2016 49th Hawaii International Conference on System Sciences

Do Facebook Likes Lead to Shares or Sales? Exploring the Empirical Links

between Social Media Content, Brand Equity, Purchase Intention, and
Engagement
Constantinos K. Coursaris Wietske van Osch Brigitte A. Balogh
Michigan State University Michigan State University Michigan State University
[email protected] [email protected] [email protected]

Abstract With one in four people using a social network as

Prior studies report organizations’ increasing of 2013, social media can produce a huge audience
fiscal investments in social media marketing for brands [4; 5]. Consumers interact with Facebook
communications reflecting anticipated benefits for posts on a daily, sometimes hourly, basis making
bottomline improvements. Yet, few studies have social media an ideal platform for brands to foster
looked at the empirical link between messaging close relationships with their consumers. For
dimensions, brand equity, consumer engagement and example, KLM Royal Dutch Airlines has used their
associated purchase intentions. Through a four-week Facebook Page not just to respond to customer
long, between-group longitudinal experiment, we questions and complaints, but also to create
assess the relationships between these factors, and opportunities for user-generated content by inviting
produce the following contributions: (i) developed their fan base to share their own photos from KLM
and empirically supported our proposed monitoring branded airplanes, a tactic that has contributed to
metric, Brand Social Media Engagement Index; (ii) their status as the “most Socially Devoted brand on
observed that brand equity is a full, direct mediator Facebook” [6].
between engaging brand content and both purchase While traditional marketing initiatives and the
intention and brand social media engagement benefits thereof for boosting brand image and
intention; and (iii) demonstrated the positive effect of purchase intention have been studied extensively, the
engaging brand content on consumer brand equity, virtual relationship between brand and consumer has
thereby establishing the value behind brand social yet to receive thorough inquiry in an attempt to link
media marketing communications to consumers. social media communications with actual purchases.
From an organizational standpoint, it becomes crucial
to demonstrate that there is a financial benefit to
social media marketing in order to justify the
1. Introduction increased marketing spending in this digital domain
[3]. [7] showed that public relations professionals
Prior studies report on organizations’ increasing believed that social media marketing is effective for
fiscal investments in social media marketing maintaining brand reputation, awareness, search
communications, reflecting the anticipated benefits rankings, and web traffic, but that increasing the
for an organization’s bottomline [1, 2]. However, number of social media platforms a company is using
while an expected and impactful trend, few studies does nothing to help user satisfaction with their
have attempted to solidify the value proposition of current consumer-brand relationship; thus, leading to
brand social media marketing communications. and reinforcing the importance of answering the
When one considers the increasingly growing role of question does social media offer a brand monetary
social media as a communication tool to reach, value [3].
engage, and interact with target markets, it becomes Traditional offline marketing theories may not be
imperative to be able to show the value behind appropriate for studying online word of mouth
investing employee time in such activities [3]. Hence, communications due to the friend role a brand takes
this study aims to answer the following research on in digital and social media [8]. In particular, as
question: social media provide a platform to bring like-minded
RQ: What is the value proposition of social media consumers together to exchange ideas and opinions
to brand marketing communications? about their favorite brands, the otherwise simple

