Journal of Business Research 134 (2021) 223–232

Contents lists available at ScienceDirect

Journal of Business Research

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

Trust me, trust me not: A nuanced view of influencer marketing on

social media
Do Yuon Kim a, *, Hye-Young Kim b
a
Department of Consumer and Design Sciences, Auburn University, AL 36849, USA
b
Retail Merchandising, University of Minnesota, Twin Cities, MN 55108-6136, USA

A R T I C L E I N F O A B S T R A C T

Keywords: Social media influencers are widely employed as a marketing strategy as they successfully attain trust from
Social media followers. Applying the social exchange theory and its principle of reciprocity, this study investigated whether
Influencer the source characteristics of an influencer (i.e., expertise, authenticity, physical attractiveness, homophily) can
Social exchange theory
function as relational resources in the formation of follower trust. Whether followers’ trust in the influencer leads
Reciprocity
Trust
to their loyalty to the influencer and desirable marketing outcomes (i.e., product attitude, purchase intention)
Relationship strength was also examined. Results revealed that trust mediated the impacts of expertise, authenticity, and homophily on
loyalty and marketing outcomes. However, physical attractiveness was not significant in building relational trust.
Also, the moderating role of relationship strength was confirmed in authenticity-trust and trust-loyalty linkages.
The findings suggest implications for the strategic use of influencer marketing and provide a better under­
standing of persuasion mechanisms manifested in influencer-follower relationships.

1. Introduction sales, but its intended goals involve much broader outcomes, such as
gaining the attention of potential customers, generating word-of-mouth
The past decade has witnessed a major change in social media effects, and creating customer engagement with a brand or a product
marketing characterized by a shift toward influencer marketing (Phua (Brown & Hayes, 2008).
et al., 2017). The size of the influencer marketing industry reached $9.7 In the increasingly competitive social media environment, influencer
billion in 2020 and is expected to grow to $15 billion by 2022 (Statistica, marketing is evolving into long-term partnerships between brands and
2021). With a market of only half a million dollars in 2015, the industry influencers (McNutt, 2021). Marketers have recognized the value of
has grown rapidly by more than 50% a year (Statistica, 2021). The continuously weaving their marketing messages into influencers’ nar­
growth of influencer marketing accelerated with the outbreak of COVID- ratives over a long timeframe to achieve greater engagement from fol­
19, during which people increasingly turned to social media for enter­ lowers (Robertson, 2020), indicating that influencer marketing can be
tainment and virtual social experiences (Etzkorn, 2021). Consequently, viewed as multi-layered relationship marketing involving influencer-
influencer marketing has become an essential part of digital marketing consumer, influencer-brand, and brand-consumer relationships. Specif­
strategy as a touchpoint for reaching a target audience. ically, influencers deliver the brand’s message to followers by leveraging
In general, social media influencers have sizable groups of followers the pre-established relationship and trust that they have cultivated. It is
and serve as experts within their content areas. According to the therefore vital to exploit relational trust, the building block of an
(Interactive Advertising Bureau, 2018), the term “social media influ­ influencer-follower relationship (Brooks & Piskorski, 2018) Evidencing
encers” is defined as those users who “have the potential to create the importance of trust for influencer marketing, one study reported that
engagement, drive conversation, and/or sell products/services with the 92% of social media users trust influencers more than traditional mar­
intended target audience; these individuals can range from celebrities to keting channels (Eyal, 2018). In global consumer surveys, 46% of re­
more micro-targeted professional or nonprofessional ‘peers’. At a spondents do not trust newspapers, magazines, TV, and radio (Ipsos,
fundamental level, influencer marketing is a type of endorsement mar­ 2019), and 63% rely significantly more on influencers’ recommenda­
keting that uses product recommendations from influencers to drive tions than on brand advertising through traditional media (Edelman,

* Corresponding author.
E-mail addresses: [email protected] (D.Y. Kim), [email protected] (H.-Y. Kim).

https://doi.org/10.1016/j.jbusres.2021.05.024
Received 25 November 2019; Received in revised form 11 May 2021; Accepted 15 May 2021
Available online 29 May 2021
0148-2963/© 2021 Elsevier Inc. All rights reserved.
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

