International Journal of Information Management 36 (2016) 360–373

Contents lists available at ScienceDirect

International Journal of Information Management

journal homepage: www.elsevier.com/locate/ijinfomgt

Predicting positive user responses to social media advertising: The

roles of emotional appeal, informativeness, and creativity
Jieun Lee, Ilyoo B. Hong ∗
College of Business and Economics, Chung-Ang University, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: As social network services become more pervasive, social media advertising emerges as an attractive
Received 8 May 2015 vehicle for augmenting advertising effectiveness. To leverage this new means of marketing, one must
Received in revised form 8 December 2015 understand what engages SNS users in a favorable online behavior (i.e., overtly indicating personal inter-
Accepted 6 January 2016
est in, or support for, the exposed message by clicking the Like or Share button in Facebook), thereby
Available online 21 January 2016
resulting in an effective advertising campaign. This research conceptualizes SNS ad effectiveness as a
concept encompassing emotional appeal, informativeness and creativity that all have a potential to con-
Keywords:
tribute to a positive online behavior. It empirically investigates the antecedents of positive user behavior
Social media advertising
Social networking site
for a SNS ad based on the theory of reasoned action, the social influence theory, and a persuasion the-
Empathy ory. It proposes and tests a conceptual model of the formation of online user’s behavioral responses with
Social influence regards to SNS advertising. The results of our empirical tests of the model reveal that informativeness and
Viral behavior advertising creativity were key drivers of favorable behavioral responses to an SNS ad and that intention
to engage in favorable user responses was positively associated with purchase intention. Based on these
findings, the paper suggests further research directions and offers implications for harnessing the full
potential of the new SNS advertising platform.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction because tracking their outcomes is hard. However, the extent to

In recent years, the increasingly widespread use of social
networking sites (SNS) has gradually shifted the advertisement
platform used by business firms from traditional media to social
media. Strong growth in mobile advertising is expected to increase
the use of digital ads, which represented up to 30% of all U.S. ad
spending in 2014 (eMarketer.com, 2014). Social media marketing
enables word-of-mouth (WOM) advertising campaigns, inducing
brand awareness and increasing sales revenue (Godes & Mayzlin,
2009; Kumar & Mirchandani, 2012), and also promotes a firm’s rela-
tionships with prospects, which makes it easy to accurately target
consumers. Social media marketing has thus attracted much atten-
tion from businesses seeking to generate increased consumption of
their commercial offerings (Wikipedia.org, 2015b).
One challenge facing marketing professionals considering cam-
paigns on SNS is ensuring that the social network ad in question
serves its goals effectively. Many marketing professionals find it
difficult to measure the effectiveness of social media campaigns
∗ Corresponding author at: Bubhakkwan Bldg., Rm. 1404, 221 Heuksuk-dong,
Dongjak-ku, Seoul 156-756, Republic of Korea. Fax: +82 2 813 8910.
E-mail addresses: jlee114@cau.ac.kr (J. Lee), ihong@cau.ac.kr (I.B. Hong).
which a social media advertisement is producing the intended ben-
efits can be gauged by examining the interactions (e.g., the number
of “Likes” or “Shares”) among social network users to assess their
perceptions of the ad (Burg, 2013). The bottom line is that in order
for a SNS ad to be considered effective, it should be able to lead a
user to favorably respond to the ad (e.g., publicly expressing his/her
support for the ideas conveyed by, or for the mood generated by, the
ad). User’s positive responses on a social network are likely to help
spread the advertisement message among the users, contributing
to the increased awareness of the featured brand. According to a
related research report (Gustin, 2012; Lipsman, Aquino, & Kemp,
2012), not only ‘exposed fans’ but also ‘friends of fans’ were found
to spend more after viewing ads on Facebook. Presumably, ads that
can induce positive perceptions and responses among social net-
work users would likely create the WOM effect, and thus result in
greater sales yielding high return on investment in SNS ad cam-
paigns.
In this regard, present research has been motivated by two
research questions (RQ): what are the qualities of a SNS ad that help
produce positive user responses (RQ1), and is positive user behavior
with regards to a SNS ad a predictor of positive purchase behav-
ior (RQ2)? Extant research has not systematically addressed these
issues. Research identifying desirable qualities of a SNS ad will pro-