1530-1605/16 $31.00 © 2016 IEEE 3545

3546
DOI 10.1109/HICSS.2016.444
relationship between brand and consumer becomes The remainder of this paper is organized as
more multifaceted in this social space. follows. First, we offer a review of the theoretical
Previous empirical research—based largely on constructs that compose the conceptual foundation of
content analyses—has suggested that through this study. Then, we discuss the research design of
increased interactions with a brand on social media, this study, including the experimental manipulation,
consumers do not merely build a relationship with the participants, instrument validation, as well as data
company but also with other consumers who are collection and data analyses. Subsequently, we
connected with that brand on social media [9]. present the findings of our study and discuss
Suddenly, it is not just the initial consumer who important implications for theory, future research,
“likes” something but a sense of community—the practice, and social media communications design.
interactions with a group of people who share a
similar interest in the brand—that may result in 2. Literature Review
increased positive attitudes toward the brand [1,
2014). When interactivity is incorporated into social
Moving beyond descriptive analyses, this study media posts, control of the content is given to the
sets out to expose the empirical link between social consumer who in turn is able to pay attention to the
media engagement—i.e., Liking, Commenting, or brand responsiveness of the content as well as other
Sharing—with a brand’s messages on social media consumers in that community [4]. Social media
and brand equity, purchase intention, as well as provides the ideal platform for brands to build social
future engagement intention, a newly introduced experience and symbolic value for the consumer,
construct, with the brand on social media; doing so thereby potentially contributing to an improved brand
advances the field in understanding whether brand image in the consumer’s mind, thus enhancing the
building through improved brand attitude via social consumer’s attitude toward the brand, or brand
media communications generates value for the brand attitude.
[3]. Using an experimental design drawing on brand The extent to which a social media message is
Facebook Page posts characterized by significantly engaging—i.e., resulting in consumer engagement in
different engagement levels (classified as either high the form of Likes, Comments, or Shares—can affect
or low), and pre/post surveys with 1177 unique brand image in two ways. First, an engaging message
responses, this study validates that highly engaging can facilitate the formation of positive brand attitude
social media messages (here, Facebook Page posts) through the manipulation of textual and/or visual
indeed have a significant positive effect (p < .001) on persuasion cues. For instance, a brand can manipulate
brand equity, and in turn significantly impact both content themes (e.g., for a messaging typology, see
consumers’ purchase intention towards the products [1]) as well as the richness of a message (e.g. by
or services offered by the brand as well as including a URL, photo, or video) that increase the
consumers’ intention for future interaction with the level of engagement associated with the message.
brand on social media. Second, an engaging message can positively affect
Based on these findings, this paper offers the brand equity by triggering a positive feeling state.
following contributions. First, this study is the first to Hereto, a brand can manipulate the appeal used in a
empirically test the relationship between social message (transformational as opposed to
media-based brand engagement and purchase informational) to elicit stronger emotional
intention and therewith to expose whether or not experiences and in turn more positive brand attitudes
investments by companies in social media marketing thus enhancing brand equity [10].
communications are associated with a potential return Consumers are able to provide a persuasive effect
on investment. Second, this study proposes a new over other consumers 22 times stronger than
construct in Engagement Intention that aims to marketers [11]. Brands provide a space for
measure an important behavioral outcome in the consumers to engage in thought provoking
context of building brand equity and loyalty in the conversations and inspire resonant stories within this
social space. Third, this study is the first to draw on trusted community of the individual’s social network
actual brand Facebook Page Posts that are classified [12, 4]. Consumption within this community is based
according to their messaging content and objectively on the consumer’s need for problem solving,
evaluated in terms of their impact (i.e. the evoked information search, feedback, and evaluation of
level of Likes, Comments, and/or Shares) before series offered in an environment where real-time
exposing the impact they may have on consumers’ accessibility and exchange of these types of
perceptions and behavioral intentions towards the information is considered the social norm [4]. By
brand. looking at social media posts that have previously

