analyze the model, participant responses were broken out into their
manipulation areas. By viewing the model this way, significance can be seen on almost all paths.
The exceptions are the path connecting Cognitive Absorption to Brand Equity (β = 0.014, p =
0.644) as well as the path connecting Engaging Brand Content to Brand Equity (β = 0.030, p =

44

0.132) when real post popularity metrics were shown. Similarly, no significance was observed for
the same two paths, from Cognitive Absorption to Brand Equity (β = 0.001, p = 0.973) and
Engaging Brand Content to Brand Equity (β = 0.013, p = 0.537) when post popularity metrics
were shown to the participants switched (established high engagement posts shown with low
engagement metrics and vice versa). For factor loadings, construct statistics, and latent variable
correlations for these manipulations, please see Appendices C, D, and E, respectively.

* significant at 0.01 level; ** significant at 0.001 level

Figure 9. PLS Model with No Post Popularity Metrics Shown

* significant at 0.01 level; ** significant at 0.001 level

Figure 10. PLS Model with Post Popularity Metrics Matching Brand Content
45

* significant at 0.01 level; ** significant at 0.001 level

Figure 11. PLS Model with Post Popularity Metrics Opposite Brand Content

Additionally, due to the significance of both Cognitive Absorption and Flow as
antecedents of brand equity in the model in which participants were not shown any post
popularity metrics what so ever (Figure 9 above), to look at the amount of unique variance
explained by each predictor, a stepwise hierarchical regression was conducted using the latent
variable loadings from the PLS output to ensure model consistency (see Table 13). In the first
approach, determinants were listed in the theoretical order as specified by the structural model,
that is, Cognitive Absorption followed by Engaging Brand Content and then followed by Flow.
As a result, this model, referred to in Table 13 as the theoretical model, attributes the (majority
of the) covariance between determinants to the first and third variables in the model, that is,
Cognitive Absorption and Flow. The R2 change for both steps in the model was significant
(p < 0.001).
In the second approach, the variables were added in the order of the statistical
significance of the predictors: Engaging Brand Content, Flow, and then Cognitive Absorption. As
a result, this model, refer to in Table 13 as Empirical Model 1, attributes the (majority of the)
covariance between determinants to Flow with significant R2 change (p < 0.001). And finally in
46

the third approach completing the steps, the variables were added in the last combination order:
Flow, Cognitive Absorption, and then Engaging Brand Content. This model, refer to in Table 13
as Empirical Model 2, again attributes the (majority of the) covariance between determinants to
Flow with significant R2 change (p < 0.001).
Comparison between the three methods as well as an average across the models is
provided in Table 12 below. The average variance reflects the unique variance of each antecedent
relative to the total explained variance of 0.367. This implies that the shared variance or
covariance between the Cognitive Absorption, Engaging Brand Content, and Flow dimensions is
over 56% (0.564).

Table 12. Stepwise Linear Regression for R2 Partitioning
Dependent Variable: Brand Equity

Antecedent

Theoretical
Approach

Empirical
Approach 1

Empirical
Approach 2

Average
(Unique
Variance)

Cognitive
Absorption

0.200**

0.004*

0.003*

0.069

Engaging
Brand Content

0.003*

0.001

0.003*

0.002

Flow

0.164**

0.364**

0.361**

0.296**

* significant at 0.01 level; ** significant at 0.001 level

Following the result presented in Table 13 above, we can conclude that the explained variance of
the integrated model aggregating the Cognitive Absorption, Engaging Brand Content, and Flow
dimensions is 36.7% (adjusted R2 is 36.6%). Even though the explained variance of the unified
model may not appear much higher than the Flow Empirical Approach 1 model alone (R2 is
36.4%) or the Cognitive Absorption Theoretical model alone (R2 is 20.0%), the results from the
stepwise regression analysis in Table 13 show that the difference in explained variance (ΔR2) as
a result of amalgamating the two types of performance is highly significant for Flow as a
contributing antecedent to Brand Equity (p < .001).
47

To double check the lack of significance between Post Popularity Metric manipulations,
ANOVAs were run between the means of each manipulation, factoring by the results of the
exposure with no manipulation as well as the results of the switched metric manipulation. See
these results in Table 13.

Table 13. ANOVAs for Post Popularity Metric Exposure Differences Compared
with True Metrics Being Shown
Construct

Manipulation

F Score

P-Value

Status

Cognitive Absorption

No Metrics

0.596

0.939

Not Significant

Switched Metrics

0.688

0.868

Not Significant

No Metrics

1.129

0.288

Not Significant

Switched Metrics

1.417

0.067

Not Significant

No Metrics

1.549

0.065

Not Significant

Switched Metrics

0.662

0.851

Not Significant

No Metrics

1.324

0.113

Not Significant

Switched Metrics

1.890

0.003

Significant

No Metrics

0.904

0.617

Not Significant

Switched Metrics

1.000

0.466

Not Significant

Flow
Brand Equity
Purchase Intention
Brand Social Media
Engagement Intention

With the lack of support for Hypothesis 4 and any community engagement metric effects
at the granular variable level, the model was run again, this time looking at interaction effects for
this construct. Using this model, significance was found (p = 0.010) if considering the
manipulation of post popularity metrics as a moderating variable on the relationship between
Engaging Brand Content and Brand Equity.

48

* significant at 0.01 level; ** significant at 0.001 level

Figure 12. PLS Model with Post Popularity as a Moderator

Finally, with the lack of support for all of Hypothesis 4 and post popularity metric effects
at the granular variable level for neither Purchase Intention nor Brand Social Media Engagement
Intention, the model was run once again; this time looking at only the metrics from the first two
posts shown to participants, to capture any effects with initial responses for one Brand Awareness
post for each brand.
This did not help establish post popularity metrics as an effect in the model. By viewing
the model this way, significance was only found on the same three paths for both brands. For
McDonald’s first Brand Awareness Post: Brand Equity to Purchase Intention (β = 0.765, p =
0.000), Brand Equity to Brand Social Media Engagement Intention (β = 0.610, p = 0.000), and
Flow to Brand Equity (β = 0.606, p = 0.000). For Delta Airline’s first Brand Awareness Post:
Brand Equity to Purchase Intention (β = 0.753, p = 0.000), Brand Equity to Brand Social Media
Engagement Intention (β = 0.571, p = 0.000), and Flow to Brand Equity (β = 0.502, p = 0.000).

49

* significant at 0.01 level; ** significant at 0.001 level

Figure 13. PLS Model, Brand Awareness McDonald’s Post

* significant at 0.01 level; ** significant at 0.001 level

Figure 14. PLS Model, Brand Awareness Delta Airlines Post

50

With the one present significant effect of Post Popularity as a an interaction construct on
Engaging Brand Content and Brand Equity, one final model test was run to see if perhaps the Post
Popularity Manipulation causes an effect on Brand Equity antecedents, Cognitive Absorption and
Flow. Using the Brand Awareness posts for both brands, there was a significant effect found on
the relationship from Post Popularity to Flow for McDonald’s (β = 0.084, p = 0.017) and Delta (β
= 0.085, p = 0.009).
And finally, the idea was entertained that perhaps the Post Popularity manipulation could
have an effect on only high engagement posts or only low engagement posts. But alas, neither of
these isolations showed any significant effects for a Post Popularity relationship.

POST-HOC TEST FOR BRAND EQUITY, PURCHASE INTENTION,
AND BRAND SOCIAL MEDIA ENGAGEMENT INTENTION EFFECTS
Since the model was built, the Brand Social Media Engagement and Purchase Intention
variables were tested for effects. While no literature pointed to either of these effects being a
possibility due to the introduction of Brand Social Media Engagement Intention as a relatively
new construct, one could reason that an effect is possible. A consumer could be trying to reinforce
their purchase decision by commenting on it in social media or the reverse direction, social media
interaction could spawn purchase intent. Using the main model with community engagement as
a reflective variable, it can be seen that Brand Social Media Engagement Intention does have an
effect on Purchase Intention (t = 8.743, p = 0.000). Further research and discovery is needed to
fully understand why these relationships exist and to understand the full impact on Brand Social
Media Engagement Intention on brand and purchasing constructs.

FURTHER EXPLORATORY INQUIRES
Due to the lack of significance for most of the original hypotheses, further inquiries were
conducted using and modifying the proposed theoretical model. Using the Multi-Group Analysis
51

(MGA) feature in Smart PLS, the following investigations were conducted with recoded
manipulations to see if this could show some sort of Post Popularity manipulation affect. For this
new coding, manipulation categories were no Post Popularity metrics shown, low metrics shown,
and high metrics shown. With this new idea, the point of the manipulation recoding was to see if
Post Popularity metrics had anything to do with the engagement, or if it was the content that
caused people to interact and show higher Brand Equity and Purchase Intent towards the brand.
Therefore, showing low or high metrics now includes the posts that had the metrics switched. So
the low category would represent the original low engagement post (as previously categorized as
“real” metrics showing) as well as the high engagement post (which would have previously been
categorized as “switched” metrics showing).
First, using all posts, Post Popularity data groups were created and run on a clean model,
with no interactors or moderators just to see where the groups would lead. For exploratory
purposes, for this model, only 15% of the data was used for this Smart PLS multi-group analysis.
For this manipulation, no significance could be seen between the groups. Overall bootstrapping
with all of the data showed significance except for the post engagement level path to cognitive flow
(p = 0.109). Using the multi-group analysis function in Smart PLS, the R2 means and path
coefficients can be seen in Table 14 and Figure 15. Significance between the groups can only been
seen between the group that was shown no metrics versus the group that was shown low metrics
(p = 0.003), or the group the was shown high metrics (p = 0.044), both for the path from Brand
Equity to Brand Social Media Engagement Intention (See Table 14 & Figure 15).

52

Table 14. All Posts, Post Popularity Groups,
MGA Community Metric R2 Means
Metrics Shown
Construct

None

Low

High

Cognitive Absorption

0.011

0.031

0.022

Flow

0.005

0.021

0.004

Brand Equity

0.409**

0.282**

0.295**

Purchase Intention

0.624**

0.710**

0.685**

Brand Social Media
Engagement Intention

0.405**

0.239**

0.291**

* significant at 0.01 level; ** significant at 0.001 level

MGA βΔ = none:low,
none:high
* significant at 0.01 level;
** significant at 0.001 level

Figure 15. All Posts, MGA Post Popularity Metric Path Differences

For the second idea, Post Popularity data groups were again used, but this time it was also
included as an independent variable affecting Cognitive Absorption, Flow, and Brand Equity.

53

Additionally, while the model was built, Post Popularity as a moderator on the relationship
between Cognitive Absorption and Brand Equity and Flow and Brand Equity were considered
again with only 15% of the data. While this idea showed significance in the clean model but not
along any paths that involved Post Popularity, the findings were replicated here, and no significant
effects were found surrounding the Post Popularity metrics as either an independent variable or
as a moderating effect. And ultimately, overall, this did not work as it caused a single matrix error
and could not run the group comparisons.
For the next idea, the previous idea was replicated using engagement data groups. For this
metric, the question “Do you currently Like the McDonald’s Facebook Page?” was recoded. The
answers “I don’t think so” and “No” were put into a group for “No,” while the answers “I think so”
and “Yes” were put into a group for “Yes.” Again, Post Popularity was included as an independent
variable affecting Cognitive Absorption, Flow, and Brand Equity; and Post Popularity was used as
a moderator between Cognitive Absorption and Brand Equity and Flow and Brand Equity, and
only 15% of the data was used (See Table 15 & Figure 16). This idea leads to the conclusion that
Post Popularity is not a moderator and does not impact Brand Equity, Cognitive Absorption, or
Flow whether a post is showing low or high Post Popularity metrics.
Table 15. All Posts, Engagement Groups,
MGA R2 Means
McDonald’s Page Liked
Construct

No

Yes

Cognitive Absorption

0.003

0.005

Flow

0.005

0.003

Brand Equity

0.335**

0.333**

Purchase Intention

0.671**

0.666**

Brand Social Media
Engagement Intention

0.330**

0.304**

* significant at 0.01 level; ** significant at 0.001 level

54

MGA βΔ = no:yes
* significant at 0.01 level;
** significant at 0.001 level

Figure 16. All Posts, Engagement Groups, MGA Path Differences

Then the idea was entertained that perhaps the purchase decision was too great for Delta
Airlines posts to be accurately incorporated into the model, so why not take these posts out and
just considers the very first Brand Awareness post from McDonald’s. Since the Culture scale was
included and the results already showed a normal distribution, this was used as a grouping factor
and Post Popularity as a moderator again for Cognitive Absorption, Flow, and Brand Equity.
Again, this idea does not produce significant relationships, and therefore leads to the conclusion
that Post Popularity is not a moderator impacting the relationship from engagement to Cognitive
Absorption, Flow, or Brand Equity, and there are no significant differences between participants
who were more individualistic or collectivist in their group interaction preferences.

55

Table 16. McDonald’s Brand Awareness Post,
Culture Groups, MGA R2 Means
Culture Groups
Construct

Individualist

Collectivist

Cognitive Absorption

0.015

0.013

Flow

0.015

0.015

Brand Equity

0.311**

0.343**

Purchase Intention

0.528**

0.608**

Brand Social Media
Engagement Intention

0.286**

0.385**

* significant at 0.01 level; ** significant at 0.001 level

MGA βΔ = Individualist:Collectivist
* significant at 0.01 level;
** significant at 0.001 level

Figure 17. McDonald’s Brand Awareness Post,
Culture Groups, MGA Path Differences

56

Finally, the idea was entertained, since all of the elements were present, to consider
engagement groups, and let Culture be a moderator, and Post Popularity an independent variable,
on only the McDonald’s first Brand Awareness posts.
Table 17. McDonald’s Brand Awareness Post,
Engagement Groups, MGA R2 Means
McDonald’s Page Liked
Construct

No

Yes

Cognitive Absorption

0.002

0.002

Flow

0.003

0.003

Brand Equity

0.234**

0.439**

Purchase Intention

0.529**

0.608**

Brand Social Media
Engagement Intention

0.329**

0.392**

* significant at 0.01 level; ** significant at 0.001 level

MGA βΔ = no:yes
* significant at 0.01 level;
** significant at 0.001 level

Figure 18. McDonald’s Brand Awareness Post,
Engagement Groups, MGA Path Differences

57

This led to the conclusion that Culture is not a moderator, and Post Popularity does not have a
significant effect on Brand Equity or its antecedent variables considered in this study.

SIMPLIFIED MODEL
When all else fails, perhaps the model should have been backed up and looked at piece by
piece to begin with. For one final idea and attempt to finding something, the Culture variable was
set aside. The model was broken into two simpler models: one for just the Brand Equity and its
antecedents, and one for engagement levels of the posts, Brand Equity, and Purchase Intent and
Brand Social Media Engagement Intent. This latter model is similar to the prelim study model,
with the exception of Brand Attitude being exchanged for Brand Equity.
Because the idea was entertained earlier that perhaps flight purchases were too high of a
purchase decision to expect any change of brand engagement based on one post, only the
McDonald’s three posts were used for the simplified model. Sample size remains the same
(n = 858), but the number of exposures is now three per person as opposed to six. To double check
that there is significance between initial interactions with McDonald’s brand and after all of the
post exposures, another ANOVA test was run to compare the means between these two groups.
Again significance was found (F = 213.316, p < 0.001).
For the first simplified model, there was a significant relationship between Engaging
Brand Content and Brand Equity when low (p = 0.015) or high (p = 0.049) metrics were shown
to participants. When looking at these two groups together (the third set for Δβ in Figure 19
below), when participants were shown low Community metrics there was a stronger path
coefficient (0.081), but when shown high metrics, there was a negative relationship (path
coefficient = -0.066). Additionally, for the relationships from Brand Equity to Purchase Intention
and from Brand Equity to Brand Social Media Engagement Intention, the paths are significant
across all groups (p < 0.001).