http://dx.doi.org/10.1016/j.ijinfomgt.2016.01.001
0268-4012/© 2016 Elsevier Ltd. All rights reserved.
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
vide useful implications for producing positive user responses, and
hence intended business benefits of the ad. A few studies (Ducoffe,
1996; Gao & Koufaris, 2006; Logan, Bright, & Grngadharbatla, 2012;
Taylor, Lewin, & Strutton, 2011) have examined the predictors of
user attitudes to SNS advertising, but they offer little assistance
in predicting attitudes to the positive user responses that help
improve the brand awareness on a social network. It is imperative
to understand what drives users’ favorable online behavior (e.g.,
clicking the Like or Share button for a SNS ad) if a firm is to increase
the potential effectiveness of an ad campaign. In this research, we
use the term ‘positive user behavior’ to refer to a user’s behav-
ioral responses associated with disclosing and spreading the user’s
personal interest in, and/or support for, a SNS ad.
The main purpose of this paper is to identify the antecedents of
the behavioral intention to engage in positive user behavior for an
SNS advertisement and propose a conceptual model of how posi-
tive user responses to an advertisement are formed on a SNS. The
conceptual model is built around the theory of reasoned action,
the social influence theory, and a persuasion theory. We thus
gain insight into the predictors of viral behavioral intention with
respect to SNS advertisement and the impact of these predictors
(e.g., attitude, subjective norm, herd behavior, privacy concerns,
etc.) on consumers’ behavioral responses including purchase inten-
tion. Predictors believed to play a sizable role in forming users’
behavioral responses will be drawn from the related literature. The
findings of this research will offer useful implications for firms seek-
ing a favorable return on investments from their SNS campaigns.
The rest of the paper is organized as follows. First, we review
the related literature to gain a basic understanding of underly-
ing concepts that will be used to construct a theoretical model
and hypotheses. The literature on empathy expression, advertis-
ing effectiveness, social influences, and privacy concerns would
provide insights into potential antecedents to the viral behavioral
intention. Next, we introduce the data collection method, sample,
and measures used to test the model. Then, we present the results
and discuss the key findings of the study. Finally, we provide impli-
cations as well as limitations and future research directions.
2. Theoretical background and hypotheses
One theoretical backbone that underlies the present research is
the theory of reasoned action (Fishbein & Ajzen, 1975). The the-
ory of reasoned action (TRA) posits that an individual’s behavior is
conceptualized as a consequence of the behavioral intention that
is affected by both the individual’s attitude concerning the behav-
ior and the subjective norm concerning it (Fishbein & Ajzen, 1975).
Over the last decade, numerous studies have modeled user behav-
ior on SNS based on the TRA. However, these studies have focused
on either SNS use (Kwon & Wen, 2010; Pelling & White, 2009;
Sledgianowski & Kulviwat, 2008) or continued SNS use (Choi, 2013)
with little attention paid to user behavior concerning SNS ads. In
our research, we use the TRA to model the user’s online behav-
ior associated with a SNS ad. The TRA provides a powerful model
to predict consumer’s volitional behavior and intentions based on
salient information and beliefs about the likelihood that perform-
ing a particular behavior will lead to a specific outcome (Madden,
Ellen, & Ajzen, 1992). And user’s online behavior including reveal-
ing personal reaction to a SNS ad is regarded as a type of volitional
behavior that can be adequately modeled and predicted using infor-
mation and beliefs about the use of a SNS. Specifically, a positive
user response to a SNS ad is viewed as a behavioral outcome jointly
formed by the user’s attitude toward the user response and the sub-
jective norm about that user response. It is presumed further that
the user’s attitude is determined by beliefs a consumer may have
as to the effectiveness of a SNS ad, such as emotional appeal.
361
Meanwhile, user behavior associated with SNS ads may also be
explained from the perspective of the social influence theory. SNS
advertising differs from conventional advertising in the way the
advertising content spreads among the individuals whom the ad is
targeted at. Electronic word of mouth (eWOM) enabled by a SNS
accelerates the dissemination of the advertising message over a
social network. Interpersonal influence that exists in a SNS serves
as a driving force for users’ engagement in eWOM so that SNS users’
online dissemination behavior including opinion giving, opinion
seeking and pass-along responses tend to be driven by the acts
of peer users (Chu, 2009). While the TRA includes a social influ-
ence construct called subjective norm that denotes an individual’s
perception of the pressure to conform to the expectations of impor-
tant others, the theory is not extensive enough to incorporate the
bandwagon effect that refers to the tendency that the probability of
individual adoption increasing with respect to the proportion who
have already done so (Colman, 2014; p. 77). It is important that
both dimensions be taken into consideration to assess the social
influence.
In summary, the TRA and the social influence theory rep-
resent two prime theories used in this research to predict the
user’s positive responses to a SNS ad. The theoretical framework
underpinning the present research encompasses both internal and
external drivers of the behavioral intention to express empathy
(e.g., to click the Like or Share button) with regards to a SNS ad.
Internal drivers are a user’s own evaluations of the outcome of a
given behavior, while external drivers are the user’s perceptions
of other people’s behavior or evaluations such as social pressures
inducing the user to perform that behavior. The TRA provides both
an internal driver (i.e., attitude) and an external driver (i.e., sub-
jective norms). However, the external driver incorporated into the
TRA is not comprehensive in that it only includes normative social
pressure without paying attention to informational social pressure.
Thus, we combined the TRA with the social influence theory so
as to include both informational and normative social influences.
The resulting research model now incorporates relevant internal
and external drivers that are likely to influence the intention about
empathy expression.
In addition to the two theories mentioned above, this research
employs a theory of persuasion called elaboration likelihood model
(ELM) to illuminate the influence of emotion on attitude change.
Petty and Cacioppo (1986) view persuasion as a process in which
the success of influence depends chiefly on the way the receiver
makes sense of the message. They state that ELM introduces two
distinct pathways that one can use when trying to persuade others.
Central route is associated with cognitive processing of the mes-
sage content, while peripheral route relies more on peripheral cues
(e.g., speech tones, credibility of the persuader, etc.) for a short
cut than on the content itself. The ELM theorizes that peripheral
messages appealing to a given peripheral cue can enable powerful
influence especially when message receivers are either uninter-
ested in, or unable to process, the content (Petty & Cacioppo, 1986).
Given peripheral route processing, emotional appeals may make a
positive impact. Thus, the ELM will provide a theoretical guideline
in hypothesizing the relationship between emotional appeal and
attitude toward empathy expression.
Given the aforementioned research aim, it was imperative that
we identify key constructs that play a role in leading to a positive
user evaluation of the empathy expression behavior. A review of
the literature resulted in the research model shown in Fig. 1, which
outlines the antecedents and their impact on behavioral intention
to express empathy about the given ad. Hypotheses H1, H6, H7
and H9 involve the relationships between the behavioral inten-
tion to express empathy and its antecedents. As the TRA posits, the
behavioral intention is influenced by attitude and subjective norm.
The model incorporates the positive impact of social influences
362 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
Fig. 1. Structural model.
(i.e., perceived herd behavior and subjective norm) as well as the
negative impact of privacy concerns on the behavioral intention.
Perceived herd behavior, a construct of informational social influ-
ence, denotes the tendency for individuals to mimic the actions,
whether rational or irrational, of a larger group. And privacy con-
cerns refers to an individual’s concerns over the possible exposure
of private information. As hypotheses H6, H7 and H8 depict, per-
ceived herd behavior is likely to both directly and indirectly affect
intention to express empathy; note that subjective norm plays the
mediating role between perceived herd behavior and intention to
express empathy. Next, the three hypotheses including H3 through
H5 encompass relationships between attitude and the related atti-
tudinal beliefs. Relating these hypotheses to the research questions
presented earlier, hypotheses H1 and H3 through H8 will help
address the RQ1, whereas hypothesis H2 is expected to find an
answer to RQ2. In the following subsections, we will present the
individual hypotheses.
2.1. Behavioral intention to express empathy
2.1.1. The empathy expression concept
It would be worthwhile to first introduce the concept of empa-
thy expression. SNS users who have been exposed to an ad generally
go through an evaluative phase where a user develops an affective
evaluation response (for example, good, bad, angry, sad, etc). And
most SNS including Facebook and Twitter provide an affective eval-
uation mechanism, known as ‘one-click social plugins’, whereby the
user can express his or her affective response in the form of positive
evaluation (e.g., like) or negative evaluation (e.g., dislike). One-click
social plugins (e.g. Like, Google +1, Retweet, and Share) are buttons
placed on social media or other sites through which users can share
their interest or convey their attitude about various content (Swani,
Milne, & Brown, 2013). In case of Facebook, the affective response
can be expressed only as positive evaluation (i.e., like), and thus, no
expression will be regarded as either negative or neutral evaluation
of the given ad. The positive affective evaluation is, in nature, virtual
empathy that a user has about the ad message (Rosen, 2012). Empa-
thy is generated from the perspectives of a user when he or she
merges with others’ feelings involuntarily (Escalas & Stern, 2003).
Psychologists conceptualize empathy as “an emotional response
that stems from another’s emotional state or condition and that is
congruent with the other’s emotional state or situation” (Escalas &
Stern, 2003; p. 567). Empathy in a social network context captures
empathic reactions that a user develops to a SNS ad. Hence, for
our research, empathy is defined as the user’s ability to sense and
understand the emotions, moods, and temperaments of others with
regards to an ad that has been exposed on a social network, coupled
with the ability to imagine what others might be thinking or feel-
ing about that ad. Once the empathy is experienced on an online
social network, it is expressed via a social plug-in. A Facebook Like
is an example of empathy expression; by clicking the Like button,
users show that they understand and share in the emotional state
or context of the user who posted the message (Rosen, 2012). Users
presented with an appealing advertisement on a SNS are likely to
be driven by a social need to share in the emotional state of the
users who have already clicked the Like button. Thus, a Facebook
user who has positive affective evaluation of a SNS ad may choose
to explicitly reveal his or her own empathy by clicking on Like.
2.1.2. Attitude toward empathy expression vs. attitude toward
advertising
In this research we took a unique approach by adopting as a
construct the attitude toward empathy expression associated with
an ad, rather than the attitude toward advertising that is com-
monly examined by most related studies. The advertising literature
defines attitude toward advertising (AAD) as “a predisposition to
respond in a favorable or unfavorable manner to a particular adver-
tising stimulus during a particular exposure occasion” (Lutz, 1985).
So the AAD concept specifically addresses the evaluations of a given
ad.
On the other hand, attitude toward empathy expression (AEE) is
conceptually rooted on AAD, but the concept is tied to a SNS context.
On a SNS, a user exposed to an ad will first evaluate the given ad (e.g.,
“It is an interesting ad”), and then evaluate the empathy expression
behavior with regards to the ad (e.g., “It is worth clicking the Like or
Share button”). The SNS user’s viral behavior will be directly contin-
gent upon the nature of the ad. That is, determinants of evaluative
perceptions of his empathy expression behavior are largely deter-
minants of evaluative perceptions of the ad. For example, if a user
perceives that an ad exposed to him is out of ordinary and heart-
touching, then no doubt he is likely to click “Like” or “Share” on
a social network so that his acquaintances will be informed of his
view of the ad.
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
It incorporates a SNS user’s attitude toward the viral behavior