3547
3546
attracted high engagement, we can capitalize on the definition that brand equity is the effect of brand
social interaction of, essentially, strangers within this knowledge on consumer response to brand marketing
brand community and gauge if consumers do in fact to simply, the customer’s perceived value [13]. For
place a higher value on these posted based on their this study, we will use [14]’s definition of brand
community support to accompany the initial marketer equity as the added attractiveness a brand name
contributions [4; 11]. From here, we seek to establish confers on a product or service (see Table 1). We
a connection with consumer brand equity. therefore propose the following hypothesis to provide
additional insight into the link between social media
2.1. Brand Equity, Attitude, and Image communications and brand equity:
H1: The higher the elicited engagement level of a
Brand equity is regarded by organizations as the brand’s post, the more positive the brand equity
incremental utility or value added by the brand name,
which contributes to the company’s long-term 2.2. Purchase Intent
profitability [13]. For consumers, the perspective of
interest within this study, brand equity refers to the Abundant evidence exists in the consumer
added attractiveness a brand name confers on a research literature that the more positive the brand
product or service [14]. Brand equity can therefore be attitude, the higher the consumers’ purchase intention
discussed in relation to the investor, the [14]. Purchase intention here refers to the behavioral
manufacturer, the retailer, or the consumer, and one inclination of consumers to plan to purchase a certain
must consider the perspective to be used for analysis product or service in the future [18]. Positive
[15]. Furthermore, it is composed of two related purchase intention, in turn, is viewed as an important
concepts: brand image and brand attitude. Hence, any antecedent to actual purchase action [19].
attempts to measure brand equity must recognize this However, asking purchase intent questions has
multidimensional nature [16,14], and the been shown to prime purchase processing decisions
abovementioned two dimensions will be defined in favorable ways toward the brand. In a 1993 study
next. by [20], purchasing rates increased by 84 percent for
Drawn from both the tangible and intangible armature personal computer shoppers just by
associations a consumer makes between a cluster of measuring their purchase intent. A 1996 study by
attributes and the brand name, brand image is defined [21] found that simply asking intent questions caused
as the consumer’s perception of a specific brand [14; purchase rates to increase for brands that already had
16]. Few marketing studies have examined the high market shares. Consumers in the late states of
connections between brand image and brand equity, the decision-making process are more likely to have
and how they in-turn drive a consumer’s purchase already formed attitudes and purchase intention thus
intention; even fewer studies have explored these prompting retrieval of preexisting cognitions toward
relationships in the social media realm, thus the brand when asked about purchase intent [21].
increasing the value of this study [16]. Brand attitudes are a precursor for customer
The second dimension of brand equity, i.e. brand activities such as recommending the brand to others
attitude, is a consumer’s internal evaluation of a as well as making repeat purchases [22]. This is
brand [14]. In essence, that evaluation becomes a consistent with brand equity being an asset to the
function of salient beliefs about the brand, those from business and thus leading to more positive future-
memory and considered by the consumer in a given term financial performances [14; 3] through
situation [16]. Brand attitude is one of the most incremental sales. Therefore, we propose the
widely examined constructs in consumer behavior following hypothesis:
and an important concept in marketing research for H2: The more positive the brand equity, the
the past twenty years. By better understanding the greater the consumer’s brand purchase intention
underlying causal dynamics of attitude formation and
image interpretation, organizational managers are 2.3. Social Media Engagement Intention
equipped with greater clarity that can inform
successive, and likely more effective, marketing Consumers do not forward or participate in the
communications [17]. curation of social media content unless there is
Brand equity provides the insulation needed to something for them to build on their own profile or
protect brands from competitors [15]. High equity strengthen their personal brand image [5]. In addition
has been associated with consumer satisfaction, brand to the link between brand equity and purchase
preference, premium price, and high profit values intention, we propose that brand equity will
[14]. Definitions have ranged from Keller’s 1993 positively affect consumers’ intentions to engage

3548
3547
 with a brand’s social touch points. Such engagement 3. Proposed Theoretical Model
may take the form of Liking, Commenting, or
Sharing a brand’s profile (e.g., Facebook Page) or Based on the above discussion of the four
message (e.g. Facebook Page Post). In essence, if a constructs comprising this study—namely engaging
brand creates engaging content resulting in positive brand messaging on Facebook Pages; brand equity;
impressions about and attitudes toward the brand, we purchase intention; and brand social media
anticipate that consumers will be more likely to engagement intention—as well as the associated
continue visiting the brands’ social media platforms three hypotheses, Figure 1 summarizes the proposed
and continue to interact with posts generated by the research model that will be tested in this
brand. experimental study.
Building engagement intention requires a shift
from marketer led brand messaging to include the Figure 1. Proposed Research Model
consumer in the messaging creation strategy and
establishing a co-creation of meaning [4]. 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 [4]. Such interaction increases brand tangibility
and could result in repeat consumption of the brand 4. Methodology
as well as interaction with the social media
community established by the brand [4]. Still, we 4.1. Manipulation
lack an understanding of both approaches to
measurement and assessments of whether or not To explore the effect of consumer engagement
social media message consumption actually increases with brand Facebook Page posts on brand equity,
purchase intention and/or further social media-based purchase intention, and intentions to engage with a
engagement intention. Thus, we propose the brand’s social media presence, actual posts were
following hypothesis: selected from three brands, namely Delta Airlines,
H3: The more positive the brand equity, the Walmart, and McDonald’s, to allow for the most
greater the brand social media engagement intention diverse category selection possible. These brands
were chosen for three reasons. First, they are among
Construct Definition Reference the top 110 Fortune companies ranked by gross
Brand The added [14] revenue, enjoy strong brand equity, and maintain a
Equity attractiveness to the considerable social media presence. Second, these
customer that a brand brands represent different levels of purchase-decision
name confers on a involvement, as follows: low involvement as found
product or service. for McDonald’s and other fast-moving consumer
Brand Image The consumer’s [14] goods (FMCGs), mixed involvement as is the case
perception of a with Wal-Mart and other Big Box Retailers, and,
specific brand high purchase-decision involvement found in Delta
Brand The consumer’s [14] Airlines and other expensive and greater risk-bearing
Attitude internal evaluation of goods and services. Third, the purchase cycle was
a brand one additional consideration in our case selection
Purchase The degree to which a [18] process resulting in a selection of brands representing
Intention consumer considers both consumables and durables. Purchase cycle—
purchasing a product. consumables versus durables—is a key factor in
Brand Social The degree to which a Defined and brand messaging when considering purchase-decision
Media consumer considers developed frequency and the need for tailored messaging.
Engagement engaging with a for this Selection of actual Facebook Page posts by the
Intention brand’s social media study above three brands was then performed by referring
touch point. to the messaging typology proposed by [1] and
Table 1. Constructs and Definitions controlling for the content of posts. Four content
categories—Brand Awareness, Product Awareness,
(Product) Promotional, and Engagement—were
identified as being most closely aligned with the four