2019). This new environment encourages social media marketers to influencers’ social impact on followers. By sharing valuable information,
develop a new set of marketing competencies and knowledge of building entertainment, and attractive characteristics, the influencer can achieve
and managing trust between influencers and followers. dependence and consequent social impact among followers. Therefore,
However, existing marketing literature does not offer prescriptive the reliance of followers on the influencer gives the influencer social
guidelines for this current marketing phenomenon. Research regarding power.
relationship development and the leverage of trust between influencers The social exchange dynamics follow the principle of reciprocity
and followers are limited given the overemphasis of past studies on (Gouldner, 1960; Malinowski, 1922) The reciprocity norm suggests that
transactional outcomes such as followers’ perceptions, attitudes, and social exchange occurs by reward, and the likability of the exchange is
behavioral intentions (Jin et al., 2019; Lou & Yuan, 2019; Schouten determined by the value of the reward (Gouldner, 1960). The exchange
et al., 2020; Sokolova & Kefi, 2020; Xiao et al., 2018). Therefore, how of resources continues toward equilibrium where both parties are
relational trust and loyalty can be established and capitalized has not rewarded equally. In the influencer-follower interaction, the influencer
been clarified. Furthermore, prior studies investigating the source is likely to upload content that can receive a greater number of views
credibility of influencers have restrictively explored the antecedents of and likes, which reflects the value of the reward. The number of fol­
influencer credibility (Breves et al., 2019; Lou & Yuan, 2019; Sokolova lowers and their engagement level also depend on the value of the
& Kefi, 2020) without understanding how and when these antecedents influencer’s content. The mutual reinforcement between influencer and
can be utilized effectively. Although evidence about the role of trust in followers will be continued to meet the balance of exchange.
online influencer marketing has begun to accumulate (Brooks & Pis­ Furthermore, the exchange of resources contributes to psychological
korski, 2018), empirical validations are limited and a clear picture has ties and motivates individuals to maintain relationships (Blau, 1964).
yet to emerge. Those who engage in a reciprocal relationship can establish a high level
In an attempt to fill this gap and in response to the current practice of of trust, affective regard, and behavioral commitment (Molm, 1990;
influencer marketing on social media, this study applies the social ex­ Molm et al., 2000). Such bonding constructed upon reciprocal ex­
change theory (Homans, 1961) Specifically, this study builds on the changes result in a greater perception of unity and harmony in the
reciprocity principle (Blau, 1964) of the social exchange theory to fully relationship (Molm et al., 2007). The reciprocity also affects the stability
capture the nature of influencer-follower relationships and the persua­ of relationship. If the exchanges between partners are continuously
sion mechanism of influencer marketing. In sum, the basic premise of unbalanced, the relationship becomes unstable. According to the reci­
this study is that the successful exchange of social resources between procity principle, this study proposes that trust and loyalty are the
relationship partners (i.e., the influencer and followers) can build psy­ relationship outcomes of the successful social exchanges between
chological ties and motivate partners to maintain the relationship. influencers and followers. Influencers will be able to obtain interper­
Satisfactory exchanges between influencer and followers will positively sonal trust and loyalty from the followers through continuous exchange
reinforce the relational bonding and enrich the quality of the relation­ activities. When influencers succeed to meet the expectation of followers
ship over time, generating greater customer engagement organically and are rewarded with views, likes, and comments, a stronger bonding
(Venkatesh, 2020). With this in mind, this study aims to examine of the relationship will be formulated.