(i.e., the behavior of overtly indicating and spreading his or her
personal reaction to a given ad). As Alhabash and McAlister (2014)
note, online content on a social network like Facebook or Twitter
becomes viral as it is communicated and disseminated in a digital
form creating the WOM effect. Thus, for our research we define AEE
as “a user’s evaluations of the behavior of disclosing and spread-
ing his or her positive reactions to a SNS ad.” In this regard, AEE
can influence the effectiveness of an advertising campaign, since
the WOM advertising success is dependent on users’ AEE and viral
behavior.
As we can see in Table 2, this construct measures a user’s over-
all attitude about both the SNS advertisement and the behavior of
empathy expression with regards to that ad. Since one-click social
plugin (e.g., the Like button) available on a SNS encourages more
frequent WOM behavior (Swani et al., 2013), SNS users would be
generally inclined to express their affective evaluation of the SNS
ad to their social network acquaintances, contributing to the brand
awareness. Thus, the attitude toward empathy expression would
be a more appropriate construct to examine for our research than
the attitude toward advertising.
2.1.3. How attitude relates to intention about empathy expression
There exists little research delving into the relationship between
attitude and intention in the context of empathy expression. The
TRA theory postulates that the intention to perform a behavior
is determined by the attitude to performing it as well as subjec-
tive norm (Fishbein & Ajzen, 1975). Attitude refers to the degree
to which a person has a favorable or unfavorable evaluation or
appraisal of the behavior in question (Ajzen, 1991). Behavioral
intention is the cognitive representation of a person’s readiness to
perform the behavior.
With our research, the behavior under consideration is a user’s
expression of empathy about an SNS ad. Hence, intention to express
empathy is defined as a user’s cognitive representation of his readi-
ness to express empathy about a given SNS ad. Provided that a user
has a positive disposition about the behavior of empathy expres-
sion, he or she is likely to develop a strong intention toward that
behavior. In their study on Facebook users’ motivation for the
Like-clicking behavior, Chin and Lu (2015) found that there was
a significant positive relationship between attitude and intention
under consideration. The formation of the user’s positive attitude
may reply on his or her values or beliefs that a solid view or opin-
ion that he or she comes up with about a matter of concern should
be communicated to or exchanged with others. If he strongly feels
that it is desirable to show support for an interesting ad by clicking
Like, then with no doubt he is likely to have an inclination to click
it without hesitation so that his reaction may be shared with his
acquaintances. Likewise, it can be posited that an individual expe-
riencing intense empathy will have a favorable attitude toward the
expression of empathy and be highly inclined to express empa-
thy by clicking Like for the SNS ad too. Accordingly, the following
hypothesis is proposed:
H1. Attitude toward empathy expression is positively related to
the intention to express empathy.
Before examining how intention to express empathy is related
to purchase intention, it would be worthwhile to understand how
the number of likes is related to sales amount. The relationship
between Facebook Likes and sales is controversial. While Facebook-
based marketers claim that the more Likes a Facebook ad is given,
the greater sales the ad is likely to generate, some say that there are
many companies today offering to sell ‘cheap fans’ and therefore
the number of likes may not reflect the true purchase intention
of potential buyers (Rubin, 2012). Nevertheless, a single click on a
Facebook Like for a daily deal was associated with 4.5 additional
363
voucher sales, equivalent to an increase in revenue of $468 (Li &
Wu, 2013) and one click to Like is estimated to create at least $8 in
revenues (Chompon.com, 2011). Thus, a positive relationship can
be established between the number of Likes and sales volume.
Then would intention to click Like as an empathic reaction
to a SNS ad lead to intention to buy? In their empirical study
on a company-hosted Facebook community, Pöyry, Parvinen, and
Malmivaara (2013) found that both browsing and contribution by
community members were positively related to intentions to pur-
chase a community product. Their finding implies that active user
behaviors on Facebook ad pages can translate into purchase inten-
tion. Liking captures the user’s affective reaction to a page (Gerlitz
& Helmond, 2011). When a user is inclined to click the Like for an
ad after experiencing a positive empathic reaction to the ad, the
user’s Like is likely to represent personal support for the ideas or
theme embedded in the ad. Thus, the user will now have an ele-
vated interest in the product or service featured in the SNS ad. And
when the time comes to make a purchase, the user is more likely to
choose that brand than others. Therefore, the intention to purchase
a product or service featured in an SNS ad may be a consequence
of positive user responses on a social network (e.g., intention to
engage in positive affective evaluation of an SNS ad). Almost 25% of
consumers participating in one survey replied that they had inten-
tion to purchase the brand that their friends liked (eMarketer.com,
2012). If a user learns that his or her friends liked a SNS ad, he or
she will be motivated to click the Like button, thereby increasing
the user’s intention to purchase that brand. Thus, persuading con-
sumers to like would be a useful strategy for marketers who try
to augment advertising effectiveness. Based on these theoretical
grounds, we propose the following hypothesis:
H2. Intention to express empathy is positively related to purchase
intention.
2.2. Advertising effectiveness
Research on advertising effectiveness highlights advertising
creativity (Reinartz & Saffert, 2013), informativeness and enter-
tainment (Ducoffe, 1996; Gao & Koufaris, 2006), irritation (Logan
et al., 2012), and social cues (Liu, 2014) as antecedents of con-
sumer attitudes to an advertisement. A review of the related
literature indicates that informativeness and entertainment as well
as advertising creativity are recognized as key sources of adver-
tising effectiveness (Ducoffe, 1996; Taylor et al., 2011). While
these related studies are designed to determine antecedents of
attitude toward advertising, our research focuses on identifying
antecedents of attitude toward empathy expression. As we men-
tioned earlier, attitude toward empathy expression is a construct
newly created specifically for the SNS setting. Since the construct
is based on the attitude toward advertising concept, it would
make logical sense to choose among the antecedents of attitude
toward advertising. At the same time, it would be necessary to take
into considerations potential determinants of a SNS user’s attitude
toward the viral behavior (i.e., empathy expression) with regards
to the ad.
We argue that both informativeness and advertising creativity
often emphasized in the advertising literature importantly affect
the user’s behavior in a SNS environment. SNS users are more
interested in advertising that has informative value and creative
content (Reinartz & Saffert, 2013; Taylor et al., 2011). On the other
hand, while the existing studies emphasize entertainment as a key
determinant of attitude toward advertising, users on a SNS react
more favorably to a persuasive message or other online content
if it has greater emotional appeal (Alhabash, McAlister, Quillam,
Rifon, & Richards, 2013). Thus, the present research argues that
SNS advertising effectiveness should be evaluated via emotional
364 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
Table 1
Demographic profile of the sample.
Attribute Value
Gender Female
Male
Age 20–23
24–26
27–30
Facebook membership in years Less than 6 months
6 months to one year
One to two years
More than two years
Facebook usage frequency Daily
Weekly
Less often
Average duration of Facebook use Less than 30 min
30 min to one hour
More than one hour
appeal instead of entertainment; emotional appeal is an integra-
tive concept inclusive of entertainment, as it encompasses a wide
range of emotions including fun. Therefore, we determine that the
constructs relevant to our research questions are emotional appeal,
informativeness, and advertising creativity, which are believed to
play key roles in the formation of behavioral responses to SNS
advertising (please see Section 3.2 for the procedure used to derive
these dimensions).
2.2.1. Emotional appeal and attitude
Emotional appeal is a method of persuasion designed to create
an emotional response to a message by using emotional content
(e.g., horror movie, sad story, triumphant music, etc.). As the ELM
predicts, emotional appeals will more effectively persuade an indi-
vidual who has little motivation or little ability to cognitively
process a message (Petty & Cacioppo, 1986). Emotional appeal deals
with the ways in which emotional responses are triggered as a
function of the message’s motivational relevance to individuals
(Alhabash et al., 2013). Emotion is a person’s state of feeling in
the sense of an affect. Emotion can be either positive (e.g., love,
joy) or negative (e.g., fear, anger, sadness). Related studies in the
field of communication recognize emotion as a vital element that
can importantly affect the effectiveness of a message (Xie, Donthu,
Lohtia, & Osmonbekov, 2004). In their empirical study on social
network WOM, Swani et al. (2013) found that emotional content
in an advertising content generates more WOM; for example, their
study revealed that a message containing emotional content gener-
ated 16 estimated mean Facebook Likes, whereas 8 estimated mean
Likes were observed without emotional content. When a persuasive
message is received, affective responses (i.e., feelings and moods
evoked by the ad) are combined with cognitive responses (i.e., ratio-
nal evaluation of the ad) to form the attitude toward the message
(Batra & Ray, 1986). Messages with strong emotional appeals have
been found to be more likely to be shared online (Alhabash et al.,
2013). Thus, the related studies suggest that the stronger emotional
appeals a SNS ad possesses, the more positive attitude a user is
likely to develop toward the ad.
H3. Emotional appeal is positively related to attitude toward
empathy expression.
2.2.2. Informativeness and attitude
Informativeness is defined as the “ability to inform users about
product alternatives that enable them to make choices yielding the
highest value” (Rotzoll, Haefner, & Sandage, 1990). Informative-
ness is a perceptual construct that is measured via self-reported
Frequency Percentage (%)
151 36.4
264 63.6
185 44.6
184 44.3
46 11.1
41 9.9
14 3.4
86 20.7
274 66.0
237 57.1
109 26.3
69 16.6
308 74.2
74 17.8
33 8.0
items (Pavlou, Liang, & Xue, 2007). And it encompasses rational
appeal due to its ability to help a consumer make an informed
judgment about the acceptance of the message, and therefore is
conceptually distinct from ‘emotional appeal.’ In their early study
on information content in television advertising, Resnik and Bruce
(1977) stated that an informative ad enables a consumer to make a
more intelligent buying decision. Consumers will find an advertise-
ment valuable, if it gives accurate portrayals of products (Andrews,
1989; Taylor et al., 2011). Studies have found informativeness to
be important in the formation of consumer attitudes to electronic
commerce websites (Gao & Koufaris, 2006) as well as to television
advertising (Resnik & Bruce, 1977) and to SNS advertising (Taylor
et al., 2011). Informative advertising messages on a SNS will draw
users’ attention and motivate them to associate the ad with a pos-
itive image and at times to drive them to spread the messages to
friends over the social network volitionally. Overall, the related lit-
erature agrees that informative SNS ads provide a user with the
ability to make an informed judgment for a future purchase, and
that the user’s perception of this favorable aspect of the ad leads
to the formation of a positive attitude toward the viral behavior as
well as toward the ad. Based on the above theoretical grounds, we
propose the following hypothesis.
H4. Informativeness is positively related to attitude toward empa-
thy expression.
2.2.3. Advertising creativity and attitude
Advertising creativity is the extent to which an ad is original
and unexpected (Haberland & Dacin, 1992). Divergence and rele-
vance are known to be the leading characteristics of advertising
creativity; divergence is associated with elements that are novel,
different or unusual, whereas relevance is concerned with elements
that are meaningful, appropriate, useful, or valuable to the audi-
ence (Smith, MacKenzie, Yang, Buchholz, & Darley, 2007). Creative
messages grab more attention and lead to positive attitudes about
the featured products (Reinartz & Saffert, 2013). Furthermore, a
“euro invested in a highly creative ad campaign had, on average,
nearly double the sales impact of a euro spent on a noncreative
campaign” (Reinartz & Saffert, 2013, p. 5). In their experimental
study, Smith et al. (2007) found that creative ads increase motiva-
tion to process the ad and improve the attitude toward the ad and
that the positive affect transfers to the brand. Thus, advertising cre-
ativity is strongly linked to the effectiveness of an ad. On the above
theoretical grounds, we propose the following hypothesis:
H5. Advertising creativity is positively related to attitude toward
empathy expression.
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
2.3. Social influences
Research on social influence suggests that there are two psycho-
logical needs that lead humans to conform to the expectations of
others: the need to be right (informational social influence) and the
need to be liked (normative social influence) (Deutsch & Gerard,
1955). Informational social influence is “the influence to accept
information obtained from another as evidence about the true state
of some aspect of the individual’s environment,” whereas norma-
tive social influence is “the influence to conform to the expectations
of another person or group” (Burnkrant & Cousineau, 1975; Deutsch
& Gerard, 1955).
Social influence can affect an individual’s attitudes and behav-
iors. Kelman (1958) states that changes of attitudes produced
by social influence may occur through three different processes
including internalization, compliance, and identification. Internal-
ization is the adoption of induced behavior because it is congruent
with his value system and can help attain his goals. Compliance
is a change in behavior but not necessarily attitude (e.g., obedi-
ent behavior). Finally, identification is the change of attitudes or
behaviors due to the influence of someone respected or liked (e.g.,
a celebrity) (Kelman, 1958). In the buying context, these three pro-
cesses are associated with the Deutsch and Gerard’s social influence
types. Internalization would occur if a consumer determines that
information from others (e.g., product reviews) helps achieve the
cost reduction goal; hence the internationalization process will
be involved when one accepts informational social influence. On
the other hand, compliance and identification are applicable to the
normative social influence situation. Compliance is likely to occur
when an individual decides to behave according to the expecta-
tions of his acquaintances (i.e., his subjective norms). For example,
a consumer may decide to adopt a mobile instant messaging system
because his online acquaintances strongly expect him to. Mean-
while, identification would be involved when a consumer decides
to purchase a bicycle featured in an ad depicting a renowned sports
hero using the product. This consumer’s behavioral change can be
described as a consequence of identifying himself with the sports
hero appearing in that ad.
We shall show below how these two forms of social influence,
informational and normative, are related to each other as well as
to behavioral intention to express empathy on a social network.
2.3.1. Perceived herd behavior and intention to express empathy
Informational social influence refers to “the pressure to accept
information from another as evidence of reality” (Deutsch & Gerard,
1955) and is commonly known as “informational cascades” or “herd
behavior.” Informational influence operates through the process of
internalization in which an individual takes information from oth-
ers if it increases his or her knowledge needed to reduce uncertainty
(Bearden, Netemeyer, & Teel, 1989; Kelman, 1958). The consumer
behavior literature suggests that product evaluation is a typical
setting in which informational social influence is likely to operate
(Cohen & Golden, 1972). A product presentation would be internal-
ized only if it is perceived as providing useful information about the
product (Burnkrant & Cousineau, 1975). Thus, under uncertainty,
consumers observe the actions of others and then follow their
behavior without regard to their own information. In this research,
we define perceived herd behavior as the perceived social pressure
to conform to the behaviors of others as a result of exposure to their
online behaviors.
Users exposed to a social network ad who find that many peo-
ple have liked it are likely to click the Like button just to “hop on
the bandwagon.” The social psychology literature defines the band-
wagon effect as the social phenomenon whereby the rate of uptake
of beliefs, ideas, fads and trends increases the more that they have
already been adopted by others (Colman, 2014). When a SNS user
365
observes that the behavior of expressing empathy about a message
presented on a SNS like Facebook has become pervasive, he or she
will be inclined to be part of this fad. In particular, the numbers
of Likes and Shares are signals to viewers about the quality of the
message and whether it is worth viewing (Alhabash et al., 2013).
Thus, the stronger social pressure the user perceives to accept the
widespread Like-clicking behavior of others on a social network,
the more intense the intention to express empathy. Based on the
above logical grounds, we propose the following hypothesis.
H6. Perceived herd behavior is positively related to intention to
express empathy.
2.3.2. Subjective norm and intention to express empathy
Meanwhile, normative influence can affect an individual’s atti-
tudes, norms, and values (Burnkrant & Cousineau, 1975). According
to Bearden et al. (1989), normative social influence is divided into
value expressive and utilitarian influences, depending on which
of Kelman’s processes of attitude change it is related to. Value
expressive influence is motivated by a desire to enhance self-image
by association with a referent like a sports hero; hence, it oper-
ates through referent identification (Bearden et al., 1989; Kelman,
1958). On the other hand, utilitarian influence is manifested in
an individual’s need to comply with the expectations of others to
achieve rewards or avoid punishments (Burnkrant & Cousineau,
1975; Kelman, 1958).
Normative influence has been frequently studied as the “sub-
jective norm” within the general framework of the TRA (Fishbein
& Ajzen, 1975). Subjective norm is the perception or opinion about
what important others believe an individual should do (Ajzen &
Fishbein, 1980). Therefore, from the perspective of Kelman’s the-
oretical framework, subjective norm would likely accompany the
compliance process, and it reflects utilitarian influence.
Related studies on technology acceptance have found that sub-
jective norm plays an important role when a user decides on the
acceptance of a new information system; for example, it positively
affects the behavioral intention to adopt B2C e-commerce (Crespo
& Bosque, 2008) or instant messaging (Lu, Zhou, & Wang, 2009).
Likewise, subjective norm is expected to play an even larger role in
determining the intention to use a SNS, another form of informa-
tion system, since a user’s online usage behavior including posting
a comment or clicking the “Like” or “Share” button tends to be quite
transparent to other users on the social network. In particular, the
reward that motivates the user to comply with the expectations of
these acquaintances is often the maintenance of a favorable rela-
tionship with a referent group. Hence, a user who learns that many
acquaintances have already clicked the Like button for a given Face-
book ad will perceive intense pressure to click “Like” for the ad as
well, and comply with the expectation to become part of the ref-
erent group who clicked “Like” for the ad. Based on this line of
reasoning, we propose the following hypothesis:
H7. Subjective norm concerning empathy expression is positively
related to intention to express empathy.
2.3.3. Perceived herd behavior and subjective norm
Cohen and Golden (1972) argued that both informational and
normative influences in a product evaluation situation are present
under the following two conditions: (a) quality is somewhat
ambiguous because of a lack of clear standards, and (b) one’s own
ability to discriminate is not thought satisfactory. Thus, perceived
herd behavior and subjective norm can both play a role in shaping
a user’s online behavior.
Perceived herd behavior conceptualized as informational influ-
ence operates through internalization where the chief motivation
is to obtain knowledge from a credible source and reduce uncer-
tainty. A lack of information, an ambiguous situation, or premature
366 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
demands for action or decision lead the person to substitute seem-
ingly competent information from others for his own search for
direct evidence (Cohen & Golden, 1972). When a user is first
exposed to a SNS ad, he or she would be unsure as to what is
the most appropriate behavioral response to the ad on a social
network. To resolve the uncertainty, the user would observe the
statistics—e.g., how many Likes or Shares there are. If the user found
high figures, the user would likely perceive strong herd behavior
and no doubt click Like or Share to imitate the behavior of the
referent group.
Whereas perceived herd behavior is based on a basic, low-level
motivation (i.e., reducing uncertainty and attain an individual’s
goal), subjective norm is associated with an advanced, high-level
motivation (that is, maintaining a favorable relationship with a ref-
erent person). One will likely choose to accept influence either to
establish or enhance a favorable reward-punishment relationship
with a certain individual or because of a desire to identify with such
an individual (Cohen & Golden, 1972; Kelman, 1958). If a user per-
ceives that herd behavior occurs intensely, the user is quite likely
to comply with expectations of his online friends. In their study on
virtual communities, Bagozzi and Dholakia (2002) found that the
attitude and behavior of a community member were apparently
influenced by referent groups within the community. SNS mes-
sages are disseminated among a larger number of “friends” on a
social network much faster than on an offline network. Thus, the
WOM process will be facilitated.
In sum, the basic need to jump on the bandwagon to reduce
uncertainty and achieve a user’s personal goal is likely to trigger
the advanced need to comply with the norms established by the
referent group. Thus, a user who has accepted influence of herd
behavior associated with expression of empathy will likely accept
influence of subjective norm as well. Based on the above grounds,
we propose the following hypothesis.
H8. Perceived herd behavior is positively related to subjective
norm concerning empathy expression.
Issues of privacy on a SNS are becoming an increasing con-
cern. Some studies have reported that concern about privacy was
the main reason why some SNS users closed their SNS accounts
(Stieger, Burger, Bohn, & Voracek, 2013). Privacy concern was found
to have a negative impact on self-disclosure, meaning that the
higher the users’ concern for privacy, the less information the user
will disclose on SNSs (Zlatolas, Welzer, Hericko, & Hölbl, 2015).
Furthermore, SNS users’ private information accumulated on the
server in the course of system use (including conversations, pro-
files, affective evaluations, and so forth) are illegally collected and
used for unintended purposes such as surveillance or monitoring.
For example, some employers use SNS to gather negative informa-
tion about job candidates that may be subsequently used to reject
a candidate (Bowers, 2008). Our social media “audience” is bigger
than we actually know; our followers or friends are not the only
ones that can see information about us (Wikipedia.org, 2015a). For
that reason, those concerned about the possibility that others might
learn about their ad likes and infer their character, values or prefer-
ences from them are likely to refrain from clicking the Like button.
Furthermore, existing research has found that privacy concerns are
negatively related to attitude to SNS advertising (Taylor et al., 2011).
Therefore, we hypothesize the following:
H9. Privacy concern is negatively related to intention to express
empathy.
3. Research method
3.1. Data collection and sample
An online survey was conducted using the Google Forms tool
to collect data for empirical analysis. Our research employs a self-
reported data collection method based on experimental design. An
experimental design is designed to collect data by applying a treat-
ment (or a stimulus) and observing the response. In our research,
a Facebook ad page has been presented to the experiment partici-
pants as a treatment, and a questionnaire has been used to capture
the perceptions of our participants after they had been exposed to
the Facebook ad page.
A total of 420 undergraduate students at a major university in
Korea, took part in this survey for course credit. The survey lasted
for the period of one week, and an additional week was given to
induce the non-respondents to participate. College students were
deemed appropriate as participants because Facebook has become
popular by solely catering to college students. The percentage of
U.S. college-aged adults (i.e., aged 18–24) who use Facebook con-
stituted 92.4% in October 2014 (McDermott, 2014). Besides, a recent
survey found that Facebook is the most popular SNS in Korea and
that young adults (i.e., aged 18–34) represented 65% of all Korean
Internet users in 2011 (Socialbakers, 2011). In this study, partici-
pants ranged from 20 to 30 years old (M = 23.86, SD = 2.03). Males
comprised 63.6% of the respondents. The relatively high proportion
of males is attributed to Korea’s corporate culture where business
firms still hire more males than females. The Korean firms’ employ-
ment practice seems to have affected the gender composition of
college students majoring in business administration. Five invalid
responses were removed, leaving 415 responses for analysis. The
respondents’ profile is shown in Table 1.
The stimulus used for our experiment was a captured still image
of a recent Facebook ad posted by a carpool service company named
Tikle. Tikle is a Korean sharing-economy company founded in 2012