3549
3548
stages of the Buying Behavior Model: Consider, From there, participants were randomly assigned
Search, Choose, and Buy [1]. For each of these into either the low or high level of Facebook Page
categories, two brand Facebook Page posts were post engagement group. For the next four weeks,
selected that were significantly different in the level participants took part in short experimental sessions
of elicited engagement, so as to serve as the in a computer lab, where they were exposed to
manipulation in either the low or high engagement Facebook Page posts from McDonalds, Walmart, and
condition. All identifiable metrics – including the Delta Airlines, corresponding to the engagement
number of Likes, Comments, and Shares actually level appropriate for the treatment group they were
elicited - and dates associated each brand Facebook randomly assigned to. The rank-order in which brand
Page post were removed prior to including them in Facebook Page posts were presented was randomized
this experiment. to avoid response biases. After each brand exposure,
To calculate the level of elicited engagement, we they were asked to respond to a set of survey
summed the raw number of Likes, Comments, or questions for each Facebook Page post in regards to
Shares associated with each post to determine an their brand perceptions (i.e. image and attitude),
absolute engagement score for an individual brand purchase intention, and future engagement intention
Facebook Page post. Hence, eight actual brand posts with brand social media.
were selected from each official Facebook Page of
Delta Airlines, Walmart, and McDonald’s (n=24), 4.3. Instrument Validation
four that elicited a low level and four that elicited a
high level of engagement. These eight posts were The questionnaire used in this study consists of
deliberately selected not only on the basis of their scales measuring the constructs from the research
engagement scores, but also to ensure internal model, as summarized in Table 2. One formative
consistency with respect to all characteristics of the construct was used, namely brand equity, while both
post except engagement, namely the post’s content of the reflective constructs in the research model, i.e.
category and subcategory as well as media type (e.g. purchase intention and brand social media
does the post include text only or a URL, photo, or engagement intention, had significant factor loadings
embedded video). greater than 0.5 to ensure construct validity [23] and
To illustrate, if a highly engaging Brand were further validated by adequate item-to-total
Awareness (subcategory: Operations) Facebook Page correlations at above the 0.35 threshold as suggested
post from Delta Airlines that included a photo was by [24].
selected, another post with the exact same message Previously validated scales for brand attitude and
characteristics (content = Brand Awareness; brand image were used to measure the second order
Operations; richness = photo) was selected that construct of brand equity (see Figure 2). Brand
displayed a low engagement score. As a manipulation attitude was adapted from the scale used by [17].
check, we performed a one-tailed t-test analysis of This scale captures respondents’ perceptions about a
the 24 brand posts (12 in the low engagement brand along seven-point scales anchored between two
condition vs. 12 in the high engagement condition) to semantic differential adjective pairs, namely
determine if the brand Facebook Page posts in the “Bad/Good” and “Extremely Dislike/Extremely
two conditions were indeed significantly different in Like.” With respect to the construct of brand image,
the extent to which they generated engagement with it was adapted from the scale used by [16]. This scale
the post in the form of Likes, Comments, and/or captures respondents’ perceptions about a brand
Shares. The result (p=0.038) offers support for the along seven-point scales anchored between five
significantly different levels of engagement evoked semantic differential adjective pairs, namely
between the low and high engagement conditions. “Colorless/Colorful,” “Dominating/Submissive,”
“Excitable/Calm,” “Thrifty/Indulgent,” and
4.2. Participants “Modest/Vain.”