whether the specific characteristics of the influencer can function as The social exchange theory was applied to the interpersonal
social exchange resources in the formation of trust. In so doing, the communication (Gatignon & Robertson, 1986) to explain how people
source credibility and attractiveness models are comprehensively inte­ communicates based on the exchange of costs and benefits. According to
grated to identify the drivers of relational trust. Furthermore, whether this perspective, a speaker who provides information or other valuable
followers’ trust in the influencer leads to loyalty to the influencer and to resources can acquire support, gratitude, recognition, and social status
desirable marketing outcomes (i.e., product attitude and purchase from the listener. It explains the mechanism by which influencers can
intention) is examined. Finally, the moderating role of relationship exert social impacts on other social media users. It also suggests that the
strength (strong vs. weak) is tested in the interrelationships proposed in exchange of interpersonal influence relies on source characteristics and
the model to provide a more nuanced view of influencer marketing. the perceived intent of the message. Specifically, the impact of source
characteristics and intention are developed into two models (Harmon &
2. Literature review and hypothesis development Coney, 1982): the source credibility (Hovland, Janis, & Kelley, 1953)
and source attractiveness (McGuire, 1968)Mills & Aronson, 1965).
2.1. Social exchange theory of interpersonal communication Grounded in this theoretical notion, this study suggests that credible and
attractive source characteristics are exchanged for follower trust.
Social exchange theory (Homans, 1961; Thibaut & Kelley, 1959)
provides a theoretical framework for influencer marketing. According to 2.2. Source credibility model: expertise and authenticity
this theory, human behavior is explained by the “exchange of activity,
tangible or intangible, and more or less rewarding or costly, between at Source credibility is defined as the image of a speaker held by a
least two people” (Homans, 1961, p.13). When a person invests re­ listener at a given time (Andersen & Clevenger, 1963). Aristotle referred
sources in a relationship, the partner is expected to behave in a way that to communicator credibility as a listener’s favorable disposition toward
rewards the person. The exchange of resources in social interactions is the speaker, indicating that a communicator’s positive quality can in­
mutually reinforced by each actor’s behavior (Blau, 1964; Homans, fluence receivers’ acceptance of the message (Ohanian, 1990). The di­
1961) In the influencer marketing context, an influencer uploads con­ mensions of source credibility are largely grouped into two categories:
tent (e.g., videos, images, and stories) that is useful, enjoyable, and speaker expertise and speaker intent (Hovland et al., 1953)
appealing to followers. Then, the followers indicate their satisfaction Expertise indicates the “extent to which a communicator was
and appreciation through activities such as liking, sharing, commenting, perceived to be a source of valid assertions” (Hovland et al., 1953), p.
and subscribing activity as rewards (O’Donell, 2018). 21). A speaker’s expertise can be perceived from the quantity and
Social exchange theory also accounts for the mechanism by which quality of information, the degree of ability, education, and professional
social status and power are generated through social interaction (Cook & achievement, and the validity of the speaker’s judgment (Giffin, 1967;
Emerson, 1978; Emerson, 1976; Skvoretz & Willer, 1993). When a (McGuire, 1968) A speaker’s credibility establishes authoritativeness,
person offers valuable resources dominantly, the relationship partner further contributing to perceived information quality and validity (Gif­
becomes reliant on the interaction. The interpersonal interdependence fin, 1967). Therefore, expertise as source credibility results in opinion
results in relative power and social structure in the dynamics of social agreement and attitudinal change in marketing communication.
exchange (Cook & Yamagashi, 1992). Interdependence also applies to The intention of a speaker is defined as the “degree of confidence in