to concentrate on carpool brokering services based on internet. The
participants were provided via e-mail with a link to a web-based
page including this image and the online questionnaire. On this web
page, they were instructed to answer the online questions based
on the still image. The image included a message urging drivers to
share vacant automobile seats and save gas expenses. The advertis-
ing copy reads “Try our carpool service for yourself. You can share
vacant seats and save money. The number one carpool company,
Tikle” (see Appendix A).
Unlike their Western counterparts, a majority of Korean college
students tend to use public transportation like subway trains, buses
or taxis for their routine trips, and hence rarely own an automobile.
For that reason, they are increasingly interested in emerging car-
pool services that not only provide comfortable rides for affordable
costs, but also help save fuel expenses. Thus, we determined that it
was a suitable SNS ad for our study. In addition, this carpool ad has
been chosen from a pool of existing ads on Korean Facebook web-
site (http://ko-kr.facebook.com), based on its relatively high figures
of “Likes” and “Shares”.
The stimulus chosen for this study is measurable in each of the
dimensions of advertising effectiveness. First, the stimulus has a
potential to generate a sense of guilt (i.e., a type of emotion) on the
part of some respondents, while individuals may experience dif-
ferent emotions depending on their values and background. Many
Korean cities have a severe traffic congestion issue while fuel prices
are unusually high. The ad image portrays multiple drivers who
each have no passenger in a car. The situation presented by the
image is against the social norm that people should help save the
limited energy resource. A respondent viewing the Facebook ad
would likely have guilty feelings if he or she were in the driver’s
place, since these solo occupants are not taking into considera-
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
Table 2
Measures for the research model.
Construct Definition
Emotional appeal The extent to which emotional
responses are triggered when exposed
to a SNS ad
Informativeness The degree to which a SNS ad offers
information users perceive as useful
Advertising creativity The extent to which an SNS ad is
original and unexpected
Privacy concerns Perceived negative consequences that
may result from disclosing personal
information to others on a SNS
Perceived herd behavior Perceived social pressure to conform to
the behaviors of others as a
consequence of exposure to their
online behaviors on a SNS
Subjective norm Perceived social pressure to comply
with a behavior expected of a user by
important referents
Attitude toward empathy expression Predisposition to respond in a
favorable or unfavorable manner about
clicking Like for a SNS ad
Intention to express empathy Perceived likelihood that a user will
engage in the behavior of clicking Like
for a SNS ad
Purchase intention Perceived likelihood that a user will
buy a given service in the near future
tions economic or environmental impact of their behavior. Based
on the ELM theory, it may be inferred that users viewing this ad
will take a peripheral, rather than central, route to processing the
message such that the sense of guilt aroused by the ad will leave
an intense impression about the ad. Second, the present stimulus is
367
Measure items Sources
—After seeing this ad, I had intense Davis (1983);
feelings Escalas and Stern
—I was emotionally attracted by the (2003)
key message of this ad
—The emotional aspect of this ad leads
me to like the ad
—Information obtained from the SNS Logan et al. (2012);
ad would be useful Pavlou et al. (2007)
—I would learn a lot from using the
SNS ad
—I think the information obtained
from the SNS ad would be helpful
—The SNS ad is unique Haberland and Dacin
—The SNS ad is really out of ordinary (1992)
—The SNS ad is intriguing
—The SNS ad is surprising
—I am concerned that my online usage Buchanan et al. (2007)
behavior might be known to others
while I use the Internet
—I am concerned that my Facebook
usage details might be known to other
Facebook users
—I am concerned that my
acquaintances might learn about my
liking the SNS ad
—My decision to click “Like” for this ad Huang and Chen
is likely to be influenced by the (2006);
number of “Likes” for this ad Yoo et al. (2008)
—If I find that many of my Facebook
acquaintances “Liked” this ad, then I
would be more willing to click “Like”
for this ad
—The more “Likes” this ad attracts, the
more preferable it is to click “Like” for
this ad
—It is wise to adopt the choice of other
Facebook users when deciding
whether to click the “Like” for this ad
—My Facebook acquaintances would Ajzen and Fishbein
find it reasonable for me to click “Like” (1980)
for this ad
—My Facebook acquaintances would
find it necessary for me to click “Like”
for this ad
—My Facebook acquaintances would
find it worthwhile for me to click
“Like” for this ad
—My Facebook acquaintances would
expect me to click “Like” for this ad.
—I feel good about clicking “Like” for Holbrook and Batra
this ad (1987)
—I feel positive about clicking “Like”
for this ad
—I feel favorable about clicking “Like”
for this sort of ad
—My overall attitude toward this ad is
positive
—I find it wise to click “Like” for this ad Alhabash et al. (2013);
—I am inclined to click “Like” for this ad Eckler and Bolls (2011)
—I might purchase this service MacKenzie, Lutz, &
—It is worth purchasing this service Belch (1986)
informative enough to provide useful information that would help
a consumer make a right purchase decision. For example, it informs
that the use of a carpool service can help save expensive fuel costs
through a more efficient utilization of empty seats. Furthermore, it
shows that carpool services can be used not only for daily commut-
368 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
ing but also for special occasions such as weekend leisure trips, ski
trips, etc. It also tells that Tikle is the leading player in the domestic
market that attracts a large number of drivers who look for rides to
lower their fuel expenses. Finally, the advertising stimulus chosen
for this study is creative to some extent. It shows two contrasting
situations in one still image: one where the road is packed with
cars and the other where the road is occupied by the same num-
ber of drivers as the cars. Such an unusual image will likely bring
to the viewer senses of uniqueness, unexpectedness and extraor-
dinariness. In summary, because of the emotional, informative and
creative aspects embedded in the ad image, this ad is likely to suc-
ceed in capturing some attention of those whom the ad is exposed
to.
3.2. Measures
First, interviews were conducted using a focus group to gain
insight into how users like Facebook advertising. A total of 17
college students who use Facebook regularly were chosen and
assigned to the focus group. Questions were asked to learn what
aspects of a SNS ad lead them to give a positive user response, such
as clicking Like or Share on Facebook. Answers to the questions
included impressiveness, eye-catching appeal, relevance, extra-
ordinariness, heart-touching potential, among others. Then, based
on the focus group interview results and the related literature, key
factors likely to lead to the positive user response to a SNS ad were
identified as emotional appeal, informativeness, and advertising
creativity.
Measurement items were mainly adapted from the related prior
studies to fit the context of SNS advertising. Advertising effec-
tiveness is comprised of three dimensions. First, emotional appeal
refers to the extent to which emotional responses are triggered
when exposed to a SNS ad, and the items are based on Davis
(1983) and Escalas and Stern (2003). Davis (1983) proposed a multi-
dimensional measure of empathy consisting of Perspective Taking
(PT), Fantasy (FS), Empathic Concern (EC), and Personal Distress
(PD). Of these four scales, EC and PD deal with emotional reactions,
and EC is particularly relevant to our emotional appeal construct.
Our emotional appeal scale is also rooted on the Ad Response Empa-
thy (ARE) scale developed by Escalas and Stern (2003) to capture
the empathic reaction to a television commercial. Second, informa-
tiveness denotes the degree to which a SNS ad offers information
users perceive as useful, and the scale was adapted from Logan et al.
(2012) and Pavlou et al. (2007). Finally, advertising creativity refers
to the extent to which an SNS ad is original and unexpected, and
the scale was adapted from Haberland and Dacin (1992). The pri-
vacy concerns construct (Buchanan, Paine, Joinson, & Reips, 2007)
was measured using a scale customized to capture the aspects of
privacy associated with a user’s sensitive information (e.g., political
or ideological or social views) that may be revealed to others when
they click the Like or Share button. The items used to measure per-
ceived herd behavior were taken from previous research (Huang &
Chen, 2006; Yoo, Kim, Moon, & Choe, 2008) and modified for this
study. The scale for subjective norms was adapted from Ajzen and
Fishbein (1980), and was used to measure perceived social pres-
sure to comply with a behavior expected of a user by important
referents. The scale for attitude toward empathy expression was
modified from Holbrook and Batra (1987) to measure a predisposi-
tion to respond in a favorable or unfavorable manner about clicking
Like for a SNS ad. Intention to express empathy refers to perceived
likelihood that a user will engage in the behavior of clicking Like for
a SNS ad. The scale to measure intention to express empathy was
adapted from Alhabash et al. (2013) and Eckler and Bolls (2011).
All items were measured on a five-point Likert-type scale, ranging
from “strongly disagree” (1) to “strongly agree” (5). Table 2 lists the
survey items used to measure the constructs and their operational
definitions.
4. Results
4.1. Measurement model
We conducted a confirmatory factor analysis using AMOS 21.0
to test the measurement model. Goodness of fit was tested using
six common model-fit measures, and all the indices indicated a
good fit between the hypothesized model and the observed data
(x2 /df = 1.77, RMSEA = 0.043; NFI = 0.924; CFI = 0.965; TLI = 0.958;
RMR = 0.050, SRMR = 0.043). The overall fit for the measurement
model was acceptable according to conventional cutoff criteria
(Kline, 2005).
The measurement model was assessed for reliability and valid-
ity. The reliability of the constructs was assessed using Cronbach’s
alpha and composite reliability (CR) values. As shown in Table 3,
the Cronbach’s alphas for the constructs were between 0.756 and
0.909 and the CR coefficients were above 0.716, all exceeding the
0.70 threshold proposed by Fornell and Larcker (1981). Therefore,
the constructs used in this study represented acceptable reliability.
Next, convergent validity and discriminant validity were tested.
First, convergent validity was assessed by examining factor load-
ings of each observed variable. All factor loadings were significant
and exceeded the recommended 0.50 threshold (Hair, Black,
Babin, Anderson, & Tatham, 2006). Convergent validity was further
assessed by examining CR and average variances extracted (AVE)
values. CR values were above 0.70 and AVE values were above 0.50,
indicating that both CR and AVE values exceeded the corresponding
cutoff criteria (Hair et al., 2006). CR values of the constructs ranged
from 0.716 to 0.912 and AVE values ranged from 0.508 to 0.801. This
demonstrated fair convergent validity (see Table 3). Furthermore,
the square root of AVE obtained for each construct was larger than
was the correlation of the construct with each of the remaining
constructs (see Table 4), confirming discriminant validity (Fornell
& Larcker, 1981). Therefore, convergent and discriminant validity
as well as reliability proved to be adequate for the measurement
model.
The data used for this study was gathered based on the self-
reported perceptions of the respondents. There is potential for
common method bias for all self-report data. So before examining
the hypothesized model, common method bias was assessed. To
assess the severity of common method bias, Harman’s single-factor
test (Podsakoff & Organ, 1986) was performed. For the Harman’s
single-factor test, all the variables data in our model were entered
into an unrotated exploratory factor analysis. Results from this
test showed that the most covariance explained by one factor was
33.81% which was below the 50% threshold, meaning that common
method biases are not a severe concern for this study (Podsakoff &
Organ, 1986). Furthermore, an additional test was performed to
examine the common method bias in the data set using the tech-
nique of unmeasured common latent factor, as demonstrated in
Podsakoff, MacKenzie, Lee, and Podsakoff (2003) and Liang, Saraf,
Hu, and Xue (2007). We conducted this test using AMOS. It was
found that differences between standardized coefficients without
a common latent factor and standardized coefficients with a com-
mon latent factor were less than 0.2. Therefore, common method
bias is not a major concern in our data set.
Last, a multi-collinearity check was conducted. A review of the
statistics literature indicates that we can use variation inflation
factor (VIF) to determine if there exists an overlap between two
variables (If VIF exceeds 10, then the items in the two variables
have overlaps). We conducted the VIF analysis, and the resulting fig-
ures were 1.070 (the minimum) and 2.052 (the maximum), which
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373 369