Participants were recruited from a large Figure 2. Proposed Second Order Construct
Midwestern institution and asked to complete a for Brand Equity, Construct Scales
baseline survey 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, baseline perceptions were assessed
for brand image, attitude, and purchase intention..

3550
3549
 The purchase intention construct was adapted As shown in Table 3, discriminant validity was
from [18] and measured along seven-point scales supported by confirming that the square root of the
anchored between three semantic differential variance shared between a construct and its items was
adjective pairs, namely “Unlikely/Likely,” greater than the correlations between the construct
“Improbable/Probable,” and “Impossible/Possible.” and any other construct in the model [26].
Brand social media engagement intention was
defined in this study by considering the entire set of a Construct Brand Purchase Brand Social
user’s potential interaction behavior with a brand Equity Intention Media
Facebook Page post, and creating a corresponding Engagement
Intention
scale item, for a total of five items. These items were
Brand Equity n/a
measured along seven-point Likert scales anchored as (formative)
“Highly Unlikely,” “Unlikely,” “Somewhat Purchase 0.615 1.00
Unlikely,” “Neither Likely Nor Unlikely,” Intention
“Somewhat Likely,” “Likely,” and “Highly Likely.” Brand Social 0.430 0.231 1.00
Media
During the questionnaire, each construct’s items Engagement
were randomized to prevent systemic response bias. Intention
Tests for nonresponse and common method biases * Diagonal values are the square root of the AVE
were not significant, revealing that no such biases Table 3. Latent Variable Correlations
existed. The factor loadings for all items were above
.847 (and thus above the .7 cut-off for convergent Finally, as brand equity is a formative construct,
reliability) used in this study are summarized in validity of these constructs cannot be assessed
Table 2. through an inspection of convergent validity,
discriminant validity, and cross-loadings, but rather
Construct Latent Cronbach Composite Convergent Discriminant should be assessed by evaluating the Variance
Mean Alpha (α) Reliability Validity Validity
(AVE) (√AVE)
Inflation Factors (VIFs). Using SPSS, the VIF of
Brand Brand Attitude and Brand Image was computed and
0.431 Formative Formative Formative Formative
Equity revealed an “excellent value” of 1.660 for both, this
Purchase
Intention
0.616 0.913 0.945 0.851 0.922 is below the acceptable threshold of 3.3 [28] and well
Brand below the commonly acceptable value of 10 [28].
Social
Media
1.954 0.923 0.941 0.763 0.873
Engage- 5. Results
ment
Intention
Using the variance-based PLS method, we then
Table 2. Construct Statistics tested the research model shown in Figure 1. PLS
allowed us to specify the relationships between the
The quality of measures was tested with the various endogenous and exogenous constructs in the
partial least squares (PLS) approach to structural model (structural model), as well as with their
equation modeling via SmartPLS 2.0. The results of underlying items (measurement model). Thus, data
the tests for convergent validity [25] discriminant analysis provided support for both how well the items
validity [25, 26], construct means, and Cronbach’s α measured each construct, and how well the
can be found in Table 3, which shows that all hypothesized relationships between constructs
constructs had adequate Cronbach’s α (i.e. > 0.80), supported the theory.
reliability (i.e. above 0.7 threshold according to [27]), The bootstrapping re-sampling method with 1,177
and convergent validity (i.e. AVE above the 0.5 cases and 2,000 re-samples were used for structural
benchmark according to [26]. model estimation. Based on the results of the model
estimation, all of the hypotheses were supported (see
Table 4). The results of the structural model
estimations, including R2 values, are presented in
Figure 3.