58

Table 18. McDonald’s Posts, Post Popularity Groups,
MGA Community Metric R2 Means for Simplified
Model
Metrics Shown
Construct

None

Low

High

Brand Equity

0.003

0.007

0.005

Purchase Intention

0.684**

0.681**

0.642**

Brand Social Media
Engagement Intention

0.383**

0.359**

0.369**

* significant at 0.01 level; ** significant at 0.001 level

MGA βΔ = none:low,
none:high,
low:high
* significant at 0.01 level;
** significant at 0.001 level

Figure 19. McDonald’s Posts, MGA Post Popularity
Metric Path Differences for Simplified Model

For the second simplified model (see Table 19 & Figure 20 below), no significant
relationships were found from Engaging Brand Content to Brand Equity. When no Post Popularity
metrics were shown versus low Post Popularity metrics being shown to participants, there is
59

significant positive relationship between Cognitive Absorption and Brand Equity. When looking
at these two groups together (the third set for Δβ in Figure 20 below), when participants were
shown no Community metrics there was a stronger path coefficient (0.075), but when shown low
Community metrics, there was again a negative relationship (path coefficient = -0.053). Although
not significant, this positive to negative relationship also holds true when considering no Post
Popularity metrics being showed compared to high Post Popularity metrics being show (path
coefficient = -0.003). Paths from Engaging Brand Content to Cognitive Absorption and from
Engaging Brand Content to Flow show significance when low Post Popularity metrics were shown
to participants (p = 0.002 and p = 0.044, respectively). Additionally, the path from Cognitive
Absorption to Brand Equity is significant when no Post Popularity metrics are shown (p = 0.034).
Interestingly, all paths from Flow to Brand Equity show significance (p < 0.001 for all). Alas, the
results are all over the place and confusing, and little can be found in relation to the Post
Popularity metrics and participant reactions to them, which leaves one to wonder why they are
included on Facebook posts to begin with.

Table 19. McDonald’s Posts, Post Popularity
Groups, MGA Community Metric R2 Means for
Brand Equity Antecedents
Metrics Shown
Construct

None

Low

High

Cognitive Absorption

0.002

0.012

0.005

Flow

0.003

0.006

0.003

0.445**

0.288**

0.342**

Brand Equity

* significant at 0.01 level; ** significant at 0.001 level

60

MGA βΔ = none:low,
none:high,
low:high
* significant at 0.01 level;
** significant at 0.001 level

Figure 20. McDonald’s Posts, MGA Post Popularity Metric
Path Differences for Brand Equity Antecedents

Additionally, it is interesting to note that if isolating the groups that initially showed no
engagement with the McDonald’s Brand from those that had previously Liked the McDonald’s
brand page, there is a significant difference, as mentioned earlier. However, it is also interesting
to note that when considering the first simplified model (see Figure 19), there is a significant
difference between these two groups from Brand Equity to Purchase Intent (p < 0.001), with the
initially no interaction with the brand Facebook Page having a stronger path coefficient 0.080
higher than those who had previously Liked the McDonald’s Brand Page on Facebook.

61

SUMMARY
After considering the different models investigated, only the simplified model in Figure
19. McDonald’s Posts, MGA Post Popularity Metric Path Differences for Simplified Model, which
matches the comparison research in Coursaris, et al, 2014a, is able to show an adequate
Standardized Root Mean Square Residuals value (SRMR = 0.065). Hu & Bentler’s 1999 study
recommends a threshold of 0.10 or of 0.08 to be considered a good fit. All other models had SRMR
values over 0.196. The rest of the Goodness of Fit metrics can be seen in Appendix H.

62

CHAPTER 5
DISCUSSION
Industry continually strives to quantify their work. A company’s return on investment is a
vital output for understanding the perplexities of investing in employee work tasks to generate
consumer interaction with the brand, on a virtually free platform at that, and show if there is any
value to the work being done to begin with. Building on previous research, this study used a
simulated Facebook environment combined with survey response to further develop
understanding of Facebook engagement metrics, consumer interactions, and the branding effects
of social media. Current industry emphasis on qualifying the value proposition of social media
activities and quantifying the return on investment in such activities provided the motivation for
this study.
In terms of significant findings, these were only found when considering the relationship
between Brand Equity and Purchase Intent (t = 118.759, p < 0.01), as well as Brand Equity and
Brand Social Media Engagement Intent (t = 64.312, p < 0.001). Now validated on a general
population sample, these results were initially also found in the Coursaris, et al. 2014a study. Due
to the length of the experimental survey, analyses were also run on only the first Brand Awareness
posts for both brands in an attempt to see if there was any significance before any participant
fatigue, if there was any, could be present. These findings held true both for the full study
observations, as well as when only the first two posts were examined.
In terms of the research question of what mechanisms account for the link between brand
equity and engaging content, flow was established as a significant predictor of brand equity (β =
0.541, p < 0.001). These findings also held true when only the first two posts were analyzed.
When the model is broken down and each component analyzed by its manipulation group,
all paths in the model can be observed as significant with the exception of the cognitive absorption
dimension and its impact on brand equity. Cognitive absorption was only found to have statistical

63

significance in the relationship with brand equity in the treatment group that was shown no Post
Popularity engagement metrics. Interestingly, only when no Post Popularity engagement metrics
are shown to participants is there also a significant relationship from Engaging Brand Content to
Brand Equity.
Similarly, Post Popularity metric manipulations cause no statistically significant effects on
either of the dependent variables as a moderator for Brand Equity on Purchase Intention (t =
0.013) or Brand Equity on Brand Social Media Engagement Intention (t = 0.884). However, Post
Popularity metrics do act as an interaction construct on Engaging Brand Content and Brand
Equity (β = 0.029, p < 0.01). So while community does not appear to have an effect on the possibly
tangible outcomes of consumer interaction and purchase intentions, it does possibly affect
whether or not engaging brand content has an effect on brand equity. Since community/ Post
Popularity determines what is engaging brand content, this interaction essentially makes sense.
From here, a final test was run to find that Post Popularity metrics have a statistically significant
effect on flow directly for both McDonald’s (β = 0.084, p = 0.017) and Delta (β = 0.085, p = 0.009)
when looking at just the first two Brand Awareness posts across all manipulation groups. Clearly
Flow is a significant construct and variable when looking at social media influences and behaviors
which warrants further researcher and discussion.
Because it is a relatively new metric, no research exists as to why a relationship between
Brand Social Media Engagement Intention and Purchase Intention exists (t = 8.743, p = 0.000)
and vice versa (t = 9.808, p = 0.000). Huang’s 2011 study showed that online involvement has a
moderating effect on purchase intent, but this is not enough to explain intention in a way that
could lead to understanding the return on investment for social media. Additionally, it was already
established that Brand Equity has a statistically significant effect on Brand Social Media
Engagement Intention (t = 64.312, p < 0.001), it turns out this vice versa relationship is also
significant (t = 25.000, p = 0.000). So clearly, there are other factors impacting this Social Media

64

Engagement Intention metric which are outside the scope of this study but would be fascinating
to learn more about what is going on and impacting this measure.
While brands increasingly rely on Facebook provided metrics – such as post Likes,
Comments, Shares, the number of Fans, the number of Followers, etc. – such metrics have yet to
be effectively validated. This study actually tested the effectiveness of these metrics by
manipulating the participant’s exposure to certain engagement metrics and posts, while observing
their interaction with the posts and engagement intent. To follow is a discussion of the study
results and setup, as well as possible future research areas to enhance future environment
simulations and studies in this area.

LIMITATIONS
While it is disappointing that this study was not able to establish Post Popularity metrics
being exposed or note exposed as a significant factor in consumer brand relations, several updates
to the study parameters might help future research to try and understand what is really going on
in the social media world and, therefore, this mode. Limitations can be found in the demographics
area, including the participants, namely, Mechanical Turk workers; the dimensions used for
surveying participants on the cognitive absorption construct; as well as the selected brands and
their relevance to participants. The discussion of these areas is to follow.
Given the length of the experimental survey, future research could either break the study
into smaller sections, perhaps testing just one piece of the Buying Behavior Model (Consider,
Search, Choose, and Buy) at a time and ensuring full understanding of consumer effects, or
building a longitudinal study to break the responses up over time (Wind, 1978). As it currently
stands, perhaps the results from Coursaris, et al. 2014 were a stronger representation of consumer
brand effects. That study was able to establish that prolonged exposure to high engagement brand
posts does show an increase in brand attitude (Coursaris, et al., 2014).

65

DEMOGRAPHICS & PARTICIPANTS
Having no baseline measurements for Brand Equity or Purchase Intent creates an area
that cannot be analyzed with these results. With no established baseline in these variable areas,
there is no way to tell if participant Brand Equity and/or Purchase Intent changed during the
course of the experiment, after multiple post exposures for each brand. This decreases some of
the replicability to allow this study to be compared to the Coursaris, et al., 2014 results which were
able to establish prolonged exposure to brand messaging as an indicator of increased brand
attitude and purchase intention. This leaves an opportunity for a future study.
This research was originally proposed with a collective versus individualistic cultural
component as part of the analysis, which is why these scales were included in the survey
questionnaire. Using this scale, since there was a normal distribution of these individuals
throughout the 858 participants data used, initial analyses were run to see if perhaps there was
such a individualistic versus collective society effect, while incorporating the Post Popularity
manipulations. Much to the researcher’s surprise, at least for this United States data from Amazon
Mechanical Turk workers, there is not a significant difference between these two cultural groups.
As soon as the Italian data is available for analysis, the comparisons will begin between the two
countries to see if perhaps a more collective society is a stronger reflection of cultural differences
as well as to see if they are more susceptible to Post Popularity metric exposures.
However, while there appeared to be a normal distribution in the cultural and personality
scale areas for all participants, this turns out to not be necessarily true when looking at these scales
by manipulation group. The same can unfortunately also be said for the overall participant age
distributions in each manipulation group. Demographic distribution break downs by age in each
of these areas can be seen in Appendix G. Significant differences can be found between the group
that was shown no Post Popularity Metrics and the group that was shown true Post Popularity
Metrics (high values for high elicited engagement posts, low values for low elicited engagement
posts).
66

While it was exciting to use a general population sample for this research, one thing that
might be considered a down fall of using Mechanical Turk participants is that they could be looked
at as completing the survey just for the money. But in fact, in a 2010 study by Paolacci, Chandler,
& Ipeirotis, only 13.8% of United States based workers report Mechanical Turk as their primary
source of income, 61.4% said that earning additional income was an important driver for
participation, and 69.6% of respondents consider Mechanical Turk a fruitful way to spend their
extra time. Still, as mentioned earlier, attrition can be higher for Mechanical Turk studies, and
researchers must be careful to compensate correctly to value participants time (Rand, 2011).
Ultimately, it is a “bang for your buck” type system. Participants want to complete the most
Human Intelligence Tasks (HIT), in the shortest amount of time, in an attempt to make the most
money.
Consequently, this could have contributed two reasons that there might have been a
possible “fatigue” in this study. First, perhaps participants did not think they were getting
adequately compensated. To this point, participants were compensated $1.10 for participation,
which at the time, was considered more than fair. Second, perhaps participants could have been
looking to move on to the next HIT sooner rather than later, in which case they were breezing
through the study, not giving the questions the time they deserved. Rader & Gray’s 2015 study
mentions that Amazon Mechanical Turk workers might be more attentive toward things in their
News Feed. Maybe participants just knew they would not proliferate the content put before them
in this study. Without further metrics and/or interviews to investigate, conclusions cannot be
made at this time.

COGNITIVE ABSORPTION & BRAND EQUITY
Additionally, cognitive engagement was not measured in this study. With several
Facebook engagement metrics already built into this research to validate Coursaris, et al’s (2016)
Weighted Engagement formula, an additional one was not added. However, by omitting cognitive
67

engagement as a specific measure included in the survey, this research lacks measuring the
enjoyment aspect of the cognitive absorption construct. While the state of playfulness is
considered identical to the flow experience according to Agarwal & Karahanna’s 2000 study,
Webster & Ho’s 1997 study emphasized the enjoyment aspect of cognitive engagement as the flow
experience minus the control aspect. It could basically be considered more free form
participation/enjoyment. Engagement is also considered multidimensional and to encompass the
intrinsic interest, curiosity, and focus in the five dimension conceptualization of the cognitive
absorption construct (Agarwal & Karahanna, 2000). In Agarwal & Karahanna’s 2000 study,
cognitive engagement was positively affected by perceived media richness, interactivity of
teaching style, and classmates’ attitudes towards the technology being used. Essentially, not
including cognitive engagement scales in this research was an oversight, which will be corrected
in future research to increase validation of the new metrics and strengthen the investigation into
the Cognitive Absorption construct.
Also it is important to note, that upon further consideration of the Chang & Liu (2009)
Brand Equity scale, perhaps this wasn’t the best choice. The scale was chosen as a shorter, reliable
way to investigate brand equity. However, while the questions touch on the areas that comprise
brand equity, they do not fully encompass the scale items that are traditionally used to measure
brand equity. Therefore, for this study, brand equity cannot measure all of the different facets that
traditionally make up brand equity (i.e. brand attitude, brand image; Faircloth, Capella, and
Alford, 2001). This therefore hinders this study’s ability to further investigate what might be going
on with the lack of significance areas since there is an inability to breakdown the components
from brand equity, which is usually treated as a second order construct.