Table 3
Statistics of Construct Items.

Factor loadings Cronbach’s ␣ Composite reliability Average variance extracted Mean (S.D.)

Emotional appeal 0.756 0.716 0.508 2.47

EMO1 0.727 (0.899)
EMO2 0.626
EMO3 0.777

Informativeness 0.816 0.821 0.609 2.97

INFO1 0.841 (0.855)
INFO2 0.663
INFO3 0.826

Creativity 0.903 0.878 0.686 2.52

CREA1 0.788 (0.967)
CREA2 0.785
CREA3 0.878
CREA4 0.859

Perceived herd behavior 0.889 0.865 0.650 2.51

HERD1 0.864 (1.038)
HERD2 0.895
HERD3 0.763
HERD4 0.686

Subjective norms 0.909 0.912 0.718 2.51

SN1 0.827 (0.882)
SN2 0.892
SN3 0.895
SN4 0.771

Attitude to empathy expression 0.841 0.870 0.557 3.25

ATT1 0.716 (0.728)
ATT2 0.637
ATT3 0.754
ATT4 0.862

Privacy concerns 0.841 0.801 0.664 2.54

PC1 0.778 (1.032)
PC2 0.959
PC3 0.685

Intention to express empathy 0.889 0.830 0.801 2.79

INT1 0.905 (1.217)
INT2 0.885

Purchase intention 0.802 0.751 0.670 2.58

PI1 0.798 (1.061)
PI2 0.839

X2 = 595.004 (p = 0.000, df = 337); RMSEA = 0.043; NFI = 0.924; CFI = 0.965; TLI = 0.958; RMR = 0.050; SRMR = 0.043.

Table 4
Correlations between constructs (AVE and squared correlations).

EMO INFO CREA HERD ATT SN PC INTL PI

EMO 0.713*
INFO 0.529 0.780*
CREA 0.510 0.459 0.828*
HERD 0.325 0.299 0.261 0.806*
ATT 0.343 0.414 0.390 0.219 0.746*
SN 0.467 0.527 0.473 0.318 0.507 0.847*
PC −0.023 −0.103 0.010 −0.150 0.068 −0.025 0.815*
INT 0.429 0.474 0.442 0.238 0.559 0.562 0.098 0.895*
PI 0.455 0.558 0.348 0.238 0.432 0.442 0.059 0.612 0.819*

EMO—emotional appeal; INFO—informativeness; CREA—ad creativity; HERD—herd behavior; ATT—attitude toward empathy expression; SN—subjective norm; PC—privacy
concerns; INT—intention to express empathy; PI—purchase intention.
*
The numbers in the diagonal row are square roots of the average variance extracted.