3551
3550
Hyp. From To Path t- Status intention, and intention to engage with future social
Coeff. Value media content produced by the brand.
1 Engaging Brand 0.074 2.539* Supported
Brand Equity * 5.2 Post-Hoc Test: Purchase Intention, Brand
Content
Social Media Engagement Intention Effects
2 Brand Purchase 0.615 33.024 Supported
Equity Intention ***
3 Brand Brand 0.430 21.874 Supported Following the validation of the proposed model,
Equity Social *** we wanted to further explore if any effects occurred
Media between the purchase intention variable and the brand
Engage- social media engagement variable. While no
ment literature pointed to either of these effects being a
Intention possibility due to the introduction of Brand Social
* significant at p = .05 ** significant at p = .01 *** significant at p = .001
Media Engagement Intention as a construct in this
Table 4. Hypothesis Testing paper, one could reason that an effect is possible. A
consumer could be trying to reinforce their purchase
Figure 3. PLS Model decision by commenting on it in social media or, the
reverse direction, social media interaction could
spawn purchase intent. Alas, neither effect was
shown as significant. When considering Purchase
Intention’s effect on Brand Social Media
Engagement Intention, this effect was found to be not
significant (p=0.110). Conversely, Brand Social
Media Engagement Intention’s effect on Purchase
5.1 Post-Hoc Analyses of Experimental
Intention was not found to be significant (p=0.191).
Manipulation Hence, we are to conclude that one ‘intention’ (i.e.
purchase vs. engagement) does not drive the other.
To conclude if the effect on brand equity is
indeed due to the experimental treatment—i.e., the
manipulation of the level of engagement—we 6. Discussion
conducted a t-test comparing brand equity
assessments in the pre-experiment (i.e., baseline) The current industry emphasis on qualifying the
survey versus the post-experiment survey as well as value proposition of social media activities and
between the low and high treatment conditions (i.e., quantifying the return on investment in such activities
those exposed to low engaging versus high engaging provided the impetus for this study. Brands are
messages). B increasingly relying on the use of metrics, such as the
For the comparison of brand equity scores number of Facebook Page fans, to evaluate the
between pre- and post-tests, we found that there was success of brand communications on this space. The
a significant difference for participants in the high first aim of this study set out to empirically test the
engagement condition (p = .000), but not for those in relationship between social media-based brand
the low engagement condition (p= .054), thus engagement and purchase intention while
lending support for the fact that it is not just establishing engagement metrics. To the best of our
continued exposure to messages, but rather continued knowledge, this paper is the first attempt at offering
exposure to engaging messages that has a positive empirical evidence of this relationship and the
effect on brand equity. mediating role of brand equity. As reported in the
Furthermore, a one-way ANOVA for all three Results section above, support was obtained for the
dependent variables in the model showed that positive effect of a brand’s Facebook Page post
participant in the high engaging treatment group engagement level on brand equity, as well as on the
displayed significantly higher levels of brand equity latter’s dimensions of brand image and brand
(F=4.787; 0 = .03), purchase intention (F = 3.988, p = attitude.
.05), and future brand social media engagement Second, and building on prior research, our
intention (F=3.923; p = .05) than those in the low findings offer support for the positive effect of brand
engaging treatment group. This further validates that equity on purchase intentions. Third, this study
the more engaging a brand’s social media content, the proposes a new construct in brand social media
more positive a person’s attitudes toward the brand in engagement intention that measures an important
terms of overall brand image and equity, purchase behavioral outcome—continued intention to engage
with the brand on social media through Liking,