SELECTED BRAND RELEVANCE & SIMULATED ENVIRONMENT
Additional explanation of possible variance for the constructs can be attributed to
environmental factors outside of the focus of this study. This study concentrated on proactive,
68

brand initiated communications and posts selected for the engaging brand content stimulus
condition of this study were hand-picked to be positive in nature, thus essentially sanitizing the
data against negative response. Should a participant have a preexisting negative attitude towards
a brand, this was not accommodated for. This study did not look at increasing loyalty or spending
over time, but if it had, consumers who are shown to have already engaged with a brand might
show an kinship towards that brand and therefore no change would be observed, and vice versa
(Baird & Parasnis, 2011).
Similarly, while the brands picked for this study were done so to provide the largest variety
of purchase making decisions possible for analysis, this study did not consider all of the possible
factors that the realism of the brands could cause. For example, one participant voluntarily
submitted email feedback after completing the survey, in which she indicated that her “sole
decision maker when I fly is the cost:”
I wish there had been places to comment why I answered some of the questions
the way I did. For example, I fly often, but my decision maker on what airline I use
is price and only price. Therefore while I would be happy to use Delta, if United
had a better price, I am not going to think twice about using Delta.
– Michelle
Michelle also indicated that her responses to the survey were based on the Delta brand in front of
her. Therefore her results would align with the intent of the study. But while a great idea, and
using pre-tested posts for engagement positioning, perhaps the specific brands should have been
disguised under dummy brands. This would have allowed the study to maintain the purchase
decision making objectives of looking at consumer engagement intents of low purchase
involvement fast moving consumer goods versus airline tickets which generally require more
investment in high purchase involvement decisions, have regional considerations, and could have
travel frequency implications on the results. However, the definition of brand equity as the brand
name’s added attractiveness to the consumer for the product or services, as defined by Chang &
Liu (2009) was used for this study (see Table 1). This definition possibly removes the ability to
use a dummy brand name since so much emphasis was potentially put on the brand name in this
69

research to impact brand equity. Ultimately, perhaps there are just too many external factors that
cannot be controlled for to effectively use airlines posts in conjunction with fast moving consumer
goods posts. Ultimately, similar posts were used to compare results to previous research, but this
small tweak of the brand and industry selection might be something to consider for future
research.
Additionally, in the sanitization of the data, while the metrics were kept on the brand posts
in the manipulations that were shown these numbers, all comments were removed from the post.
In doing this, the posts were neutralized to one aspect of the user-generated content that might
cause participants to want to interact with the post. As referenced earlier, Bruhn, Schoenmueller,
& Schafer’s 2012 study also showed that user-generated content on social media platforms has a
greater influence on brand awareness and brand image in the tourist industry. While the removal
of these comments would help participants conclude for themselves if they would like to interact
with the post without peer influence, something for another study, the lack of including them
could have affected the results in testing overall brand equity.
It was also considered, that perhaps the single exposure setting of this study was not as
conducive to analyzing large purchase decision products, such as Delta. For such an investment
in time and money for people to consider making, three post exposures in one sitting isn’t going
to change a purchase decision right then and there; especially when there is established feedback
that cost is too important of a factor when making these types of decisions. Perhaps a longitudinal
study, and/or spaced out exposure to brand posts like in the Coursaris, et al. 2014a study could
provide better results in this area.
Also interesting to note, are the types of social networking sites and usages the participants
indicated. Only 3.8% and 16.6% of participants engage in Virtual Social Worlds (i.e. Second Life)
and Virtual Game Worlds (i.e. World of Warcraft), respectfully. Considering that these are
possibly areas with higher cognitive absorption and/or flow possibilities, one has to wonder if
perhaps this research should have worked harder to target participants in these areas to see what
70

kinds of effects that group produces. Similarly, this research looked at posts from the Fast Food
and Airlines Facebook Page categories, of which the sample population had only 27% and 13.5%,
respectively, of participants indicating that they had Liked these pages. Due to the time of year
the study was conducted, in the summer, not during any type of sport season start, such as
baseball, it could be suggested that sports were not a “top of mind” consideration for people to be
thinking of, yet 35% of participants indicated that they had Liked a sports team page, making it
the third highest preference indicated. However, looking at these results, perhaps this research
should have looked at Media/Television Actors or posts from Television network pages. This
would have referenced 82.6% of respondents and perhaps would have been more immediately
relatable when talking about community effects. Not to mention, when discussing a community
of people-type environment with participants who are people, this would have been posts by a
brand that were also about a realm of people. Just an idea; it is not sure if this disconnect had
anything to do with the results.
Considering these results, which are summarized earlier in Table 9, future research could
test this research model on media/television actor, television networks/channel, sports teams,
and/or big box retailers Page posts to see if conclusive results for post popularity metric
manipulations could be observed. Ironically, Facebook Page Posts by Walmart were part of the
original analysis for comparison with this study, but they were removed with the idea that keeping
purchasing decision extremes (McDonald’s and Delta Airlines) would produce better results.
Ideally, an adaptive environment would have suited this type of study better. By asking
participants what pages they have previously liked on Facebook, the survey questions could adapt
what brands are displayed to the participants. This would increase the relevancy to them, and
possibly generate more interaction and/or Engagement Intent with the posts displayed before
them.
On seventeen posts, fifteen participants clicked Like and then Unlike. While not
statistically significant by any means, it was interesting to see that apparently two participants are
71

either forgetful or tend to change their mind, and nine of the seventeen posts were ones with the
metrics switched (five high engagement posts showing low engagement metrics, four low
engagement posts showing high engagement metrics). Five of these answer changed received
comments, which are best summarized by one of the participants who received the switched
metric manipulation: “Yeah, no.”
That being said, perhaps environmental updates could have enhanced the experience with
this experimental survey platform for participants. One suggestion would be to create an adaptive
environment that asks more upfront questions to the participants. Using their response, the
simulated page would update to their preferences. This would allow areas of the simulation to be
more relevant to them, and possibly increase interaction with the created Facebook Page.
Included in these questions could be to name a few of their friends which would then allow for
displayed posts to not only show Post Popularity metrics, but to also incorporate friend names to
help build the immediacy of the peer interaction with the posts.
Another update to the experiment would be to create a lab experiment in which
participants were invited to more of a usability study environment, and a researcher would walk
them through using the simulated environment. This would allow for an in person confirmation
that their interactions with the tested posts absolutely do not connect to their personal Facebook
account, and therefore can be interacted with freely; if that concern was part of the problem. The
environment itself was not usability studied for this research project; perhaps there is something
in the interface that could have been updated to facilitate more self-interaction with the posts by
participants. All considerations for future studies!

THEORETICAL IMPACT
Interestingly, this study was not able to establish engaging brand content as having an effect
on brand equity. Coursaris, et al. (2014) was able to establish a relationship from engaging brand
content to brand attitude, an antecedent of brand equity. This study was, however, able to establish
72

engaging brand content as affecting the proposed antecedents of brand equity, cognitive absorption
and flow. This would be great, except for the fact that only flow was a significant predecessor for
brand equity (β = 0.541, p < 0.001) based on the original model from these study results.
SELF-REPORT VS SIMULATION
In terms of evaluating self-report metrics against a simulated environment, this research
is not able to do that. If it is considered that each participant was exposed to six brand Facebook
posts, and each post had six possible interactions in the simulation, that’s 36 possible clicks per
participant. Times 858 participants, is 30,888 possible interactions. Seven posts had hyperlinks,
so seven times 858 participants plus 30,888 is equal to 36,894 total possible interactions in the
simulated environment. Through the entire study, for 858 participants, only 634 interactions
were performed, resulting in a 1.72% response rate. Granted, the whole point of the
simulation/study posts was to ask if participants would interact with the posts. But when it is
considered that in the self-report metrics, the results can cover 634 interaction intentions within
the first two posts, these measurements are not comparable. While participants do have a
tendency to incorrectly measure their own self actions (Bailenson, et al., 2004), it is doubted that
less than two percent completion rate could be anything comparable as a threshold of self-report
versus behavior accuracy. In actuality, this probably could have been predicted. Gummerus, et
al.’s 2013 study found that only a small portion of customers actively interact with social media
content and other member’s commentary. Most people just use brand pages as information
sources for consumption, not for contributing too (Gummerus, et al., 2013).
One explanation for the lack of interaction with the simulated environment could be
different uses and thought processes for interacting with social media. When scrolling through
news feeds, a participant could read something, make a mental note that they like it, and in their
mind – that’s it! They have interacted with the post, even though they failed to leave the physical
breadcrumb of that thought on the post for the rest of the world to see. This could fall under Online
Disinhibition Effect, in which face to face social customs are no longer considered necessary by
73

some people on the internet, such as acknowledging understanding/reading something and
therefore engaging online (Suler, 2004). As Baird & Parasnis (2011) put it, despite the rapid
growth of social media adoption, few consumers regularly engage by responding to posts and
interacting with the media. Some people could be more interested in not spreading their digital
thumbprint and therefore do not leave interactions on pages so that their profile is not associated
with that digital trail. All of these ideas require further investigation into psychological processes
and mental models that are being used while people interact with social media in general and/or
Facebook. But they would make an exciting future study!
Another explanation of this lack of consistency between to the simulation and self-report
areas comes from Berkowitz’s 1989 studies in experimental social psychology. He proposes that
attitudes people find easy to explain, are therefore easier to self-report on because they should be
unaffected by any reasons. While it would make sense that whether or not a Facebook post should
be shared or interacted with should be a gut decisions, perhaps there are additional social
pressure/awareness issues that should be further analyzed and looked at which were not the scope
of this research. Or, at the very least, the scales for gaging brand attitude used in Coursaris, et al.
(2014) could have provided some insight into this area of social psychology in a social media
platform. This is a very long way of saying that for this project, sorry, but the simulation did not
work and was therefore not further analyzed.

POST POPULARITY/COMMUNITY METRICS
It is surprising to find very limited significant relationships within the manipulation to
display Post Popularity metrics. Consumers are able to provide 22 times the persuasive effect than
marketers can on other consumers (Goh, Heng, & Lin, 2013). Why is it then that no effects were
observed from this manipulation as either a reflexive variable or as a moderating effect?
Perhaps the purest test of some type of effect as to whether or not Post Popularity should
be visible to Facebook users, is the fact that when isolating the manipulation group that was shown
74

no metrics, the model can be validated for significant relationships among all of the paths.
Looking at the isolated model this way, the various Coursaris, et al. papers findings using student
samples can be validated on a general population sample, which was one of the desired objectives
of this research. But it leaves one to wonder what the point of showing these Post Popularity is
for. Could they be there to show return on investment by letting people as well as brands know
how far their interactions have reached? While this works conceptually for a competitive analysis,
these engagement measures are not the true impressions and reach metrics only shown to the
actual Page owner. Is there something more to the psychological aspect of showing these metrics
that this study was just not able to capture? There is definitely more to be looked into here to
validate that showing these metrics is not a complete waste of coder time and bogging down user
cognitive resources in the social media environment.
Several ideas are entertained as to why this lack of effect might have happened. First,
perhaps brand posts alone do not provide enough of a problem solving aspect for information
seeking to attract consumer need to process information (Davis, Piven, & Brazeale, 2014; HennigThurau, et al., 2010). Facebook Page posts put forward more informative information and appear,
or not, depending on the Facebook algorithm, in a user’s News Feed. Perhaps this passive strategy
for unobtrusively consuming brand information, given that most users see brand pages as a source
of information only (Gummerus, et al., 2013), is not sufficient to create strong enough information
processing for common social pressures to take effect. However, the cognitive flow model asserts
that higher rates of interaction with a technology can increase the number of user challenges
which might then allow information processing, thus confusing the matter once again (Cheng,
Chieng, & Chieng, 2014; Oh & Sundar, 2015). Huang (2011) asserts that involvement is a key
antecedent of flow, which would explain why flow always has a significant effect on brand equity.
And thus, we are stuck in circular logic.
Second, perhaps the follower base for the brand pages was too large to have an immediate
effect on Post Popularity to participants. While the brands and posts selected were attempting to
75

accommodate all possible purchase situations (high involvement, infrequent purchasing with Delta;
low involvement, frequent purchasing of fast-moving consumer goods with McDonald’s), since this
was the first look into community effects on consumers in a social media environment, no attempt
was made to vary the online brand community structure. Since it is millions of strangers engaging
in these posts, perhaps this did not create enough relevancies to affect the participants. Or, both
brands have to stay relevant to a large number of people. Maybe this makes it harder for individuals
to feel connected to the brand since they do not market to their specific interests directly. Both Pages
are administered by McDonald’s and Delta Airlines themselves, allow for user feedback and posting
to the Page, and maintain a follower base of more than one million. Additionally, Wirtz, et al. (2013)
proposed a framework for categorizing brands by funding and governance structures based on the
community or the brand/firm controlling these entities (see Figure 21). To the best of our
knowledge, McDonald’s and Delta Airlines are administered and controlled by the brand in both
the areas of funding and governance, thus putting them in the top right corner of the quadrants.
Perhaps if a brand on the other side of the Funding/Governance quadrants, towards more
community management and funding and farther away from Brand/Firm management had been
analyzed, a community effect might have been observed.

76

Figure 21. Funding and Governance Combinations
of Online Brand Communities

The issue with brand community size could, however, also work with Dunbar’s 1998 idea
of weak versus strong ties idea. Robin Dunbar (1998) calculated that humans can handle a social
group size of 150 people (with a confidence interval of 95-230 people) and maintain stable, strong
relationships with these people. This would suggest that users can better handle smaller
communities, and that perhaps a community effect might have been observed in the low metric
manipulations. So can engagement metrics translate to community size? Perhaps, but maybe not.
Engagement metrics are also a factor of the entire Facebook reach for a post. But it does bring
into question the community size – which this study did not vary so that an overall effect could
first be tested. But then, if we look at a more recent counter argument to Dunbar's research and
apply the idea that weak ties are more important in a social media, modern world, where large
networks are king, then the present research again does not make sense that there is no effect of
displaying Post Popularity metrics on brand communication posts (Dunbar, 1998; Morgan, 2010).
Another idea could be the difference between online asynchronous activities, and bonding
synchronous activities. Online activities are asynchronous and therefore do not fulfill the same
bonding and social needs. Because of the lack of physical proximity of participants in an online
77

dialog, it is difficult to test for this effect and to design an experiment that can accurately gage if
this is somehow causing a validity issue. One can only wonder if the LOL acronym or other forms
of written humor have become acceptable, psychological replacements for the act of laughing, and
if the two cause similar or different brain wave activities and bonding.
Along a similar line, in 2010, Krienen, Tu, and Buckner found that the medial prefrontal
cortex (MPFC) of the human brain activates and reacts differently to people we know and that we
pass judgement on people based on how similar they are to us. Huang (2011) shows the same
relationship with organizations that people identify with brand characteristics they can easily
relate to. Although his credibility has been questioned in recent years, Lehrer proposed in 2010
that perhaps Krienen, Tu, and Buckner’s results are the reason why some people check their
Facebook page so regularly: we care about people we know, and perhaps it activates our MPFC as
well. While just questions at this point, it would be interesting to see if in fact social media can
cause the same neurological and psychological responses as synchronous, in person bonding
activities.
Further, to tie these last few points together, Animesh, Yang, and Oh conducted a study in
2011 in which they observed that regular communication with other participants in a virtual
reality environment made it easier to spontaneously have conversations and overcome any
difficulties in the formation of social ties. This, in turn, they noticed had an impact on decreasing
any inhibitors of cognitive flow, the sense of intrinsic enjoyment from interacting with a
technology (Animesh, Yang, & Oh, 2011; Animesh, et al., 2011). Conversely, infrequent
participants had stronger barriers to create social ties and decreased sense of flow. So basically,
this establishes flow as related to social presence (Animesh, Yang, & Oh, 2011). While flow was
considered in this study as a predecessor to brand equity, it might be interesting to see what
relations Post Popularity metrics have on flow as a dependent variable, and if these effects and
influences hold true in a social media environment. The results of this research already establish

78

flow as a statistically significant antecedent to brand equity thus forming the basis of cognitive
flow research in the social media realm.
However, ultimately it is possible that perhaps Post Popularity was operationalized
correctly, but conceptually misguided. The definition of community engagement, specified in
Table 1, as “the amount of Likes, Comments, Shares a post receives on Facebook” still makes sense
for what was being tested in this study. However, perhaps more emphasis should have been placed
on the use of the word engagement as opposed to community. In retrospect, page 8 of this
dissertation references “The extent to which a social media message is engaging—i.e., resulting in
consumer engagement in the form of Likes, Comments, or Shares…” Perhaps the correct phrasing
was in the introduction all along. Community implies that there is a presence of people and
perhaps they are interacting and communicating together. This is how most of the brand
community literature reads (Wang, Yu, & Wei, 2012; Erdogmus & Cicek, 2012; Alhabash, et al.,
2013). Pletikosa Cvijikj & Michahelles (2013) describe Likes, Comments, and Shares in terms of
"engagement and participation," construct to describe participant interactions with posts.
Ultimately, the term "popularity" was selected to represent the amount of Likes, Comments, and
Shares collectively on a post. However, this word still does not seem accureate when considering
how Facebook creates exposure of the post by based on its current algorithm settings, because the
reach can vary so much. However, it was used in this study to represent the preliminary analysis
into community interactions, and further researcher should investgate how to term these metrics
as they appear on Facebook Posts.