suggests that there is no overlap. That is, there is no issue with
multi-collinearity which refers to a situation when variables are
so highly correlated with each other that it is difficult to come
up with reliable estimates of their individual regression coeffi-
cients.
4.2. Structural paths and hypothesis tests
We constructed a structural equation model to test hypothe-
ses. The hypothesized causal paths were estimated using AMOS
21.0. The resulting indices indicated a good model fit (x2 /df = 1.78,
370 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
Fig. 2. Hypotheses testing. X2 = 625.553 (p = 0.000, df = 352); RMSEA = 0.043; NFI = 0.920; CFI = 0.963; TLI = 0.957; RMR = 0.071; SRMR = 0.0584; * p < 0.05; ** p < 0.01; *** p < 0.001.
RMSEA = 0.043; NFI = 0.920; CFI = 0.963; TLI = 0.957; RMR = 0.071;
SRMR = 0.058) when evaluated against cutoff criteria in the litera-
ture (Kline, 2005). Fig. 2 displays the standardized path coefficient,
path significance, and explained variance (R2 ) for each path. Seven
hypotheses were supported while two were rejected. The results
of the structural model illustrated the vital role of informative-
ness and advertising creativity in determining users’ attitudes
to empathy expression. Informativeness (ˇ = 0.31, p < 0.001) and
advertising creativity (ˇ = 0.21, p < 0.01) were both found to have a
significant positive effect on attitude to empathy expression. How-
ever, the effect of emotional appeal on attitude proved insignificant
(ˇ = 0.16, p = 0.11). Thus, H4 and H5 were supported while H3 was
not. Meanwhile, the results revealed that perceived herd behav-
ior had a significant positive effect on subjective norm (ˇ = 0.32,
p < 0.001). However, perceived herd behavior was not significantly
related to intention to express empathy (ˇ = 0.06, p = 0.17). Sub-
jective norm (ˇ = 0.40, p < 0.001) and privacy concern (ˇ = –0.10,
p < 0.05) were significantly associated with intention to express
empathy. Thus, H8 and H9 were supported, whereas H7 was not.
Attitude to empathy expression had a significant positive effect on
intention to express empathy, supporting H1 (ˇ = 0.45, p < 0.001).
Finally, a positive relationship was observed between intention to
express empathy and purchase intention (ˇ = 0.69, p < 0.001), sup-
porting H2. It was revealed that 54% of the variance in intention
to express empathy was explained by a combination of emo-
tional appeal, informativeness, advertising creativity, perceived
herd behavior, attitude, subjective norm, and privacy concerns.
Emotional appeal, informativenss, and advertising creativity were
found to explain 35% of the variance in attitude toward empathy
expression.
5. Discussion and implications
5.1. Discussion
This research sought to develop a theoretical model specifying
the antecedents of positive user responses to an advertisement on
a SNS and to test the proposed model using structural equation
modeling. Our findings empirically support the proposed model.
The results of this study reveal two significant predictors of
attitude toward empathy expression: informativeness and adver-
tising creativity. The role of informativeness can be understood
from the instrumental, or utilitarian, perspective. Muk and Chung
(2014) found that utilitarian values of advertising on social net-
works enhance users’ positive attitudes toward brand pages. In our
research, information about the utilitarian aspects of a product or
service can strongly support consumers’ functional needs (Khan,
Dhar, & Wertebroch, 2005). The quality information presented in
SNS ads, such as a product’s benefits, can therefore help consumers
make optimal purchases. Meanwhile, the importance of advertising
creativity is associated with the fundamental human need to con-
sume something new, original, out of the ordinary, and imaginative
(Haberland & Dacin, 1992). A creative SNS ad will capture con-
sumers’ attention, driving them to express their positive affective
evaluation of the ad. Specifically, we found that the more infor-
mative and creative users perceive an SNS advertising message to
be, the more positive their attitude to empathy expression is. Thus,
users are likely to develop a positive attitude to the expression of
favorable affective evaluation of a SNS ad when the ad provides
information that addresses their functional needs (e.g., making an
optimal choice of product/service) and when it is creative enough
to provide hedonic value.
Contrary to our expectations, emotional appeal failed to influ-
ence user attitudes toward empathy expression. In forming a
positive user attitude toward empathy expression for a SNS ad,
emotional appeal is less likely to play a substantial role. That is,
emotional appeal may be important in itself, however it is not
expected to be influential enough to generate a positive attitude
toward empathy expression. Another possible explanation for this
insignificant relationship is that the SNS advertisement chosen
for this study (i.e., the carpool service ad) features a utilitarian
(i.e., functionally driven) service, and therefore, emotional appeal
appears to be less important, hardly giving rise to empathy in aver-
age users. Instead, rational aspects (e.g., informativeness) are likely
to play a larger role in forming user attitude for utilitarian services
than emotional aspects. This suggests that the product category
may moderate the relationship between emotional appeal and user
attitude.
The perceived herd behavior construct was designed to exam-
ine perceptions of informational social pressure. It was found that,
while perceived herd behavior has an indirect impact on inten-
tion to express empathy, it is positively related to subjective norm,
which is in turn positively related to intention to express empathy.
This finding implies that users who recognize that many people
appear to like a given SNS ad will be likely to perceive that impor-
tant others expect them to like the ad as well, ultimately leading
them to intend to like the ad. Theoretically, a user who decided to
reduce uncertainty associated with SNS usage by following the herd
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
behavior (i.e., adopting the prevailing viral behavior on Facebook)
is likely to move on to maintain favorable relationships with his
Facebook friends by complying with their expectations concerning
the behavior of clicking “Like” or “Share” as a reaction to a SNS ad.
Thus, top priority is given to uncertainty reduction and the next to
social relationship management. Furthermore, as the TRA predicts,
subjective norm positively affects intention to like the ad, imply-
ing that one may intend to like an SNS ad, not just because many
people have liked it but because one’s acquaintances expect one to
like it.
As expected, privacy concerns negatively influenced a user’s
viral intention for a SNS ad. When a SNS user strongly perceives
that his privacy might be severely violated in the event that his
friends make inferences about his ideology, character or prefer-
ences from the user’s viral behavior, the user is less likely to intend
to express empathy. Meanwhile, this finding appears to be some-
what compatible with the finding of another study (Chen & Kim,
2013) that examined the effect of privacy concerns on SNS usage.
Chen and Kim (2013) found that privacy concerns is not likely to
entirely thwart a user’s intense desire to use an SNS, but to limit
the user’s online usage to the extent that it may misrepresent or
expose his private matters to other users on a social network. The
respondents in our study may have perceived that their act of click-
ing Like for a SNS ad could be observed by their Facebook friends
and would give them the impression that they supported for-profit
commercials. If such a perception grows strong, an individual will
most likely avoid any overt behavior that is believed to invade his
or her privacy.
This research also found that the more positive a participant’s
attitude toward empathy expression and the more strongly partic-
ipants perceive the social pressure to express empathy, the greater
their intention to express empathy for the given SNS ad. Moreover,
SNS responses such as intention to express empathy were found to
influence purchase intention.
5.2. Implications
Our findings provide useful insights into what contributes to
the formation of user attitudes to empathy expression and behav-
ioral intention to express empathy with regards to a SNS ad. As we
shall see below, they offer theoretical and practical implications for
social media advertising.
The first theoretical implication is that it enhances our under-
standing about advertising effectiveness in a SNS context by
proposing a new construct called attitude toward empathy expres-
sion (AEE), a concept extended from attitude toward advertising
(AAD) traditionally studied in the marketing domain. We argue
that AEE may have better predictive power in explaining the viral
behavioral intention characterized by the SNS environment. Even
though we did not empirically analyze the relationship between
the two constructs (i.e., AAD and AEE), the findings suggest that
AEE appears to be a suitable construct in a SNS advertising context.
Second, the results demonstrate that the TRA and social influence
theory are robust theories that can help researchers predict the fac-
tors influencing users’ positive viral responses (e.g., clicking Like
or Share) and a subsequent marketing response (e.g., purchase
intention). These two theories are complementary in that they
together provide a comprehensive perspective on user behavior
on a SNS by enabling us to consider both personal and social fac-
tors. Third, while the relationship between perceived herd behavior
and intention to express empathy expression proved insignificant,
the relationship was mediated by subjective norms. Our results
helped enhance our understanding of the theoretical relationships
among these two forms of social influences and intention to express
empathy.
371
On the other hand, our research results also point to practical
implications. First, the finding on the relationship between inten-
tion to express empathy and purchase intention provides a key
implication for advertisers and marketers; Positive SNS responses
(e.g., a user’s viral intention and behavior) can help generate a
direct marketing response (e.g., purchase intention), thereby help-
ing to augment the sales. Users will be inclined to purchase a good
or service featured in an SNS ad when they have a strong inten-
tion to engage in a viral behavior for a given ad, suggesting that
a well-formulated advertising strategy focused on increasing the
number of likes and shares will eventually pay off. Second, our
results determined three antecedents of behavioral intention to
express empathy about a SNS ad; namely, attitude toward empathy
expression, subjective norm, and privacy concerns. In particular,
informativeness and advertising creativity were found to be sig-
nificant predictors of attitude toward empathy expression while
emotional appeal was not. Therefore, to increase the likelihood
that users will react favorably to an ad, advertisers and marketers
should make their advertising messages more informative and cre-
ative to the extent that such messages will effectively induce users
to express their empathy. However, the insignificant relationship
between emotional appeal and attitude toward empathy expres-
sion appears to be attributed to the product/service class used in
this study. So marketers and advertisers need to pay attention to
the product or service class to be featured in their SNS ad when
creating an advertising campaign, since emotional appeal may or
may not contribute to positive user responses depending on the
product/service category.
6. Limitations and research directions
Though this research contributes to the literature by identi-
fying the antecedents of user responses to SNS advertising and
subsequent marketing responses, it has a few limitations. First,
the predictors we used in our model may not be exhaustive. Fur-
ther research is needed to explore other factors that may lead to
favorable responses to SNS advertising. Second, college students
in their 20s are active SNS users but may not be representative of
the consumer population. Future research may investigate and test
the nomological relationships in our model using a sample con-
sisting of real-world consumers. Third, it would be worthwhile to
replicate this study using a motion picture (e.g., a YouTube video)
instead of a still picture. A recent study found that media type
may affect attitudes to an advertisement on a mobile device and
that a video ad leads to a more positive attitude to advertising
than does a still-image ad (Sung & Cho, 2012). Fourth, this study
did not include the ‘behavior’ construct in the research model,
because it focused on understanding the relationships between
intention and its antecedents. However, the incorporation of the
construct as a dependent variable would have enabled far-reaching
implications for practitioners as well as academics. Moreover, with
such a model setup it would have been possible to examine the
moderating role of privacy concerns in the relationship between
intention and behavior, rather than merely illuminating the direct
effect of privacy concerns on intention. Fifth, while the present
research focuses on ‘liking’ as a user’s behavioral response to a SNS
ad, it would be desirable to examine alternative responses such