3552
3551
Commenting, and/or Sharing—in the context of not eliminate any participants based on an initially
building greater brand equity and loyalty. There was low brand attitude or brand image scores.
strong support for the positive effect of brand equity This study has made significant strides in
on consumers’ intentions to engage with a brand on advancing our understanding of social media in the
social media. context of social media-based marketing
In addition, an interesting finding emerges from communications and contributes both to theory and
the results reported earlier. First, we observe that the practice. With respect to theory, three main
direct effect of brand equity on purchase intention is contributions are put forth. First, we have extended
approximately twice as strong as that on brand social our understanding of purchase intention antecedents
media engagement intention. Put simply, the greater by identifying an important factor in brand social
equity a brand enjoys in the consumer’s mind, the ( media messaging engagement. This antecedent may
~twice) more likely outcome is a future be considered as the digital behavioral manifestation
purchase than a subsequent interaction with the of a consumer’s positive attitudes and associations
brand’s social media content in the future. This can with a brand. Second, by proposing a new
be explained by revisiting the definition of each construct—brand social media engagement
construct and when one considers the alignment intention—this study further identifies an important
between equity and the target behavior: purchasing digital proxy for brand equity and loyalty. Third, by
from the brand is highly aligned with the “added drawing on actual brand Facebook Page Posts that
attractiveness to the customer that a brand name were classified according to their messaging content,
confers on a product or service”, while the latter is message richness, message appeal, and objectively
only a subset of the numerous considerations a social evaluated in terms of their impact (i.e. the evoked
media user entertains and in particular the extent to level of Likes, Comments, and/or Shares), this study
which the brand social media content in itself is sheds further light on the theoretical mechanisms
likely to contribute to the consumer’s personal brand through which the level of engagement associated
upon Liking, Commenting, and/or Sharing the Page with a post affects brand attitude, namely both
post and the outcomes that follow from his through the manipulation of visual and textual
association with both the brand and the post content. content as well as through the instigation of positive
This conclusion is further supported by the high level feelings emerging from the message’s
of variance explained (R2=37.8%) in purchase transformational appeal. With respect to practice, as
intention (R2=37.8%) contrasted with the lower yet aforementioned, this study helps to empirically
significant level of variance explained in brand social validate the link between social media engagement,
media engagement intention (R2=18.5%). Hence, brand attitude, and in turn purchase intent. Social
brand Community Managers would be wise to design media marketing is not just a platform; it is an area of
and disseminate content that is not only aligned with communication on a deeper level with a brand’s
the marketing communication objectives but is also audience [5]. With the amount of social network
crafted in a manner that will motivate the consumer’s users expected to almost double from 1.7 billion
interaction with it – whether that is because of its people in 2013 to 2.55 billion people in 2017, it is
visual, aesthetic, or otherwise hedonic appeal or its with due cause that firms learn the ins and outs of
informational, practical or otherwise utilitarian value their effects of using this type of communication with
proposition. their consumers [4; 17; 16].
Additional explanation of possible variance for Although attempts to monetize Facebook Likes
the constructs can be attributed to environmental and other forms of social media engagement are still
factors outside of the focus of this study. We myopic in our view, this paper does show that social
concentrated on proactive, brand-initiated media messages can have a positive effect on
communications for this study. It is important to note consumers’ brand attitude and in turn on their
that brand Facebook Page posts selected for the intention to purchase products or services offered by
engaging brand content stimulus condition of this the brand. However, the most interesting practical
study were manually identified to ensure the elicited implication is that it is not continuous exposure to
engagement was positive in nature, thus essentially social media messages, but rather the exposure
sanitizing the data against negative responses. Should overall to strong, engaging messages that has a
a participant have a preexisting negative attitude positive effect on brand attitude and in turn purchase
toward a brand, this was not accommodated for. We intent. This became evident from the fact that
used a t-test to check for brand equity improvement respondents in the low engagement condition did not
over the course of the four-week long study and display a significant change in brand attitude
being exposed to engaging brand content but we did following continued exposure to brand messages,

3553
3552
whereas respondents in the high engagement [2] Moorman, C. 2013. “The CMO Survey. Highlights and
condition displayed significantly more positive brand Insights, February 2013.” Retrieved on August 24, 2013,
attitudes following their repeated exposure to from:
engaging brand messages. Finally, another important http://cmosurvey.org/files/2013/02/The_CMO_Survey_Hig
hlights_and_Insights_Feb-2013-Final2.pdf.
practical implication is that brand attitude not only
affects consumers’ intention to purchase from the [3] Aaker, D.A., & Jacobson, R. 2001. “The Value
brand, but also their intention for continued Relevance of Brand Attitude in High-Technology
engagement with the brand, thus potentially resulting Markets,” Journal of Marketing Research (38:11), pp. 485-
in a virtuous cycle of positive consumer engagement, 493.
brand attitude development, purchase intention, and
ultimately brand loyalty. [4] Davis, R., Piven, I., & Breazeale, M. 2014.
“Conceptualizing the brand in social media community:
The five sources model,” Journal of Retailing and
6.1 Challenges and Future Research Consumer Services (21), pp. 468-481.