COGNITIVE ABSORPTION
The overall structure of cognitive absorption as a construct needs further consideration. A
relatively new construct to communications and technology use, part of what made initial
literature discovery confusing was that the name cognitive absorption is both an overarching
construct, as well as a dimension itself in the construct, along with flow and cognitive engagement.

79

Even Agarwal & Karahanna, who did a lot of the initial research in this area, acknowledged this
when they questioned whether flow and engagement are the same dimensions conceptually
and/or are empirically distinct in their paper from 2000. Fifteen years later, Oh and Sundar
(2015) draw a comparison between cognitive absorption in both the persuasion and humancomputer interaction areas. In human-computer interaction discussions, cognitive absorption is
the term used. In persuasion studies, the same construct is called elaboration or cognitive
involvement (Oh & Sundar, 2015). However, elaboration is described as a divergent process of
building issue and experience related arguments, while absorption is described as a convergent
process to focus on a single point for the discussion (Oh & Sundar, 2015).
To make matters worse, occasionally cognitive arousal is used interchangeably with
cognitive absorption. While cognitive arousal is defined by thinking and mental stimulation to
facilitate learning, cognitive absorption is a deep state of involvement or giving complete and
absolute attention, specifically with technology. This study looked at cognitive absorption, which
makes sense someone could give their complete and uninterrupted attention to highly engaging
brand Facebook posts. Whereas, if this study had been looking to see if messages were educating
consumers to enhance brand equity, perhaps it should have used a cognitive arousal scale instead.
As mentioned earlier, it is a shame that cognitive engagement dimension questions were
not included to measure the cognitive absorption dimension with participants. Considering that
the cognitive engagement area encompasses the enjoyment aspect of cognitive absorption
(Agarwal & Karahanna, 2000), this study possibly missed a key area the social media experience.
Further consideration should also be given to the area of environment interaction. Oh and
Sundar (2015) mention that a consequence of interacting with a site can be that the user pays
more attention to the browsing task than what is being requested of them. Therefore, what
enhances cognitive absorption can be blurred into the environment, lost in whichever element the
consumer is focusing on. Eye tracking software and heat maps would be needed to further analyze
this phenomenon. Or, more elaborate browsing capabilities, both through interaction as well as
80

the ability to scroll through the News Feed of the Facebook environment might be needed.
Therefore, while the environment in this research was carefully analyzed for any validity issues
and to make sure a true Facebook user experience was being replicated, since the research was on
the Page’s post and not any of the peripheral area, this area was not coded for interaction
capabilities. Therefore, unfortunately, there could still be a problem with ecological validity in this
study.

PRACTICAL IMPACT
Practical implications from these results are as follows. Social marketing is an area to
communicate on a deeper level with a brand’s audience (Hosea, 2011). With the amount of social
network uses expected to almost double from 1.7 billion people in 2013 to 2.55 billion people in
2017, it is with due cause that firms learn the ins and outs of their effects of using this type of
communication with their consumers (Davis, Piven, and Breazeale, 2014; Mitchell and Olson,
1981; Faircloth, Capella, and Alford, 2001). Stronger Brand Equity, Purchase Intent, and Brand
Social Media Engagement Intent results are produced when no Post Popularity metrics are shown.
While the Post Popularity metrics serve as a way for companies to conduct competitive analyses,
further research is necessary to conclusively decide if these metrics are in fact impacting consumer
brand relations. Perhaps it is more pertinent for an organization to focus on their reach and how
well they are doing to satisfy Facebook’s EdgeRank so that posts can make it into their audience’s
news feed. Or, interaction with a post is more heavily based on peer connections than this study
was able to test for.
This study confirms that engaging content has an effect on consumer Brand Equity,
Purchase Intent, and Brand Social Media Engagement Intent within the general public, and not
just a student sample as used in Coursairs, et al., 2014a. therefore, practitioners should
concentrate their efforts on creating engaging content to attract Facebook users to stop just
scrolling through their news feed, and also interact with posts. These results were highlighted in
81

the significance of Flow, and future research should look into how to break into a cognitive flow
cycle enough to encourage this interaction. Ultimately, this study was able to find that if that initial
Brand Equity can be built, there is a relationship that this will influence Purchase Intention and
Engagement Intention thus possibly creating a cycle of positive consumer engagement, purchase,
and ultimately brand loyalty.

CONCLUSIONS
The brands chosen for analysis in this study were based on previous research in the area,
in an attempt to validate findings in a general population sample with comparable data. Previous
Coursaris, et al., studies have used a student population which, while an excellent starting point,
can be construed as an educated, younger demographic, not representative of the general
population. That being said, it might be interesting to identify brands to analyze with the
consumer base before setting up the experiment as Broyles, Schumann, and Leingpibul (2009)
did. Such a set up would increase relevancy to the participants, and ensure they already have an
interest in the brands being discussed (Baird & Parasnis, 2011). Building in such a selection bias
could allow for an even more neutral sample to allow for testing effects of antecedents of brand
equity as well as brand social media engagement intention and community exposure metric
effects. Once it is known if such relationships truly exist, then taking that model back to the
general public would be a better starting place.
The fact that little can be found in terms of the relationship of Post Popularity metrics to
consumer perceptions towards brand posts and brand relationships is puzzling, and leads to so
much more interesting research to expand upon this study with. Considering there is a stronger
relationship between Brand Equity and Brand Social Media Engagement Intention when no
metrics are shown to the participant, why are these metrics even there? Has Facebook included
them strictly so that other companies can do competitive analyses among industry brands? How
nice of them! It just doesn’t make sense; Facebook could just shut down a development area and
82

save on the overhead costs for implementing these metrics on the page visible to the user. There
has to be a reason as to why they are included and further research into consumer cognitive
overload and cognitive enjoyment should be done.
This research provides several key findings to marketing practitioners and scholars. First,
it validates Coursaris, et al.’s 2014a study results on a general population sample, as opposed to
the student sample used in the original study. Second, this study began investigation into
understanding antecedents of brand equity. While cognitive absorption could not be confirmed
as a statistically significant influencer, several reasons were stated as to why this area needs
further deciphering and research to begin with. Cognitive flow was established as a highly likely
predictor of brand equity. In a technological environment like Facebook, it makes sense that users
would be so lost in the enjoyment of use and thus cognitive flow, that they would develop higher
brand equity and, in turn, brand loyalty (something also for a future test!). Third, surprisingly,
the presence of community metrics, whether true metrics or switched, was not established as a
significant relationship in the model. This leads to the conclusion that perhaps interactions with
a post are purely content driven, with limited, if any, environmental factors influencing consumer
engagement in the Facebook environment. Fourth, active interaction was not established as an
equivocal reflection of self-report metrics for engagement intention.
Brown, Broderick, and Lee (2007) mention that perhaps traditional, offline marketing
theories cannot be used to appropriately study online word of mouth communications. The brand
takes on too much of a friend role, thus creating a peer scenario instead of a persuasive, higher
entity (Wang, Yu, & Wei, 2012). The relationship between brand and computer is multifaceted
(Brown, Broderick, & Lee, 2007), and this was more than the current research structure could
accommodate. But it made for a wide range of excellent future research ideas!
Further understanding of the Facebook environment is still needed. For while this study
was not able to establish community/Post Popularity as a significant indicator of brand social
media engagement intention, several companies have actually started pulling away from
83

Facebook advertising as being “too targeted” and therefore ineffective (Terlep & Seetharaman,
2016; Pan, 2012). If social media’s community engagement metrics are not effective, and
consumers and advertisers start pulling away, what does this lead Web 3.0 towards?

84

APPENDICES

85

APPENDIX A. MANIPULATION - BRAND FACEBOOK PAGE POST
SELECTIONS’ ENGAGEMENT METRICS

Table 20. Manipulation - Brand Facebook Page Post Selections’
Engagement Metrics
McDonalds
Brand Awareness
Product
Awareness
Engagement

Delta Airlines
Brand Awareness
Product
Awareness
Engagement

Engagement
Condition

Likes

Comments

Shares

Elicited
Engagement*

Low

2,108

181

41

2,330

High

7,904

616

267

8,787

Low

17,856

344

343

18,543

High

34,181

2,181

828

37,190

Low

9,932

497

366

10,795

High

20,499

571

581

21,651

Engagement
Condition

Likes

Comments

Shares

Elicited
Engagement

Low

390

0

26

416

High

1,844

101

269

2,214

Low

524

1

38

563

High

1,949

135

195

2,279

Low

160

0

2

162

High

2,723
314
715
3,752
* Sum of raw number of Likes, Comments, and Shares

86

APPENDIX B. SIMULATED FACEBOOK ENVIRONMENT

Figure 22. Example of Simulated Facebook Environment

87

APPENDIX C. FACTOR LOADINGS

Table 21. Factor Loadings
Loadings

Resources

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intent

Items
When I was reading the post, I was
able to block out all other
Distractions.
When I was reading the post, I felt
totally immersed in the post.
When I was reading the post, I felt
completely absorbed in the post.
When I was reading the post, my
attention did not get diverted very
easily.
My imagination is aroused when I
interact with the post.
I feel curious when interacting with
the post.
The interaction with the post is
interesting.
I am absorbed in the interaction in
the post.
It’s fun to interact with the post.
Even if another [food service/airline]
offers the same quality of services as
[McDonald’s/Delta Airlines], I would
prefer to use the services of
[McDonald’s/Delta Airlines].
If there is another [food
service/airline] as good as
[McDonald’s/Delta Airlines], I prefer
to go to [McDonald’s/Delta Airlines].
It makes sense to use the services of
[McDonald’s/Delta Airlines] instead
of services of any other [food
service/airline] even if they are the
same.
The likelihood that I would purchase
from [McDonald’s/Delta Airlines]
The probability that I would consider
buying from [McDonald’s/Delta
Airlines]
My willingness to buy from
[McDonald’s/Delta Airlines]
For this particular type of purchase, I
would use [McDonald’s/Delta
Airlines]

88

No
Manipulation

No Post
Popularity
Metrics
Shown

Metrics
Switched

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

Formative

0.973

0.975

0.976

0.976

0.978

0.980

0.959

0.966

0.965

0.957

0.965

0.963

0.956

0.967

0.961

0.955

0.967

0.957

0.940

0.951

0.950

Table 21. Continued

Resources
Purchase
Intent

Brand Social
Media
Engagement
Intention

Items
My intention would be to purchase
from
[McDonald’s/Delta Airlines]
Considering this [McDonald’s/Delta
Airlines] Facebook Page post, how
likely are you to do each of the
following?
Like this post
Comment on this post
Share this post on my wall
Share this post on a friend’s wall
Like the [McDonald’s/Delta Airlines]
Facebook Page

89

No
Manipulation

No Post
Popularity
Metrics
Shown

Metrics
Switched

0.956

0.964

0.963

0.849
0.849
0.914
0.881

0.863
0.861
0.910
0.875

0.850
0.843
0.909
0.871

0.877

0.902

0.873

APPENDIX D. CONSTRUCT STATISTICS

Table 22. Construct Statistics: No Manipulation
Mean

Cronbach’s
Alpha (α)

Composite
Reliability

Convergent
Validity
(AVE)

Discriminant
Validity
(√AVE)

Cognitive Absorption

4.808

Formative

Formative

Formative

Formative

Flow

3.989

Formative

Formative

Formative

Formative

Brand Equity

3.784

0.968

0.979

0.940

0.970

Purchase Intention

4.221

0.975

0.980

0.908

0.953

Brand Social Media
Engagement Intention

2.751

0.923

0.942

0.765

0.875

Construct

Table 23. Construct Statistics: No Post Popularity Metrics Shown
Mean

Cronbach’s
Alpha (α)

Composite
Reliability

Convergent
Validity
(AVE)

Discriminant
Validity
(√AVE)

Cognitive Absorption

4.681

Formative

Formative

Formative

Formative

Flow

3.817

Formative

Formative

Formative

Formative

Brand Equity

3.808

0.972

0.982

0.947

0.973

Purchase Intention

4.214

0.980

0.984

0.927

0.963

Brand Social Media
Engagement Intention

2.685

0.929

0.946

0.778

0.882

Construct

Table 24. Construct Statistics: Metrics Switched
Mean

Cronbach’s
Alpha (α)

Composite
Reliability

Convergent
Validity
(AVE)

Discriminant
Validity
(√AVE)

Cognitive Absorption

4.779

Formative

Formative

Formative

Formative

Flow

3.988

Formative

Formative

Formative

Formative

Brand Equity

3.807

0.973

0.982

0.948

0.974

Purchase Intention

4.213

0.978

0.983

0.919

0.956

Brand Social Media
Engagement Intention

2.726

0.919

0.939

0.756

0.869

Construct

90

APPENDIX E. LATENT VARIABLE CORRELATIONS

Table 25. Latent Variable Correlations: No Manipulation
Construct

Cognitive Absorption

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intention

Brand Social
Media
Engagement
Intention

n/a
(formative)

Flow

0.675

n/a
(formative)

Brand Equity

0.383

.559

1.00

Purchase Intention

0.407

.550

.791

1.00

Brand Social Media
Engagement Intention

0.453

.666

.574

.530

1.00

Table 26. Latent Variable Correlations: No Post Popularity Metrics Shown
Construct

Cognitive Absorption

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intention

Brand Social
Media
Engagement
Intention

n/a
(formative)

Flow

0.686

n/a
(formative)

Brand Equity

0.455

0.601

1.00

Purchase Intention

0.450

0.551

0.822

1.00

Brand Social Media
Engagement Intention

0.508

0.718

.602

0.548

91

1.00

Table 27. Latent Variable Correlations: Metrics Switched
Construct

Cognitive Absorption

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intention

Brand Social
Media
Engagement
Intention

n/a
(formative)

Flow

0.704

n/a
(formative)

Brand Equity

0.373

0.528

1.00

Purchase Intention

0.399

0.507

0.821

1.00

Brand Social Media
Engagement Intention

0.471

0.678

0.566

0.504

92

1.00

APPENDIX F. MATRIX OF LOADINGS AND CROSS-LOADINGS

Table 28. Matrix of Loadings and Cross-Loadings: No Manipulation
Flow
(formative)

Brand
Equity
(reflective)

Purchase
Intention
(reflective)

Brand Social
Media
Engagement
Intention
(reflective)