as sharing and writing posts which are regarded as more active
user behavior than liking. Finally, future research may examine the
moderating role of product category in the relationship between
emotional appeal and user’s attitude toward empathy expression.
It may be posited that hedonic products or services involve greater
impact of emotional appeal on attitude, whereas utilitarian coun-
terparts involve minimal or insignificant impact. Findings of such
372 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
research will augment our understanding of the factors affecting
the formation of positive user behavior with regards to a SNS ad.
Appendix A. : Advertising stimulus
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and
Human Decision Processes, 50, 179–211.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social
behavior. Upper Saddle River, NJ: Prentice Hall.
Alhabash, S., & McAlister, A. R. (2014). Redefining virality in less broad strokes:
predicting viral behavioral intentions from motivations and uses of Facebook
and Twitter. New Media & Society, 1–23.
Alhabash, S., McAlister, A. R., Quillam, E. T., Rifon, N. J., & Richards, J. I. (2013).
Between likes and shares: effects of emotional appeal and virality on the
persuasiveness of anticyberbullying messages on Facebook. Cyberpsychology,
Behavior and Social Networking, 16, 175–182.
Andrews, J. C. (1989). The dimensionality of beliefs toward advertising in general.
Journal of Advertising, 18, 26–35.
Bagozzi, R. P., & Dholakia, U. M. (2002). Intentional social action in virtual
communities. Journal of Interactive Marketing, 16, 2–21.
Batra, R., & Ray, M. L. (1986). Affective responses mediating acceptance of
advertising. Journal of Consumer Research, 13, 234–249.
Bearden, W. O., Netemeyer, R. G., & Teel, J. E. (1989). Measurement of consumer
susceptibility to interpersonal influence. Journal of Consumer Research, 15,
473–481.
Bowers, T. (2008). Employers who check out job candidates on MySpace could be
legally liable. http://www.techrepublic.com/blog/career/employers-who-
check-out-job-candidates-on-myspace-could-be-legally-liable/338 Accessed
08.08.15
Buchanan, T., Paine, C., Joinson, A. N., & Reips, U. D. (2007). Development of
measures of online privacy concern and protection for use on the Internet.
Journal of the American Society for Information Science and Technology, 58,
157–165.
Burg, N. (2013). How to measure your social media return on investment. Forbes,.
Retrieved from: http://www.forbes.com/sites/capitalonespark/2013/04/25/
how-to-measure-your-social-media-return-on-investment Accessed 10.10.15
Burnkrant, R. E., & Cousineau, A. (1975). Informational and normative social
influence in buyer behavior. Journal of Consumer Research, 2, 206–215.
Chen, H. T., & Kim, Y. (2013). Problematic use of social network sites: the
interactive relationship between gratifications sought and privacy concerns.
Cyberpsychology, Behavior and Social Networking, 16, 806–812.
Chin, C. Y., & Lu, H. P. (2015). Facebook users’ motivation for clicking the Like
button. Social Behavior and Personality, 43, 579–592.
Choi, S. (2013). An empirical study of social network service (SNS) continuance:
incorporating the customer value-satisfaction-loyalty model into the IS
continuance model. The Asian Pacific Journal of Information Technology, 23, 1–28.
Chompon.com (2011). What is the value of a social action in online commerce?
www.chompon.com Accessed 29.10.15.
Chu, S. C. (2009). Determinants of consumer engagement in electronic
word-of-mouth in social networking sites. In Unpublished dissertation. The
University of Texas at Austin.
Cohen, J. B., & Golden, E. (1972). Informational social influence and product
evaluation. Journal of Applied Psychology, 56, 54–59.
Colman, A. M. (2014). The bandwagon effect. In Oxford dictionary of psychology (4th
ed.). New York: Oxford University Press.
Crespo, A. H., & Bosque, I. R. (2008). The effect of innovativeness on the adoption of
B2C e-commerce: a model based on the theory of planned behaviour.
Computers in Human Behavior, 24, 2830–2847.
Davis, M. (1983). Measuring individual differences in empathy: evidence for a
multidimensional approach. Journal of Personality and Social Psychology, 44,
113–126.
Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational social
influences upon individual judgement. Journal of Abnormal and Social
Psychology, 51, 629–636.
Ducoffe, R. H. (1996). Advertising value and advertising on the web. Journal of
Advertising Research, 36, 21–36.
Eckler, P., & Bolls, P. (2011). Spreading the virus: emotional tone of viral
advertising and its effect on forwarding intentions and attitudes. Journal of
Interactive Advertising, 11, 1–11.
eMarketer.com (2012). The ripple effect of following a brand on social media.
http://www.emarketer.com/Article/Ripple-Effect-of-Following-Brand-on-
Social-Media/1009177 Accessed 02.11.15
eMarketer.com (2014). Total US ad spending to see largest increase since 2004.
http://www.emarketer.com/Article/Total-US-Ad-Spending-See-Largest-
Increase-Since-2004/1010982 Accessed 14.12.15
Escalas, J. E., & Stern, B. B. (2003). Sympathy and empathy: emotional responses to
advertising dramas. Journal of Consumer Reserach, 29, 566–578.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: an
introduction to theory and research. Reading, Massachusetts: Addison-Wesley.
Fornell, C., & Larcker, D. F. (1981). Structural equation models with unobservable
variables and measurement error: algebra and statistics. Journal of Marketing
Research, 18, 382–388.
Gao, Y., & Koufaris, M. (2006). Perceptual antecedents of user attitude in electonic
commerce. ACM SIGMIS, 37, 43–50.
Gerlitz, C., & Helmond, A. (2011). Hit, link, like and share. Organising the social and
the fabric of the web. In Digital Methods Winter Conference Proceedings (pp.
1–29).
Godes, D., & Mayzlin, D. (2009). Firm-created word-of-mouth communication:
evidence from a field test. Marketing Science, 28, 721–739.
Gustin, S. (2012). Do Facebook ads work? In Time. Time, Inc. http://business.time.
com/
Haberland, G. S., & Dacin, P. A. (1992). The development of a measure to assess
viewers’ judgement of the creativity of an advertisement: a preliminary study.
Advances in Consumer Research, 19(1), 817–825.
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006).
Multivariate data analysis (6th ed.). Upper Saddle River, NJ: Prentice Hall.
Holbrook, M. B., & Batra, R. (1987). Assessing the role of emotions as mediators of
consumer responses to advertising. Journal of Consumer Research, 14, 404–420.
Huang, J. H., & Chen, Y. F. (2006). Herding in Online Product Choice. Psychology &
Marketing, 23(5), 413–428.
Kelman, H. (1958). Compliance, identification: and internalization: three processes
of attitude change. Journal of Conflict Resolution, 2, 51–60.
Khan, U., Dhar, R., & Wertebroch, K. (2005). A behavioral decision theoretic
perspective on hedonic and utilitarian choice. In S. Ratneshwar, & D. G. Mick
(Eds.), Inside consumption: consumer motives, goals, and desires. New York, NY:
Routledge.
Kline, R. (2005). Principles and practice of structural equation modeling. NY: Guilford
Press.
Kumar, V., & Mirchandani, R. (2012). Increasing the ROI of social media marketing.
Sloan Management Review, 54, 54–61.
Kwon, O., & Wen, Y. (2010). An empirical study of the factors affecting social
network service use. Computers in Human Behavior, 26, 254–263.
Li, X., & Wu, L. (2013). Measuring effects of observational learning and
social-network word-of-mouth (WOM) on the sales of daily-deal vouchers. In
Proceedings of the 46th Hawaii International Conference on System Sciences (pp.
2908–2917).
Liang, H., Saraf, N., Hu, Q., & Xue, Y. (2007). Assimilation of enterprise systems: the
effect of institutional pressures and the mediating role of top management.
MIS Quarterly, 31, 59–87.
Lipsman, A., Aquino, C., & Kemp, P. (2012). The power of like 2: how social
marketing works. http://www.comscore.com/ Accessed 20.07.15
Liu, C. (2014). The impact of social cues and effectiveness in check-in advertising.
Kybernetes, 43, 984–1002.
Logan, K., Bright, L. F., & Grngadharbatla, H. (2012). Facebook versus television:
advertising value perceptions among females. Journal of Research in Interactive
Marketing, 6, 164–179.
Lu, Y., Zhou, T., & Wang, B. (2009). Exploring Chinese users’ acceptance of instant
messaging using the theory of planned behavior, the technology acceptance
model, and the flow theory. Computers in Human Behavior, 25, 29–39.
Lutz, R. J. (1985). Affective and cognitive antecedents of attitude toward the ad: a
conceptual framework. In L. F. Alwitt, & A. A. Mitchell (Eds.), Psychological
processes and advertising effects: theory, research and application. Hillsdale, NJ:
Lawrence Erlbaum Associate.
MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the ad
as a mediator of advertising effectiveness: a test of competing explanations.
Journal of Marketing Research, 130–143.
 J. Lee, I.B. Hong / International Journal of Information Management 36 (2016) 360–373
Madden, T. J., Ellen, P. S., & Ajzen, I. (1992). A comparison of the theory of planned
behavior and the theory of reasoned action. Personality and Social Psychology
Bulletin, 18, 3–9.
McDermott, J. (2014). Facebook bounces back with college kids. Digiday,. Retrieved
from http://digiday.com/platforms/facebook-bounces-back-among-college-
aged-users/ Accessed 23.12.15
Muk, A., & Chung, C. (2014). Driving consumers to become fans of brand pages: a
theoretical framework. Journal of Interactive Advertising, 14, 1–10.
Pöyry, E., Parvinen, P., & Malmivaara, T. (2013). The power of ‘Like’—interpreting
usage behaviors in company-hosted Facebook pages. In 2013 46th Hawaii
International Conference on System Sciences.
Pavlou, P. A., Liang, H., & Xue, Y. (2007). Understanding and mitigating uncertainty
in online exchange relationships: a principal–agent perspective. MIS Quarterly,
31, 105–136.
Pelling, E. L., & White, K. M. (2009). The theory of planned behavior applied to
young people’s use of social networking web sites. Cyberpsychology & Behavior,
12, 755–759.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of
persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology
(Vol. 19) (pp. 123–205). New York: Academic Press.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common
method biases in behavioral research: a critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88, 879–903.
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research:
problems and prospects. Journal of Management, 12, 531–544.
Reinartz, W., & Saffert, P. (2013). Creativity in advertising: when it works and
when it does not. Harvard Business Review, 1–8.
Resnik, A., & Bruce, L. S. (1977). An analysis of information content in television
advertising. Journal of Marketing, 41, 50–53.
Rosen, L. (2012). The power of ‘Like’. Psychology Today,. Retrieved from https://
www.psychologytoday.com/blog/rewired-the-psychology-technology/
201207/the-power [Accessed on 30.12.15]
Rotzoll, K. B., Haefner, J. E., & Sandage, C. H. (1990). Advertising in contemporary
society. Cincinnati, OH: South-Western.
Rubin, C. (2012). Do Facebook likes translate to actual sales? https://www.
americanexpress.com/us/small-business/openforum/articles/do-facebook-
likes-translate-to-actual-sales/ Accessed 19.10.15
373
Sledgianowski, D., & Kulviwat, S. (2008). Social network sites: antecedents of user
adoption and usage. AMCIS 2008 Proceedings, 83.
Smith, R. E., MacKenzie, S. B., Yang, X., Buchholz, L. M., & Darley, W. K. (2007).
Modeling the determinants and effects of creativity in advertising. Marketing
Science, 26, 819–833.
Socialbakers (2011). Facebook by the numbers: user age distribution on Facebook
in South Korea. www.socialbakers.com Accessed 05.09.15
Stieger, S., Burger, C., Bohn, M., & Voracek, M. (2013). Who commits virtual identity
suicide? Differences in privacy concerns, Internet addiction, and personality
between Facebook users and quitters. Cyberpsychology, Behavior, and Social
Networking, 16, 629–634.
Sung, J., & Cho, K. (2012). The influence of media type on attitude toward mobile
advertisements over time. Cyberpsychology, Behavior and Social Networking, 15,
31–36.
Swani, K., Milne, G., & Brown, B. P. (2013). Spreading the word through likes on
Facebook: evaluating the message strategy effectiveness of Fortune 500
companies. Journal of Research in Interactive Marketing, 7, 269–294.
Taylor, D. G., Lewin, J. E., & Strutton, D. (2011). Friends, fans, and followers: do ads
work on social networks? Journal of Advertising Research, 51, 258–275.
Wikipedia.org (2015a). Privacy concerns with social networking services. http://
en.wikipedia.org/wiki/Privacy concerns with social networking services
Accessed 08.09.15
Wikipedia.org (2015b). Social media marketing. http://en.wikipedia.org/wiki/
Social media marketing Accessed 01.10.15.
Xie, F. T., Donthu, N., Lohtia, R., & Osmonbekov, T. (2004). Emotional appeal and
incentive offering in banner advertisements. Journal of Interactive Advertising,
4, 30–37.
Yoo, C. W., Kim, Y. J., Moon, J., & Choe, Y. C. (2008). The effects of herding behavior
and perceived usefulness on intention to purchase e-learning content:
comparision analysis by purchase experience. Asia Pacific Journal of Information
Systems, 18, 105–130.
Zlatolas, L. N., Welzer, T., Hericko, M., & Hölbl, M. (2015). Privacy antecedents for
SNS self-disclosure: the case of Facebook. Computers in Human Behavior, 45,
158–167.