The main limitation of the current study was the [5] Hosea, M. 2011, January 6. “Engagement takes place in
use of a student sample. Hence, future research a shared environment,” Marketing Week, pp. 30-33.
aiming to replicate the findings of this study should Retrieved August 14, 2014 from EBSCOhost database.
use a national random sample to further validate the
relation between social media engagement and brand [6] Rezab, J. 2014. “How KLM Achieves such Stellar
attitude. Success on Facebook,” Social Bakers, Web, May 1, 2014.
Another limitation of this study is the inclusion of
[7] Agozzino, A. 2012. “Building a Personal Relationship
only a limited set of brand-related variables.
through Social Media: A Study of Millennial Students’
Although intended to enhance the parsimony of the Brand Engagement,” Ohio Communication Journal (50),
current study, future studies could replicate a similar pp. 181-204.
experiment, but include additional constructs, such as
brand preference and brand loyalty [14]. [8] Brown, J., Broderick, A.J., & Lee, N. 2007. "Word of
Although experiments are useful in establishing Mouth Communication within Online Communities:
the causal link between social media engagement, Conceptualizing the Online Social Network," Journal of
brand equity, and in turn purchase intention, they are Interactive Marketing (21:3), pp. 2-20.
also inherently limited due to their artificial nature.
[9] Hollenbeck, C.R., & Kaikati, A.M. 2012. “Consumers'
Furthermore, although purchase intention is a
use of brands to reflect their actual and ideal selves on
significant and strong predictor of actual purchase Facebook,” International Journal of Research in Marketing
behaviors, such behaviors cannot be measured in an (29), pp. 395-405.
experimental study. Hence, future studies should aim
to further assess the correlation between social media [10] Coursaris, C.K., Van Osch, W., Balogh, B.A.,
engagement and purchase behaviors, for instance, Quilliam, E.T. 2014. “Social Media Marketing:
through the longitudinal regression of social media Investigating Empirical Links Between Purchase
engagement metrics and transactional data associated Involvement, Strategy, Content, and Media Type,”
with a brand. Proceedings of the 2014 Annual Conference of the
American Academy of Advertising, Atlanta, Georgia.

7. Acknowledgement [11] Goh, K.Y., Heng, C.S., & Lin, Z. 2013. “Social Media
Brand Community and Consumer Behavior: Quantifying
This research was supported by Leo Burnett Detroit the Relative Impact of User- and Marketer- Generated
(http://www.lbdetroit.com). The content is solely the Content,” Information Systems Research, pp. 88-107.
responsibility of the authors and does not necessarily
represent the official views of Leo Burnet Detroit. [12] Briggs, T. 2010. “Social Media’s Second Act: Toward
Sustainable Brand Engagement,” The Design Management
8. References Institute, pp. 46-53.

[13] Chen, C.F. & Chang, Y.Y. (2008). Airline brand

[1] Coursaris, C.K. and Van Osch, W. 2013. “A Contextual
equity, brand preference, and purchase intentions - The
Messaging Framework: Informing the Design of Effective
moderating effects of switching costs. Journal of Air
Social Media Marketing Messages.” Proceedings of the
Transport Management, 14, 40-42.
Twelfth Pre-ICIS SIG-HCI Workshop 2013, Milan, Italy.

3554
3553
[14] Chang, H.H., & Liu, Y.M. (2009). The impact of [28] Diamantopoulos, A., & Siguaw, J. A. 2006.
brand equity on brand preferences and purchase intentions “Formative versus reflective indicators in organizational
in the service industries. The Service Industries Journal, measure development: a comparison and empirical
29(12), 1687-1706. illustration,” British Journal of Management, (17:4), pp.
263-282.
[15] Cobb-Walgren, C.J., Ruble, C.A., Donthu, N. (1995).
Brand Equity, Brand Preference, and Purchase Intent.
Journal of Advertising, 24(3), 25-40.

[16] 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.

[17] 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. 318-332.

[18] 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.

[19] Fishbein, M., Ajzen, I., 1975. Belief, attitude,

intention, and behavior: an introduction to theory and
research. Addison-Wesley Publishing Company, MA.

[20] Morwitz, V.G., Johnson, E., & Schmittlein, D. 1993.

“Does Measuring Intent Change Behavior?” Journal of
Consumer Research (20:6), pp. 46-61.

[21] 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.

[22] 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.

[23] 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.

[24] Saxe, R., & Weitz, B. 1982. “The SOCO Scale: A

measure of the customer orientation of salespeople,”
Journal of Marketing Research (19), pp. 343-351.

[25] Bagozzi, R. 1981. “Evaluating structural equation

models with unobservable variables and measurement
error: A comment,” Journal of Marketing Research (18:3),
pp. 375-381.

[26] Fornell, C., and Larcker, D. 1981. “Evaluating

Structural Equation Models with Unobservable Variable
and Measurement Error,” Journal of Marketing Research
(18), pp. 39-50.

[27] Nunnally, J. C. 1978. Psychometric Theory, 2nd

edition, New York: McGraw-Hill.

3555
3554