0.764

0.401

0.214

0.247

0.244

2

0.944

0.685

0.403

0.410

0.479

3

0.939

0.684

0.402

0.401

0.478

4

0.847

0.512

0.257

0.329

0.322

1

0.597

0.916

0.506

0.489

0.590

2

0.568

0.908

0.460

0.473

0.550

3

0.617

0.945

0.525

0.525

0.625

4

0.710

0.913

0.517

0.490

0.629

5

0.605

0.917

0.553

0.546

0.659

1

0.376

0.557

0.973

0.780

0.569

2

0.371

0.541

0.976

0.780

0.552

3

0.367

0.527

0.959

0.742

0.547

1

0.377

0.507

0.735

0.957

0.484

2

0.363

0.498

0.721

0.956

0.465

3

0.378

0.508

0.734

0.955

0.447

4

0.410

0.554

0.776

0.940

0.538

5

0.408

0.549

0.798

0.956

0.551

1

0.453

0.651

0.528

0.525

0.849

2

0.395

0.547

0.441

0.358

0.849

3

0.389

0.577

0.495

0.440

0.914

4

0.340

0.527

0.482

0.414

0.881

5

0.397

0.599

0.548

0.551

0.877

Cognitive
Absorption
(formative)
1

Construct

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intent

Brand Social
Media
Engagement
Intention

93

Table 29. Matrix of Loadings and Cross-Loadings: No Post Popularity
Metrics Shown
Flow
(formative)

Brand
Equity
(reflective)

Purchase
Intention
(reflective)

Brand Social
Media
Engagement
Intention
(reflective)

0.814

0.457

0.299

0.321

0.305

2

0.940

0.688

0.460

0.447

0.526

3

0.935

0.709

0.476

0.447

0.546

4

0.877

0.540

0.350

0.366

0.377

1

0.602

0.919

0.546

0.496

0.640

2

0.594

0.920

0.546

0.493

0.627

3

0.643

0.950

0.566

0.522

0.669

4

0.707

0.928

0.557

0.510

0.663

5

0.635

0.921

0.574

0.532

0.727

1

0.454

0.595

0.975

0.815

0.591

2

0.436

0.575

0.978

0.808

0.574

3

0.593

0.585

0.966

0.777

0.593

1

0.422

0.510

0.779

0.965

0.513

2

0.425

0.508

0.770

0.967

0.493

3

0.419

0.515

0.780

0.967

0.504

4

0.450

0.565

0.803

0.951

0.557

5

0.449

0.550

0.824

0.964

0.570

1

0.481

0.701

0.551

0.534

0.863

2

0.449

0.606

0.442

0.383

0.861

3

0.431

0.611

0.511

0.442

0.910

4

0.390

0.555

0.495

0.418

0.875

5

0.480

0.675

0.625

0.599

0.902

Cognitive
Absorption
(formative)
1

Construct

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intent

Brand Social
Media
Engagement
Intention

94

Table 30. Matrix of Loadings and Cross-Loadings: Metrics Switched
Flow
(formative)

Brand
Equity
(reflective)

Purchase
Intention
(reflective)

Brand Social
Media
Engagement
Intention
(reflective)

0.796

0.434

0.228

0.258

0.257

2

0.946

0.724

0.387

0.397

0.499

3

0.941

0.717

0.393

0.404

0.509

4

0.858

0.537

0.261

0.319

0.322

1

0.620

0.915

0.478

0.452

0.599

2

0.601

0.912

0.435

0.426

0.572

3

0.653

0.943

0.502

0.489

0.644

4

0.732

0.925

0.497

0.464

0.632

5

0.632

0.917

0.516

0.500

0.672

1

0.367

0.520

0.976

0.807

0.555

2

0.365

0.514

0.980

0.809

0.547

3

0.357

0.509

0.965

0.783

0.552

1

0.370

0.463

0.774

0.963

0.452

2

0.386

0.458

0.761

0.961

0.445

3

0.389

0.485

0.770

0.957

0.462

4

0.396

0.513

0.803

0.950

0.520

5

0.390

0.508

0.825

0.963

0.530

1

0.489

0.683

0.517

0.500

0.850

2

0.406

0.555

0.420

0.327

0.843

3

0.377

0.560

0.488

0.415

0.909

4

0.326

0.509

0.477

0.375

0.871

5

0.440

0.624

0.543

0.540

0.873

Cognitive
Absorption
(formative)
1

Construct

Cognitive
Absorption

Flow

Brand
Equity

Purchase
Intent

Brand Social
Media
Engagement
Intention

95

APPENDIX G. DEMOGRAPHICS

Mean = 4.18, s = 1.509, N = 285

Mean = 4.25, s = 1.486, N = 285

Figure 23. Age Demographics: No
Post Popularity Metrics Shown

Figure 24. Age Demographics: No
Manipulation

Mean = 4.19, s = 1.497, N = 858

Mean = 4.14, s = 1.50, N = 288

Figure 25. Age Demographics:
Metrics Switched

Figure 26. Age Demographics: All
Manipulation Groups

96

Individualistic

Collective

Individualistic

Mean = 3.49, s = 1.17, N = 285

Mean = 3.40, s = 1.229, N = 285

Figure 27. Culture Demographics: No
Post Popularity Metrics Shown

Individualistic

Collective

Figure 28. Culture Demographics:
No Manipulation

Individualistic

Collective

Collective

Mean = 3.50, s = 1.177, N = 858

Mean = 3.60, s = 1.127, N = 288

Figure 29. Culture Demographics:
Metrics Switched

Figure 30. Culture Demographics:
All Manipulation Groups

97

Extrovert

Introvert

Extrovert

Introvert

Mean = 3.94, s = 1.54, N = 285

Mean = 4.00, s = 1.502, N = 285

Figure 31. Personality Demographics:
No Post Popularity Metrics Shown

Figure 32. Personality Demographics:
No Manipulation

Extrovert

Introvert

Extrovert

Introvert

Mean = 3.93, s = 1.485, N = 288

Mean = 3.96, s = 1.508, N = 858

Figure 33. Personality Demographics:
Metrics Switched

Figure 34. Personality Demographics:
All Manipulation Groups

98

Table 31. ANOVAs for Post Popularity Metric Exposure Differences Compared
with True Post Popularity Metrics Being Shown
Demographic

Manipulation

F Score

P-Value

Status

Age

No Metrics

3.026

0.007

Significant

Switched Metrics

0.551

0.769

Not Significant

No Metrics

0.394

0.531

Not Significant

Switched Metrics

0.166

0.684

Not Significant

No Metrics

1.445

0.060

Not Significant

Switched Metrics

1.033

0.424

Not Significant

No Metrics

0.780

0.724

Not Significant

Switched Metrics

0.808

0.691

Not Significant

Gender

Culture
Personality

99

APPENDIX H. MODEL FIT COMPARISONS

Table 32. Standardized Root Mean Square Residuals (SRMR)
for Different Models Considered
Figure

Model

SRMR

Status

8

PLS Model

0.222

Not Significant

9

PLS Model with No Post
Popularity Metrics Shown

0.238

Not Significant

10

PLS Model with Post Popularity
Metrics Matching Brand
Content

0.223

Not Significant

11

PLS Model with Post Popularity
Metrics Opposite Brand Content

0.233

Not Significant

12

PLS Model with Post Popularity
as a Moderator

0.232

Not Significant

16

All Posts, Engagement Groups,
MGA Path Differences

0.232

Not Significant

17

McDonald’s Brand Awareness
Post, Culture Groups, MGA Path
Differences

0.196

Not Significant

18

McDonald’s Brand Awareness
Post, Engagement Groups, MGA
Path Differences

0.200

Not Significant

19

McDonald’s Posts, MGA Post
Popularity Metric Path
Differences for Simplified Model

0.065

Significant

20

McDonald’s Posts, MGA Post
Popularity Metric Path
Differences for Brand Equity
Antecedents

0.280

Not Significant

100

APPENDIX I. SURVEY

Please note, this survey is optimized for use with Safari, Firefox, and Internet Explorer (note that
use of Chrome is not advised).
Week 1 Survey
1. Consent
a. I Consent
b. I Decline
During this survey, you will be presented with Facebook Page posts from various brands. The
following information is used to enhance your interaction experience as if you were on your
personal Facebook Newsfeed. This is a separate platform and no interactions with the post(s) will
actually affect your personal Facebook profile.
First Name:
c. FREE RESPONSE TEXT BOX
NEW PAGE ---------------------------------------------2. Do you have regular access to the Internet via a computer, tablet, or smartphone
throughout the day?
a. Yes
b. No
3. Do you have a Facebook account?
a. Yes
b. No
NEW PAGE ---------------------------------------------4. If you have a Facebook profile, how much time (in minutes) do you spend on average
each day on Facebook? Please provide your best estimate in minutes. (For example, if
you spend one hour and a half each day on Facebook, you would enter 90)
a. FREE RESPONSE TEXT BOX
5. During the past week, which of the following social media activities did you participate
in? (Please select all that apply.)
a. Blogs
b. Social Networking Sites (e.g., Facebook, Twitter, etc)
c. Virtual social worlds (e.g., Second Life)
d. Collaborative projects (e.g., Wikipedia)
e. Content communities (e.g., YouTube)
f. Virtual game worlds (e.g., World of Warcraft)
g. I do not regularly participate in any of these activities
h. I regularly participate in social media activities but did not this past week

101

NEW PAGE ---------------------------------------------6. Have you Liked any brand's Facebook Page?
a. I don't think so
b. No
c. I think so
d. Yes
7. What types of brand Pages have you Liked on Facebook? (Please select all that apply.)
a. Airlines (Delta, Southwest, United, etc)
b. Fast food (McDonalds, Arby's, Taco Bell, etc)
c. Consumer foods (milk, ice cream, bananas, etc)
d. Consumer products (linens, kitchen supplies, bathroom supplies, etc)
e. Big Box retailers (Walmart, Target, etc)
f. Services (insurance, car repair, contractors, etc)
g. Creation/Manufacturing (artist, Etsy, guitar picks, etc)
h. Media/television actors (i.e. Marvin Zindler, Paul Harvey, Mel Gibson, Jack
Nicholson, etc)
i. Television networks/channels (i.e. local news station)
j. Sports teams (i.e. Greenbay Packers, Chicago Bulls, Detroit Red Wings, Houston
Texans, etc)
k. I do not Like brand Facebook Pages
l. Other
8. If Other selected, please specify:
a. FREE RESPONSE TEXT BOX
NEW PAGE ---------------------------------------------9. Do you currently Like the McDonald's Facebook Page?
a. I don't think so
b. No
c. I think so
d. Yes
10. Do you currently Like the Delta Airlines Facebook Page?
a. I don't think so
b. No
c. I think so
d. Yes
11. In the past week, did you visit any brand’s Facebook Page?
a. I don't think so
b. No
c. I think so
d. Yes

102

12. What types of brand Pages have you visited in the past week?
a. Airlines (Delta, Southwest, United, etc)
b. Fast food (McDonalds, Arby's, Taco Bell, etc)
c. Consumer foods (milk, ice cream, bananas, etc)
d. Consumer products (linens, kitchen supplies, bathroom supplies, etc)
e. Big Box retailers (Walmart, Target, etc)
f. Services (insurance, car repair, contractors, etc)
g. Creation/Manufacturing (artist, Etsy, guitar picks, etc)
h. Media/television actors (i.e. Marvin Zindler, Paul Harvey, Mel Gibson, Jack
Nicholson, etc)
i. Television networks/channels (i.e. local news station)
j. Sports teams (i.e. Greenbay Packers, Chicago Bulls, Detroit Red Wings, Houston
Texans, etc)
k. I do not visit brand Pages on Facebook
l. I do visit brand Pages on Facebook but did not this past week
m. Other
13. If Other selected, please specify:
a. FREE RESPONSE TEXT BOX
NEW PAGE ---------------------------------------------14. In the past week, did you visit the McDonald’s Facebook Page?
a. I don't think so
b. No
c. I think so
d. Yes
15. In the past week, did you visit the Delta Airlines Facebook Page?
a. I don't think so
b. No
c. I think so
d. Yes
NEW PAGE ---------------------------------------------Please indicate your level of agreement with the following statements, on a scale of Strongly
Disagree to Strongly Agree.
16. Being accepted as a member of a group is more important than having autonomy and
independence. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree

103

17. Being accepted as a member of a group is more important than being independent. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
18. Group success is more important than individual success. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
19. Being loyal to a group is more important than individual gain. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
20. Individual rewards are not as important as group welfare. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
21. It is more important for a manager to encourage loyalty and a sense of duty in
subordinates than it is to encourage individual initiative. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree

104

INSERT STATUS BAR: 12% then continue to next page
NEW PAGE ---------------------------------------------You will be presented with several brand Facebook Page posts. Interact with each post as if you
were on your personal Facebook News Feed, by Liking, Commenting and/or Sharing the post.
Keep in mind that this is a separate platform and no interactions with the post will actually affect
your personal Facebook profile (i.e., they will not be shown to your friends or be associated with
your profile).
Post_1

22. PostHyperLinkClicked
23. PostLikeClicked
24. PostCommentClicked
25. PostCommentText
26. PostShareClicked
27. PostShareText
28. PostShareConfirmed

or
NEW PAGE ----------------------------------------------

Considering this McDonald’s Facebook Page post, how likely are you to do each of the following?
29. Like this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
30. Comment on this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
31. Share this post on my wall (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
32. Share this post on a friend's wall (7pt Likert Scale, Very Unlikely to Very Likely)
105

a.
b.
c.
d.
e.
f.
g.

Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

33. Like the McDonald’s Facebook Page (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
NEW PAGE ---------------------------------------------34. When I was reading the post, I was able to block out all other distractions. (7pt Likert
Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
35. When I was reading the post, I felt totally immersed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
36. When I was reading the post, I felt completely absorbed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
37. When I was reading the post, my attention did not get diverted very easily. (7pt Likert
Scale, Strongly Disagree to Strongly Agree)
106

a.
b.
c.
d.
e.
f.
g.

Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

NEW PAGE ---------------------------------------------38. My imagination is aroused when I interact with the post. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
39. I feel curious when interacting with the post. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
40. The interaction with the post is interesting. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
41. I am absorbed in the interaction in the post. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
42. It’s fun to interact with the post. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
107

a.
b.
c.
d.
e.
f.
g.

Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

NEW PAGE ---------------------------------------------43. Even if another food service offers the same quality of services as McDonald’s, I would
prefer to use the services of McDonald’s. (7pt Likert Scale, Strongly Disagree to Strongly
Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
44. If there is another food service as good as McDonald’s, I prefer to go to McDonald’s. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
45. It makes sense to use the services of McDonald’s instead of services of any other food
service even if they are the same. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------46. The likelihood that I would purchase from McDonald’s is: (7pt Likert Scale, Very Low to
Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
108

g. Very High
47. The probability that I would consider buying from McDonald’s is: (7pt Likert Scale, Very
Low to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
48. My willingness to buy from McDonald’s is: (7pt Likert Scale, Very Low to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
49. For this particular type of purchase, I would use McDonald’s. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
50. My intention would be to purchase from McDonald’s. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
INSERT STATUS BAR: 25% then continue to next page
NEW PAGE ----------------------------------------------

Post_2

51. PostHyperLinkClicked
52. PostLikeClicked
109

53. PostCommentClicked
54. PostCommentText
55. PostShareClicked
56. PostShareText
57. PostShareConfirmed

or
NEW PAGE ----------------------------------------------

Considering this Delta Airlines Facebook Page post, how likely are you to do each of the following?
58. Like this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
59. Comment on this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
60. Share this post on my wall (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
61. Share this post on a friend's wall (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
62. Like the Delta Airlines Facebook Page (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
110

f. Likely
g. Very Likely
NEW PAGE ---------------------------------------------63. When I was reading the post, I was able to block out all other distractions. (7pt Likert
Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
64. When I was reading the post, I felt totally immersed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
65. When I was reading the post, I felt completely absorbed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
66. When I was reading the post, my attention did not get diverted very easily. (7pt Likert
Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------67. My imagination is aroused when I interact with the post. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
111

c.
d.
e.
f.
g.

Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

68. I feel curious when interacting with the post. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
69. The interaction with the post is interesting. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
70. I am absorbed in the interaction in the post. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
71. It’s fun to interact with the post. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------72. Even if another airline offers the same quality of services as Delta Airlines, I would prefer
to use the services of Delta Airlines. (7pt Likert Scale, Strongly Disagree to Strongly
Agree)
a. Strongly Disagree
112

b.
c.
d.
e.
f.
g.

Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

73. If there is another airline as good as Delta Airlines, I prefer to go to Delta Airlines. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
74. It makes sense to use the services of Delta Airlines instead of services of any other airline
even if they are the same. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------75. The likelihood that I would purchase from Delta Airlines is: (7pt Likert Scale, Very Low
to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
76. The probability that I would consider buying from Delta Airlines is: (7pt Likert Scale,
Very Low to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
77. My willingness to buy from Delta Airlines is: (7pt Likert Scale, Very Low to Very High)
a. Very Low
113

b.
c.
d.
e.
f.
g.

Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

78. For this particular type of purchase, I would use Delta Airlines. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
79. My intention would be to purchase from Delta Airlines. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
INSERT STATUS BAR: 41% then continue to next page
NEW PAGE ---------------------------------------------Post_3

80. PostHyperLinkClicked
81. PostLikeClicked
82. PostCommentClicked
83. PostCommentText
84. PostShareClicked
85. PostShareText
86. PostShareConfirmed

or
NEW PAGE ----------------------------------------------

Considering this McDonald’s Facebook Page post, how likely are you to do each of the following?
87. Like this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
88. Comment on this post (7pt Likert Scale, Very Unlikely to Very Likely)
114

a.
b.
c.
d.
e.
f.
g.

Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

89. Share this post on my wall (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
90. Share this post on a friend's wall (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
91. Like the McDonald’s Facebook Page (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
NEW PAGE ---------------------------------------------92. When I was reading the post, I was able to block out all other distractions. (7pt Likert
Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
93. When I was reading the post, I felt totally immersed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
115

c.
d.
e.
f.
g.

Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

94. When I was reading the post, I felt completely absorbed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
95. When I was reading the post, my attention did not get diverted very easily. (7pt Likert
Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------96. My imagination is aroused when I interact with the post. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
97. I feel curious when interacting with the post. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
98. The interaction with the post is interesting. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
116

b.
c.
d.
e.
f.
g.

Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

99. I am absorbed in the interaction in the post. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
100.
Agree)
a.
b.
c.
d.
e.
f.
g.

It’s fun to interact with the post. (7pt Likert Scale, Strongly Disagree to Strongly
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

NEW PAGE ---------------------------------------------101.
Even if another food service offers the same quality of services as McDonald’s, I
would prefer to use the services of McDonald’s. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
102.
If there is another food service as good as McDonald’s, I prefer to go to
McDonald’s. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree

117

103.
It makes sense to use the services of McDonald’s instead of services of any other
food service even if they are the same. (7pt Likert Scale, Strongly Disagree to Strongly
Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------104.
The likelihood that I would purchase from McDonald’s is: (7pt Likert Scale, Very
Low to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
105.
Scale,
a.
b.
c.
d.
e.
f.
g.

The probability that I would consider buying from McDonald’s is: (7pt Likert
Very Low to Very High)
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

106.
High)
a.
b.
c.
d.
e.
f.
g.

My willingness to buy from McDonald’s is: (7pt Likert Scale, Very Low to Very
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

107.
For this particular type of purchase, I would use McDonald’s. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
118

g. Strongly Agree
108.
My intention would be to purchase from McDonald’s. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
INSERT STATUS BAR: 55% then continue to next page
NEW PAGE ---------------------------------------------Post_4

109.
PostHyperLinkClicked
110.
PostLikeClicked
111. PostCommentClicked
112.
PostCommentText
113.
PostShareClicked
114.
PostShareText
115.
PostShareConfirmed

or
NEW PAGE ----------------------------------------------

Considering this Delta Airlines Facebook Page post, how likely are you to do each of the following?
116.
Like this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
117.Comment on this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
118.

a.
b.
c.
d.
e.
f.

Share this post on my wall (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
119

g. Very Likely
119.

a.
b.
c.
d.
e.
f.
g.

Share this post on a friend's wall (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

120.
Like the Delta Airlines Facebook Page (7pt Likert Scale, Very Unlikely to Very
Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
NEW PAGE ---------------------------------------------121.
When I was reading the post, I was able to block out all other distractions. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
122.
When I was reading the post, I felt totally immersed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
123.
Scale,
a.
b.
c.
d.
e.
f.

When I was reading the post, I felt completely absorbed in the post. (7pt Likert
Strongly Disagree to Strongly Agree)
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
120

g. Strongly Agree
124.
When I was reading the post, my attention did not get diverted very easily. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------125.
My imagination is aroused when I interact with the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
126.
I feel curious when interacting with the post. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
127.
The interaction with the post is interesting. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
128.
I am absorbed in the interaction in the post. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
121

f. Agree
g. Strongly Agree
129.
Agree)
a.
b.
c.
d.
e.
f.
g.

It’s fun to interact with the post. (7pt Likert Scale, Strongly Disagree to Strongly
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

NEW PAGE ---------------------------------------------130.
Even if another airline offers the same quality of services as Delta Airlines, I
would prefer to use the services of Delta Airlines. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
131.
If there is another airline as good as Delta Airlines, I prefer to go to Delta
Airlines. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
132.
It makes sense to use the services of Delta Airlines instead of services of any other
airline even if they are the same. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------133.
The likelihood that I would purchase from Delta Airlines is: (7pt Likert Scale,
Very Low to Very High)
a. Very Low
122

b.
c.
d.
e.
f.
g.

Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

134.
Scale,
a.
b.
c.
d.
e.
f.
g.

The probability that I would consider buying from Delta Airlines is: (7pt Likert
Very Low to Very High)
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

135.
High)
a.
b.
c.
d.
e.
f.
g.

My willingness to buy from Delta Airlines is: (7pt Likert Scale, Very Low to Very
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

136.
For this particular type of purchase, I would use Delta Airlines. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
137.
My intention would be to purchase from Delta Airlines. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
INSERT STATUS BAR: 70% then continue to next page
NEXT PAGE -------------------------------------------------------------------Post_5

138.
123

PostHyperLinkClicked

139.
140.
141.
142.
143.
144.

or

PostLikeClicked
PostCommentClicked
PostCommentText
PostShareClicked
PostShareText
PostShareConfirmed

NEW PAGE ---------------------------------------------Considering this Delta Airlines Facebook Page post, how likely are you to do each of the following?
145.
Like this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
146.

a.
b.
c.
d.
e.
f.
g.

Comment on this post (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

a.
b.
c.
d.
e.
f.
g.

Share this post on my wall (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

a.
b.
c.
d.
e.
f.
g.

Share this post on a friend's wall (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

147.

148.

149.
Like the Delta Airlines Facebook Page (7pt Likert Scale, Very Unlikely to Very
Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
124

d.
e.
f.
g.

Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

NEW PAGE ---------------------------------------------150.
When I was reading the post, I was able to block out all other distractions. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
151.
When I was reading the post, I felt totally immersed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
152.
Scale,
a.
b.
c.
d.
e.
f.
g.

When I was reading the post, I felt completely absorbed in the post. (7pt Likert
Strongly Disagree to Strongly Agree)
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

153.
When I was reading the post, my attention did not get diverted very easily. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------154.
My imagination is aroused when I interact with the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
125

a.
b.
c.
d.
e.
f.
g.

Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

155.
I feel curious when interacting with the post. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
156.
The interaction with the post is interesting. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
157.
I am absorbed in the interaction in the post. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
158.
Agree)
a.
b.
c.
d.
e.
f.
g.

It’s fun to interact with the post. (7pt Likert Scale, Strongly Disagree to Strongly
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

NEW PAGE ----------------------------------------------

126

159.
Even if another airline offers the same quality of services as Delta Airlines, I
would prefer to use the services of Delta Airlines. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
160.
If there is another airline as good as Delta Airlines, I prefer to go to Delta
Airlines. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
161.
It makes sense to use the services of Delta Airlines instead of services of any other
airline even if they are the same. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------162.
The likelihood that I would purchase from Delta Airlines is: (7pt Likert Scale,
Very Low to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
163.
Scale,
a.
b.
c.
d.
e.
f.

The probability that I would consider buying from Delta Airlines is: (7pt Likert
Very Low to Very High)
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
127

g. Very High
164.
High)
a.
b.
c.
d.
e.
f.
g.

My willingness to buy from Delta Airlines is: (7pt Likert Scale, Very Low to Very
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

165.
For this particular type of purchase, I would use Delta Airlines. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
166.
My intention would be to purchase from Delta Airlines. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
INSERT STATUS BAR: 85% then continue to next page
NEW PAGE ---------------------------------------------Post_6

167.
PostHyperLinkClicked
168.
PostLikeClicked
169.
PostCommentClicked
170.
PostCommentText
171.PostShareClicked
172.
PostShareText
173.
PostShareConfirmed

or
NEW PAGE ----------------------------------------------

Considering this McDonald’s Facebook Page post, how likely are you to do each of the following?
174.
Like this post (7pt Likert Scale, Very Unlikely to Very Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
128

e. Somewhat Likely
f. Likely
g. Very Likely
175.

176.

a.
b.
c.
d.
e.
f.
g.

Comment on this post (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

a.
b.
c.
d.
e.
f.
g.

Share this post on my wall (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

a.
b.
c.
d.
e.
f.
g.

Share this post on a friend's wall (7pt Likert Scale, Very Unlikely to Very Likely)
Very Unlikely
Unlikely
Somewhat Unlikely
Neither Unlikely Nor Likely
Somewhat Likely
Likely
Very Likely

177.

178.
Like the McDonald’s Facebook Page (7pt Likert Scale, Very Unlikely to Very
Likely)
a. Very Unlikely
b. Unlikely
c. Somewhat Unlikely
d. Neither Unlikely Nor Likely
e. Somewhat Likely
f. Likely
g. Very Likely
NEW PAGE ---------------------------------------------179.
When I was reading the post, I was able to block out all other distractions. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
129

g. Strongly Agree
180.
When I was reading the post, I felt totally immersed in the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
181.
Scale,
a.
b.
c.
d.
e.
f.
g.

When I was reading the post, I felt completely absorbed in the post. (7pt Likert
Strongly Disagree to Strongly Agree)
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

182.
When I was reading the post, my attention did not get diverted very easily. (7pt
Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------183.
My imagination is aroused when I interact with the post. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
184.
I feel curious when interacting with the post. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
130

f. Agree
g. Strongly Agree
185.
The interaction with the post is interesting. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
186.
I am absorbed in the interaction in the post. (7pt Likert Scale, Strongly Disagree
to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
187.
Agree)
a.
b.
c.
d.
e.
f.
g.

It’s fun to interact with the post. (7pt Likert Scale, Strongly Disagree to Strongly
Strongly Disagree
Disagree
Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

NEW PAGE ---------------------------------------------188.
Even if another food service offers the same quality of services as McDonald’s, I
would prefer to use the services of McDonald’s. (7pt Likert Scale, Strongly Disagree to
Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
189.
If there is another food service as good as McDonald’s, I prefer to go to
McDonald’s. (7pt Likert Scale, Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
131

d.
e.
f.
g.

Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

190.
It makes sense to use the services of McDonald’s instead of services of any other
food service even if they are the same. (7pt Likert Scale, Strongly Disagree to Strongly
Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
NEW PAGE ---------------------------------------------191.
The likelihood that I would purchase from McDonald’s is: (7pt Likert Scale, Very
Low to Very High)
a. Very Low
b. Low
c. Possibly Low
d. Neither Low Nor High
e. Possibly High
f. High
g. Very High
192.
Scale,
a.
b.
c.
d.
e.
f.
g.

The probability that I would consider buying from McDonald’s is: (7pt Likert
Very Low to Very High)
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

193.
High)
a.
b.
c.
d.
e.
f.
g.

My willingness to buy from McDonald’s is: (7pt Likert Scale, Very Low to Very
Very Low
Low
Possibly Low
Neither Low Nor High
Possibly High
High
Very High

194.
For this particular type of purchase, I would use McDonald’s. (7pt Likert Scale,
Strongly Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
132

c.
d.
e.
f.
g.

Somewhat Disagree
Neither Disagree Nor Agree
Somewhat Agree
Agree
Strongly Agree

195.
My intention would be to purchase from McDonald’s. (7pt Likert Scale, Strongly
Disagree to Strongly Agree)
a. Strongly Disagree
b. Disagree
c. Somewhat Disagree
d. Neither Disagree Nor Agree
e. Somewhat Agree
f. Agree
g. Strongly Agree
INSERT STATUS BAR: 97% then continue to next page
NEW PAGE ---------------------------------------------196.
Considering your normal daily activities and interactions, please rate your
personality in the following areas: (7pt Semantic Differential Scale)
a. Extrovert to Introvert
b. Assertive to Unassertive
c. Talkative to Silent
197.

What country do you live in?
A. FREE RESPONSE TEXT BOX

198.

Gender
a. Female
b. Male

199.

200.

a.
b.
c.
d.
e.
f.
g.
h.

Age
Under 18
18-23 years old
24-29 years old
30-35 years old
36-45 years old
46-55 years old
56-65 years old
Over 66 years old

What do you think this study is about?
a. FREE RESPONSE TEXT BOX

INSERT STATUS BAR: 100% then continue to next page
Thank you for your participation.

133

BIBLIOGRAPHY

134

BIBLIOGRAPHY

Aaker, D.A., & Jacobson, R. (2001). The Value Relevance of Brand Attitude in High-Technology
Markets. Journal of Marketing Research, 38(11), pp. 485-493.
Agozzino, A. (2012). Building a Personal Relationship through Social Media: A Study of
Millennial Students’ Brand Engagement. Ohio Communication Journal, (50), pp. 181204.
Agarwal, R., & Karahanna, E. (2000). Time Flies When You're Having Fun: Cognitive
Absorption and Beliefs About Information Technology Usage. MIS Quarterly, 24(4), pp.
665-694.
Ahmad, A., & Thyagaraj, K.S. (2014). Brand Personality and Brand Equity Research: Past
Developments and Future Directions. IUP Journal of Brand Management, 11(3), 19-56.
Alhabash, S., McAlister, A.R., Quilliam, E.T., Richards, J.I., & Lou, C. (2013). Alcohol’s Getting a
Bit More Social: When Alcohol Marketing Messages on Facebook Increase Young Adults’
Intentions to Imbibe. Mass Communication and Society, 18, pp. 350-375.
Animesh, A., Yang, S.B., & Oh, W. (2011). The Impacts of Technological and Spatial
Environments on Intention to Purchase Virtual Products. MIS Quarterly, 35(3), pp. 789810.
Bagozzi, R. (1981). Evaluating structural equation models with unobservable variables and
measurement error: A comment. Journal of Marketing Research, 18(3), pp. 375-381.
Bailenson, J.N., Aharoni, E., Beall, A.C., Guadagno, R.E., Dimov, A., & Blascovich, J. (2004).
Comparing Behavioral and Self-Report Measures of Embodied Agents' Social Presence
in Immersive Virtual Environments. Presented at the 7th Annual International
Workshop on PRESENCE: 2004. Valencia, Spain.
Baird, C.H., & Parasnis, G. (2011). From social media to social customer relationship
management. Strategy & Leadership, 39(5), pp. 30-37.
Berkowitz, L. (Ed.) (1989). Advances in experimental social psychology. New York: Academic
Press.
Bondesson, N. (2012). Brand Image Antecedents of Loyalty and Price Premium in Business
Markets. Business and Management Research, 1 (1), pp. 32-47.
Boyes, A.D., & Fletcher, G.J.O. (2007). Metaperceptions of BIas in Intimate Relationships.
Journal of Personality and Social Psychology, 92, pp. 286-306.
Briggs, T. (2010). Social Media’s Second Act: Toward Sustainable Brand Engagement. The
Design Management Institute, pp. 46-53.

135

Brodie, R.J., Ilic, A., Juric, B., & Hollebeek, L. (2013). Consumer engagement in a virtual brand
community: An exploratory analysis. Journal of Business Research, 66, pp. 105-114.
Brown, J., Broderick, A.J., & Lee, N. (2007). Word of Mouth Communication within Online
Communities: Conceptualizing the Online Social Network. Journal of Interactive
Marketing, 21(3), pp. 2-20.
Broyles, S.A, Leingpibul, T., Ross, R.H., & Foster, B.M. (2010). Brand equity's
antecedent/consequence relationships in cross-cultural settings. Journal of Product &
Brand Management, 19(3), pp. 159-169.
Broyles, S.A., Schumann, D.W., and Leingpibul, T. (2009). Examining Brand Equity
Antecedent/Consequence Relationships. Journal of Marketing Theory and Practice,
17(2), pp. 145-161.
Bruhn, M., Schoenmueller, V., & Schafer, D.B. (2012). Are social media replacing a traditional
media in terms of brand equity creation? Management Research Review, 35(9), pp. 770790.
Burton-Jones, A., & Straub, Jr., D.W. (2006). Reconceptualizing System Usage: An Approach
and Empirical Test. Information Systems Research, 17(3), pp. 228-246.
Cai, Y., Zhao, G., & He, J. (2015). Influences of two modes of intergenerational communication
on brand equity. Journal of Business Research, 68(9), 553-560.
Callarisa, L., Garcia, J.S., Cardiff, J., & Roshchina, A. (2012). Harnessing social media platforms
to measure customer-based hotel brand equity. Tourism Management Perspectives, 4,
73-79.
Calvo-Porral, C., Martinez-Fernandez, V.A., Juanatey-Boga, O., & Levy-Mangin, J.P. (2015).
Measuring the influence of customer-based store brand equity in the purchase intention.
Cuadernos de Gestion, 15(1), 93-118.
Carmines, E., and Zeller, R. (1979). Reliability and Validity Assessment. Sage Paper Series on
Quantitative Applications No. 07-017, Beverly Hills, CA: Sage Publications Inc.
Chandra, S., Srivastava, S.C., & Theng, Y. (2012). Cognitive Absorption and Trust for Workplace
Collaboration in Virtual Worlds: An Information Processing Decision Making
Perspective. [Special Issue] Journal of the Association of Information Systems, 13, pp.
797-835.
Chandran, S., & Morwitz, V.G. (2005). Effects of Participative Pricing on Consumers’ Cognitions
and Actions: A Goal Theoretic Perspective. Journal of Consumer Research, 32(2), pp.
249-259.
Chang, E., 2014."Influences of the spokes-character on brand equity antecedents," Asia Pacific
Journal of Marketing and Logistics, 26(3), pp. 494-515.
Chang, H.H., & Liu, Y.M. (2009). The impact of brand equity on brand preferences and
purchase intentions in the service industries. The Service Industries Journal, 29(12), pp.
1687-1706.

136

Chen, C.F. & Chang, Y.Y. (2008). Airline brand equity, brand preference, and purchase
intentions - The moderating effects of switching costs. Journal of Air Transport
Management, 14, pp. 40-42.
Chang, T.Z., & Wildt, A.R. (2009). Price, Product Information, and Purchase Intention: An
Empirical Study. Journal of the Academy of Marketing Science, 22(1), pp. 16-27.
Chen, W., & Lee, K.H. (2013). Sharing, Liking, Commenting, and Distressed? The Pathway
Between Facebook Interaction and Psychological Distress. Cyberpsychology, Behavior,
and Social Networking, 16(10), pp. 728-734.
Cheng, L.K., Chieng, M.H., & Chieng, W.H. (2014). Measuring virtual experience in a threedimensional virtual reality interactive simulator environment: a structural equation
modeling approach. Virtual Reality, 18, pp. 173-188.
Cho, E., Fiore, A.M., & Russell, D.W. (2015). Validation of a Fashion Brand Image Scale
Capturing Cognitive, Sensory, and Affective Associations: Testing Its Role in an
Extended Brand Equity Model. Psychology and Marketing, 32(1), pp. 28-48.
Chung, J.Y., Lee, J., & Heath, R.L. (2013). Public relations aspects of brand attitudes and
customer activity. Public Relations Review, 39, pp. 432-439.
Cleff, T., Lin, I.C., & Walter, N. (2013). Can You Feel It? – The Effect of Brand Experience on
Brand Equity. IUP Journal of Brand Management, 11(2), pp. 7-27.
Cobb-Walgren, C.J., Ruble, C.A., Donthu, N. (1995). Brand Equity, Brand Preference, and
Purchase Intent. Journal of Advertising, 24(3), pp. 25-40.
Coursaris, C.K. and Van Osch, W. (2013). A Contextual Messaging Framework: Informing the
Design of Effective Social Media Marketing Messages. Proceedings of the Twelfth PreICIS SIG-HCI Workshop 2013, Milan, Italy.
Coursaris, C.K., Van Osch, W., & Balogh, B.A. (2013). A Social Media Marketing Typology:
Classifying Brand Facebook Page Messages For Strategic Consumer Engagement. ECIS
2013 Completed Research. Paper 46.
Coursaris, C.K., Van Osch, W., & Balogh, B.A. (2014a). Exploring the Empirical Link between
Social Media Engagement, Brand Attitude, and Purchase Intention. Unpublished
manuscript submitted for publication.
Coursaris, C.K., Van Osch, W., & Balogh, B.A. (2016). Informing brand messaging strategies via
social media analytics. Online Information Review, 40(1). pp. 6-24.
Coursaris, C.K., Van Osch, W., Balogh, B.A., Quilliam, E.T. (2014b). Social Media Marketing:
Investigating Empirical Links Between Purchase Involvement, Strategy, Content, and
Media Type. Proceedings of the 2014 Annual Conference of the American Academy of
Advertising, Atlanta, Georgia.
Davis, R., Piven, I., & Breazeale, M. (2014). Conceptualizing the brand in social media
community: The five sources model. Journal of Retailing and Consumer Services, 21,
pp. 468-481.

137

de Vries, L., Gensler, S., & Leeflang, P.S.H. (2012). Popularity of Brand Posts on Brand Fan
Pages: An Investigation of the Effects of Social Media Marketing. Journal of Interactive
Marketing, 26, pp. 83-91.
Diamantopoulos, A., & Siguaw, J. A. (2006). Formative versus reflective indicators in
organizational measure development: a comparison and empirical illustration. British
Journal of Management, 17(4), pp. 263-282.
Dodds, W.B., Monroe, K.B., Grewal, D. (1991). Effects of Price, Brand, and Store Information on
Buyers' Product Evaluations. Journal of Marketing Research, 28(:3), pp. 307-319.
Drengner, J., Gaus, H., & Jahn, S. (2008). Does Flow Influence the Brand Image in Event
Marketing? Journal of Advertising Research, 48(1), pp. 138-147.
Dunbar, R.I.M. (1998). The social brain hypothesis. Evolutionary Anthropology: Issues, News,
and Reviews, 6(5), pp. 178-190.
Dwivedi, A., & Merrilees, B. (2013). Retail brand extensions: Unpacking the link between brand
extension attitude and change in parent brand equity. Australasian Marketing Journal,
21, pp. 75-84.
Elmezni, I., & Gharbi, J. (2010). Mediation of Cognitive Absorption between Users’ Time Styles
and Website Satisfaction. Journal of Internet Banking and Commerce, 15(1), pp. 1-16.
eMarketer. (2016, June 30). Nearly One-Third of the World Will Use Social Networks
Regularly This Year. Retrieved from https://www.emarketer.com/Article/Nearly-OneThird-of-World-Will-Use-Social-Networks-Regularly-This-Year/1014157
Erdogmus, I.E., & Cicek, M. (2012). The impact of social media marketing on brand loyalty.
Social and Behavioral Sciences, 58, pp. 1353-1360.
Facebook. (2017). Facebook Q1 2017 Results [PDF]. Retrieved from
https://s21.q4cdn.com/399680738/files/doc_financials/2017/FB-Q1'17-EarningsSlides.pdf
Faircloth, J.B., Capella, L.M., & Alford, B.L. (2001). The Effect of Brand Attitude and Brand
Image on Brand Equity. Journal of Marketing Theory & Practice, 9(3), pp. 61-71.
Fishbein, M., Ajzen, I., (1975). Belief, attitude, intention, and behavior: an introduction to theory
and research. Addison-Wesley Publishing Company, MA.
Fitzsimons, G.L., & Morwitz, V.G. (1996). The Effect of measuring Intent on Brand-Level
Purchase Behavior. The Journal of Marketing Research, 23(1), pp. 1-11.
Floh, A., & Madlberger, M. (2013). The role of atmospheric cues in online impulse-buying
behavior. Electronic Commerce Research and Applications, 12, pp. 425-439.
Fornell, C., and Larcker, D. (1981). Evaluating Structural Equation Models with Unobservable
Variable and Measurement Error. Journal of Marketing Research, 18, pp. 39-50.

138

Gao, L., & Bai, X. (2014). An empirical study on continuance intention of mobile social
networking services. Asia Pacific Journal of Marketing and Logistics, 26(2), pp. 168189.
Gardner, B.G., and Levy, S.J. (1955). The Product and the Brand. Harvard Business Review
March-April, pp. 33-39.
Gaskin, J. [Jason Gaskin]. (2013, May 2). SEM Series Part 2: Data Screening [Video File].
Retrieved from https://www.youtube.com/watch?v=1KuM5e0aFgU.
Gensler, S., Volckner, F., Liu-Thompkins, Y., & Wiertz, C. (2013). Managing Brands in the Social
Media Environment. Journal of Interactive Marketing, 27, pp. 242-256.
Gelb, B.D., & Rangarajan, D. (2014). Employee Contributions to Brand Equity. California
Management Review, 56(2), pp. 95-112.
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), pp. 88-107.
Grewal, D., Krishnan, R., Baker, J., & Borin, N. (1998). The Effect of Store Name, Brand Name
and Price Discounts on Consumer's Evaluations and Purchase Intentions. Journal of
Retailing, 74(3), pp. 331-352.
Gummerus, J., Liljander, V., Weman, E., & Pihlstrom, M. (2013). Consumer engagement in a
Facebook brand community. Management Research Review, 35(9), pp. 857-877.
Gupta, R., & Kabadayi, S. (2014). The Relationship Between Trusting Beliefs and Web Site
Loyalty: The Moderating Role of Consumer Motives and Flow. Psychology & Marketing,
27(2), pp. 166-185.
Ha, H. (2011). Brand Equity Model and Marketing Stimuli. Seoul Journal of Business, 17(2), pp.
31-60.
Haelein, M. & Kaplan, A.M. (2009). Flagship Brand Stores within Virtual Worlds: The Impact of
Virtual Store Exposure on Real-Life Attitude toward the Brand and Purchase Intent.
Recherche et Applications en Marketing, 24(3), pp. 58-79.
Hainla, L. (2017, March 10). 21 Social Media Marketing Statistics You Need to Know in 2017.
Dreamgrow. Retrieved from https://www.dreamgrow.com/21-social-media-marketingstatistics/
Hair, J.F., Anderson, R.E., Tatham, R.C. and Black, W.C. (1995). Multivariate Data Analysis,
With Readings. Prentice Hall, Englewood Cliffs, NJ.
Harris, L.C., & Goode, M.M.H. (2010). Online servicescapes, trust, and purchase intentions.
Journal of Services Marketing, 24(3), pp. 230-243.
Henning-Thurau, T., Malthouse, E.C., Friege, C., Gensler, S., Lobschat, L., Rangaswamy, A., &
Skiera, B. (2010). The Impact of New Media on Customer Relationships. Journal of
Service Research, 13(3), pp. 311-330.

139

Herzog, H. (1963). Behavioral Science Concepts for Analyzing the Consumer. Marketing and the
Behavioral Sciences, Perry Bliss, (Ed.), Boston: Allyn and Bacon, Inc., pp. 76-86.
Hoffman, D.L., & Novak, T.P. (1996). Marketing in hypermedia computer-mediated
environments: Conceptual foundations. Journal of Marketing, 60(3), pp. 50-68.
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, pp. 149-165.
Hosea, M. (2011, January 6). Engagement takes place in a shared environment. Marketing
Week, pp. 30-33.
Hu, L., & Bentler, P. M. 1998. Fit Indices in Covariance Structure Modeling: Sensitivity to
Underparameterized Model Misspecification, Psychological Methods, 3(4), pp. 424-453.
Huang, E. (2014). Online experiences and virtual goods purchase intention. Internet Research,
22(3), pp. 252-274.
Huang, R., & Sarigollu, E. (2014). Assessment of brand equity measures. International Journal
of Market Research, 56 (6), pp. 783-806.
Hulland, J. (1999). Use of Partial Least Squares (PLS) in Strategic Management Research: A
review of four recent studies. Strategic Management Journal, 20, pp. 195-204.
Hung, K., Chen, A.H., Pent, N., Hackley, C., Tiwsakul, R.A., & Chou, C. (2011). Antecedents of
luxury brand purchase intention. Journal of Product & Brand Management, 20(6), pp.
457-467.
Hutter, K., Hautz, J., Dennhardt, S., & Fuller, J., (2013). The impact of user interactions in
social media on brand awareness and purchase intention: the case of MINI on Facebook.
Journal of Product & Brand Management, 22(5/6), pp. 342-351.
Irshad, W. (2012). Service Based Brand Equity, Measure of Purchase Intention, Mediating Role
of Brand Performance. Academy of Contemporary Research Journal, 1(1), pp. 1-10.
Jiang, Z., & Benbasat, I. (2004). Virtual Product Experience: Effects of Visual and Functional
Control of Products on Perceived Diagnosticity and Flow in Electronic Shopping.
Journal of Management Information Systems, 21(3), pp. 111-147.
Jiang, Z., Chan, J., Tan, B.C.Y., & Chua, W.S. (2010). Effects of Interactivity on Website
Involvement and Purchase Intention. Journal of the Association for Information
Systems, 11(1), pp. 34-59.
Jin, S.A.A. (2012). The potential of social media for luxury brand management. Marketing
Intelligence & Planning, 30(7), pp. 687-699.
Kabadayi, S., & Price, K. (2013). Consumer – brand engagement on Facebook: liking and
commenting behaviors. Journal of Research in Interactive Marketing, 8(3), pp. 203223.

140

Kang, S., Hur, W. (2012). “Investigating the Antecedents of Green Brand Equity: A Sustainable
Development Perspective.” Corporate Social Responsibility and Environmental
Management, 19, pp. 206-216.
Kaplan, A.M., & Haenlein, M. (2010). Users of the world, unite! The challenges and
opportunities of Social Media. Business Horizons, 53, pp. 59-68.
Keller, K.L. (1993). Conceptualizing, Measuring, and Managing Customer-Based Brand Equity.
Journal of Marketing, 57(1), 1-22.
Kerlinger, F.N., & Lee, H.B. (2000). Foundations of Behavioral Research. Belmont, CA:
Cengage Learning.
Kietzmann, J.H., Hermkens, K., McCarthy, I.P., & Silvestre, B.S. (2011). Social media? Get
Serious! Understanding the functional building blocks of social media. Business
Horizons, 54, pp. 241-251.
Kim, M.K., Lopetcharat, K., & Drake, M.A. (2013). Influence of packaging information on
consumer liking of chocolate milk. Journal of Dairy Science, 96, pp. 4843-4856.
Kline, R. B. (1998). Principles and Practice of Structural Equation Modeling. New York: Guilford
Press.
Krienen, F.M., Tu, P.C., & Buckner, R.L. (2010). Clan mentality: evidence tha tthe medical
prefrontal cortex responds to close others. The Journal of Neuroscience, 30(41), pp.
13906-13915.
Kusinitz, S. (2014, June 6). 16 Stats That Prove Social Media Isn't Just a Fad [New Data].
HubSpot. Retrieved from https://blog.hubspot.com/marketing/social-media-roistats#sm.00001vi4zwf992etxyl6a6fascdhn
Laroche, M., Habibi, M.R., & Richard, M.O. (2013). To be or not to be in social media: How
brand loyalty is affected by social media? International Journal of Information
Management, 33, pp. 76-82.
Laroche, M., Habibi, M.R., Richard, M.O., & Sankaranarayanan, R. (2012). The effects of social
media based brand communities on brand community markers, value creation practices,
brand trust and brand loyalty. Computers in Human Behavior, 28, pp. 1755-1767.
Leger, P., Davis, F.D., Cronan, T.P., & Perret, J. (2014). Neurophysiological correlates of
cognitive absorption in an enactive training context. Computers in Human Behavior, 34,
pp. 273-283.
Lehrer, J. (2010, October 26). Why Social Closeness Matters. Wired. Retrieved from
https://www.wired.com/2010/10/why-social-closeness-matters/#comments
LePage, E. (2016, Nov. 29). All the Social Media Advertising Stats You Need to Know.
Hootsuite. Retrieved from https://blog.hootsuite.com/social-media-advertising-stats/
Lin, H. (2009). Examination of cognitive absorption influencing the intention to use a virtual
community. Behavior & Information Technology, 28(5), pp. 421-431.

141

Lobschat, L., Zinnbauer, M.A., Pallas, F., Joachimsthaler, E. (2013). Why Social Currency
Becomes a pe Driver of a Firm’s Brand Equity – Insights from the Automotive Industry.
Long Range Planning, 46, pp. 125-148.
Lowry, P.B., & Gaskin, J. (2014). Partial least squares (PLS) structural equation modeling (SEM)
for building and testing behavioral causal theory: When to choose it and how to use it.
IEEE Transactions on Professional Communication, 57(2), pp. 123-146.
Lunden, I. (2012, May 14). ITU: There Are Now Over 1 Billion Users of Social Media
Worldwide, Most on Mobile. TechCrunch. Retrieved from
https://techcrunch.com/2012/05/14/itu-there-are-now-over-1-billion-users-of-socialmedia-worldwide-most-on-mobile/.
Macinnis, D.J., & Park, C.W. (1991). The Differential Role of Characteristics of Music on Highand Low-Involvement Consumers’ Processing of Ads. Journal of Consumer Research,
18(9), pp. 161-173.
Malhotra, N. 1981. A Scale to Measure Self Concepts, Person-Concepts, and Product Concepts.
Journal of Marketing Research, 18(11), pp. 456-464.
Manara, C., & Roquilly, C. (2011). The Risk of Brand Equity Erosion in the Social Media: the
Efficacy and Limitations of Legal Instruments. Recherche et Applications en Marketing,
26(3), pp. 93-114.
MarketingCharts. (2016, December 20). Top 10 Multi-Platform Social Networking Websites &
Forums - November 2016. Watershed Publishing. Retrieved from
http://www.marketingcharts.com/updates/top-10-multi-platform-social-networkingwebsites-forums-october-2016-2-73213/
Matthews, D.R., Son, J., and Watchravesringkan, K. (2014). An exploration of brand equity
antecedents concerning brand loyalty: A cognitive, affective, and conative perspective.
Journal of Business and Retail Management Research, 9(1), pp. 26-39.
Mitchell, A.A., & Olson, J.C. (1981). Are Product Attribute Beliefs the Only Mediator of
Advertising Effect son Brand Attitude? Journal of Marketing Research, 18(8), pp. 318332.
Moore, K., & McElroy, J.C. (2012). The influence of personality on Facebook usage, wall
postings, and regret. Computers in Human Behavior, 28, pp. 267-274.
Moorman, C. (2013). The CMO Survey; Highlights and Insights, February 2013. Retrieved on
from:
http://cmosurvey.org/files/2013/02/The_CMO_Survey_Highlights_and_Insights_Feb
-2013-Final2.pdf.
Morgan, J. (2010, Jan. 25). Why Dunbar's Number is Irrelevant [Web log post]. Retrieved from
http://www.socialmediatoday.com/content/why-dunbars-number-irrelevant
Morwitz, V.G., Johnson, E., & Schmittlein, D. (1993). Does Measuring Intent Change Behavior?
Journal of Consumer Research, 20(6), pp. 46-61.

142

Nah, F.F.H., Eschenbrenner, B., & DeWester, D. (2014). Enhancing Brand Equity Through Flow
and Telepresence: A Comparison of 2D and 3D Virtual Worlds. MIS Quarterly, 35(3),
pp. 731-747.
Nah, F.F.H., Eschenbrenner, B., DeWester, D., & Park, S.R. (2010). Impact of Flow and Brand
Equity in 3D Virtual Worlds. Journal of Database Management, 21(3), pp. 69-89.
Novak, T.P., Hoffman, D.L., & Yung, Y.-F. (2000). Measuring the customer experience in online
environments: A structural modeling approach. Marketing Science, 19(1), pp. 22-42.
Nunnally, J. C. (1978). Psychometric Theory, 2nd edition, New York: McGraw-Hill.
Oh, J., & Sundar, S.S. (2015). How Does Interactivity Persuade? An Experimental Test of
Interactivity on Cognitive Absorption, Elaboration, and Attitudes. Journal of
Communication, 65, pp. 213-236.
Pan, J. (2012, May 15). Facebook and Twitter Ad Spending to Double by 2016 [REPORT].
Yahoo! News. Retrieved from https://www.yahoo.com/news/facebook-twitter-adspending-double-2016-report-154149894.html
Paolacci, G., Chandler, J., & Ipeirotis, P.G. (2010). Running experiments on Amazon Mechanical
Turk. Judgement and Decision Making, 5(5), pp. 411-419.
Park, C.W., et al. (2010). Brand Attachment and Brand Attitude Strength: Conceptual and
Empirical Differentiation of Two Critical Brand Equity Drivers. Journal of Marketing,
74(11), pp. 1-17.
Park, J.S., Hyun, J., Fairhurst, A., & Lee, K.H. (2012). Perceptions of Presence as Antecedents to
E-tail Shopping. The Research Journal of the Costume Culture, 6, pp. 451-462.
Pew Research Center. (2017, January 12). Social Media Fact Sheet. Retrieved from
http://www.pewinternet.org/fact-sheet/social-media/
Pletikosa Cvijikj, I., & Michahelles, F. (2013). Online Engagement Factors on Facebook Brand
Pages. Social Network Analysis and Mining, 3(4), pp. 843-861.
Rader, E., & Gray, R. (April 2015). Understanding User Beliefs About Algorithmic Curation in
the Facebook News Feed. Proceedings from ACM Conference on Human Factors in
Computing (SIGCHI). Seoul, Korea.
Rand, D.G. (2011). The promise of Mechanical Turk: How online labor markets can help
theorists run behavioral experiments. Journal of Theoretical Biology, 299, pp. 172-179.
Reychav, I., & Wu, D. (2015). Are your users actively involved? A cognitive absorption
perspective in mobile training. Computers in Human Behavior, 44, pp. 335-346.
Rezab, J. (2014). How KLM Achieved such Stellar Success on Facebook. Social Bakers, Web,
May 1, 2014. Retreived from https://www.socialbakers.com/blog/1873-how-klmachieves-such-stellar-success-on-social-media
Rouis, S. (2012). Impact of Cognitive Absorption on Facebook on Student’s Achievement.
Cyberpsychology, Behavior, and Social Networking, 15(6), pp. 296-303.
143

Saade, R., & Bahli, B. (2005). The impact of cognitive absorption on perceived usefulness and
perceived ease of use in on-line learning: an extension of the technology acceptance
model. Information & Management, 42, pp. 317-327.
Sashi, C.M. (2012). Customer engagement, buyer-seller relationships, and social media.
Management Decision, 50(2), pp. 253-272.
Saxe, R., & Weitz, B. (1982). The SOCO Scale: A measure of the customer orientation of
salespeople. Journal of Marketing Research, 19(3), pp. 343-351.
Schiffman, L.G., Kanuk, L.L., (2007). Consumer Behavior (Ninth ed.). Prentice-Hall Inc, NJ.
See-To, E.W.K., & Ho, K.K.W. (2014). Value co-creation and purchase intention in social
network sites: The role of electronic Word-of-Mouth and trust – A theoretical analysis.
Computers in Human Behavior, 31, pp. 182-189.
Seib, H.M., & Vodanovich, S.J. (1998). Cognitive Correlates of Boredom Proneness: The Role of
Private Self-Consciousness and Absorption. The Journal of Psychology, 132(6), pp. 642652.
Shankar, V. (2015, June 4). Announcing Facebook Lite. Facebook Newsroom. Retrieved from
https://newsroom.fb.com/news/2015/06/announcing-facebook-lite/
Shimp, T., & Sharma, S. (1987). Consumer ethnocentrism construction and validation of the
CETSCALE. Journal of Marketing Research, 24(3), pp. 280-289.
Sirois, F.M. (2014). Absorbed in the moment? An investigation of procrastination, absorption
and cognitive failures. Personality and Individual Differences, 71, pp. 30-34.
Son, J., Jin, B., & George, B. (2013). Consumer’ purchase intention toward foreign brand goods.
Management Decision, 51(2), pp. 434-450.
Srite, M., & Karahanna, E. (2006). The Role of Espoused National Cultural Values in
Technology Acceptance. MIS Quarterly, 30(3), pp. 679-704.
Statista. (2017). Number of daily active Facebook users worldwide as of 1st quarter 2017 (in
millions). Retrieved from https://www.statista.com/statistics/346167/facebook-globaldau/
Terlep, S. & Seetharaman, D. (2016, August 17). P&G to Scale Back Targeted Facebook Ads. The
Wall Street Journal. Retrieved from http://www.wsj.com/articles/p-g-to-scale-backtargeted-facebook-ads-1470760949
Thamizhvanan, A., & Xavier, M.J. (2013). Determinants of customers’ online purchase
intention: an empirical study in India. Journal of Indian Business Research, 5(1), pp. 1732.
Tian, K., & Borges, L. (2011). Cross-Cultural Issues in Marketing Communications: An
Anthropological Perspective of International Business. International Journal of China
Marketing, 2(1), pp. 110-126.

144

Trevino, L. K . and Webster, J. (1992). Flow in Computer-Mediated Communication: Electronic
Mail and Voice Mail Evaluation and Impacts. Communication Research, 19(5), pp. 539573.
Wang, X., Yu, C., & Wei, Y. (2012). Social Media Peer Communication and Impacts on Purchase
Intentions: A Consumer Socialization Framework. Journal of Interactive Marketing, 26,
pp. 198-208.
Wasko, M.M., & Faraj, S. (2005). Why Should I Share? Examining Social Capital and Knowledge
Contribution in Electronic Networks of Practice. MIS Quarterly, 29(1), pp. 35-57.
Watkins, B. (2014). An Integrated Approach to Sports Branding: Examining the Influence of
Social Media on Brand Outcomes. International Journal of Integrated Marketing
Communications, 6(2), pp. 30-40.
Webster, J., & Ho, H. (1997). Audience Engagement in Multi-Media Presentations. Data Base
for the Advances in Information Systems, 28(2), pp. 63-77,
Webster, J. & Martocchio, J.J. (1992). Microcomputer Playfulness: Development of a Measure
With Workplace Implications. MIS Quarterly, 16(2), pp. 201-226.
Webster, J,. Trevino. L. K,. and Ryan, L. (1993). The Dimensionality and Correlates of Flow in
Human-Computer Interactions. Computers in Human Behavior, 9(4), pp. 411-426.
Wei, L., Willnat, L., & Shao, L. (2012). Cultural Differences in the Use of Web. 1.0 and Web 2.0:
A Comparative Analysis of Chinese and American Youth. China Media Research, 8(4),
pp. 77-89.
Weisberg, J., Te’eni, D., & Arman, L. (2011). Past purchase and intention to purchase in ecommerce. Internet Research, 21(1), pp. 82-96.
Wind, Y. (1978) On the Interface Between Organizational and Consumer Buying Behavior.
Advances in Consumer Research, 5, pp. 657-662.
Wirtz, J., den Ambtman, A., Bloemer, J., Horvath, C., Ramaseshan, B., van de Klundert, J.,
Canli, Z.G., & Kandampully, J. (2013). Managing brands and customer engagement in
online brand communities. Journal of Service Management, 24(34), pp. 198-208.
Yoo, B., & Donthu, N. (2001). Developing and validating a multidimensional consumer-based
brand equity scale. Journal of Business Research, 52, pp. 1-14.
Zaglia, M.E. (2013). Brand communities embedded in social networks. Journal of Business
Research, 66, pp. 216-233.
Zailskaite-Jakste, L., & Kuvykaite, R. (2013). Communication in Social Media for Brand Equity
Building. Economics and Management, 18(1), pp. 142-153.
Zhou, T. (2013). Understanding continuance usage of mobile sites. Industrial Management &
Data Systems, 113(9), pp. 1286–1299.

145