224
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

the communicator’s intent to communicate assertions he considers most agreement (Chaiken, 1979; Horai et al., 1974). Physically attractive
valid” (Hovland et al., 1953) p. 21). Intention has also been referred to individuals are likely to be perceived as kind, interesting, sociable,
as trustworthiness and sincerity (Kenton, 1989). The intention perceived strong, modest, and responsive (Dion et al., 1972). These individuals are
by the message receiver can be affected by the speaker’s attempt to also perceived as motivated, decisive, informed, and logical (Dipboye
persuade and manipulate. Such intentions create inferences about the et al., 1977). Such positive perceptions of attractive communicators
speaker’s self-interest as opposed to purely transferring information. increase credibility and message acceptance.
Therefore, the perceived intention of a speaker can determine the Homophily refers to the perceived similarity in beliefs, values, ex­
persuasiveness of their message. periences, and lifestyles of the communicator by the receiver (Gilly
Specifically, this study suggests that influencers’ intentions can be et al., 1998; Lazarsfeld & Merton, 1954). Relatable communicators tend
operationalized as authenticity. In sociology, authenticity is defined as to be influential and can greatly affect recommendations (v. Wangen­
being true to oneself or to others (Vannini & Franzese, 2008; Zickmund, heim & Bayón, 2004). Perceived similarity creates positive intergroup
2007), and implies sincerity, genuineness, truthfulness, and originality feelings and reduces uncertainty (Gerard & Greenbaum, 1962; Simons
(Molleda, 2010). In marketing communication, Baker and Martinson et al., 1970). In such a situation, a listener is likely to infer that the
(2002) argued that authenticity can be determined by “whether the communicator’s attitude, interests, beliefs, and feelings are similar to
practitioner is willing to openly, publicly, and personally be identified as their own, resulting in the listener endorsing the opinion.
the persuader” (Baker & Martinson, 2002, p. 17). In social media mar­ This study proposes that source attractiveness can develop trust in
keting, authenticity corresponds to the genuine intention of the influ­ social media influencers. Specifically, physical attractiveness and
encer given they can post and recommend any product/service for homophily can be explained by a different process. First, physical
external compensation (Boerman et al., 2017; Evans et al., 2017). Social attractiveness can infer qualities of credibility, such as being well
media users are aware of the existence of sponsorship and may be informed and logical (Dipboye et al., 1977). These positive inferred
skeptical of the influencer’s reasons for promoting the product/service. traits enhance the credibility of social media influencers. Thus, the
Thus, authenticity allows followers to believe that the influencer posting physical attractiveness of a social media influencer increases followers’
is based on sincere opinion. trust. Furthermore, the perceived similarity with the speaker creates the
The credibility of a source enhances persuasion effectiveness and impression that the communicator shares a similar background with
listener advocacy (Homer & Kahle, 1990; Horai et al., 1974; Hovland & listeners (McGuire, 1968), and that the speaker’s argument is consistent
Weiss, 1951; Pornpitakpan, 2004). Credibility can induce a listener’s with that of the listeners (Simons et al., 1970). This concept of building
favorable attitude toward the source and result in receptivity to the greater trust in a speaker among listeners can be applied to social media
message. For instance, a statement generated a greater change of influencers. Therefore, homophily to the influencer leads to trust in the
opinion when it was delivered by a highly credible source with a influencer. We thus propose the following hypotheses:
trustworthy character than by a source with low credibility (Hovland &
Weiss, 1951). In addition, a communicator who exhibits expertise in a • H3: The attractiveness of an influencer positively influences trust in
certain topic can receive significant agreement from listeners (Maddux the influencer.
& Rogers, 1980). The speaker’s intention determines acceptance of the • H4: Homophily to an influencer positively influences trust in the
message. influencer.
Similarly, this study proposes that each of the two dimensions of
source credibility, expertise, and authenticity, leads to relational trust. 2.4. Trust in the influencer
Specifically, an influencer’s ability, experience, and competence induce
followers to trust the influencer. This can be evidenced by a previous Trust is defined as the “confidence in an exchange partner’s reli­
qualitative study in which social media users attend to a source’s ability and integrity” (Morgan & Hunt, 1994, p. 23). Trust can be
competence in knowledge or experience when following Instagram ce­ established when partners fulfill their promises (Grönroos, 1990).
lebrities (Djafarova & Rushworth, 2017; Djafarova & Trofimenko, Therefore, an actor should invest resources in a way that the partner
2018). Thus, the perceived expertise of the influencer will result in trust perceives as reliable in building relational trust. In communication,
in the relationship. Furthermore, perceived intention of influencers de­ trusting a speaker reflects the listener’s confidence and willingness to
termines dependability, consistency, and predictability (Giffin, 1967). rely on the message. If an actor sends behavioral and social cues that
Therefore, the authenticity of social media influencers also leads to promise future rewards, the receiver’s trust in the actor is likely to
relational trust. Accordingly, we propose the following two hypotheses: increase.
Trust is conceptualized as a relationship trait established through
• H1: The expertise of an influencer positively influences trust in the continuous interactions. Individuals can estimate and assess the value of
influencer. future exchanges based on relational trust in their partners. Therefore,
• H2: The authenticity of an influencer positively influences trust in trust can guarantee a desirable social exchange for relationship partners
the influencer. and contribute to the maintenance of relationships (Gassenheimer et al.,
1998; Mayer et al., 1995; Sirdeshmukh et al., 2002). Similarly, trust in
2.3. Source attractiveness model: physical attractiveness and homophily the influencer assures followers that their relationship with the influ­
encer will affect them positively. As a result, followers seek to maintain
Source attractiveness increases the impact of communication by this relationship and become loyal to the influencer. We therefore sug­
generating considerable attention and engagement (Mills & Aronson, gest the following hypothesis:
1965; Sternthal & Samuel, 1982). When a communicator is attractive,
the audience may be willing to like and accept the communicator’s • H5: Trust positively influences loyalty to the influencer.
message. The source attractiveness model (McGuire, 1968) considers
source characteristics such as familiarity, similarity, likability, and Furthermore, trust in the influencer makes followers regard the
attractiveness. In the context of influencer marketing, physical attrac­ relationship as rewarding and enhances message effectiveness. As fol­
tiveness and homophily were adopted as two dimensions of source lowers believe the influencer’s message will bring a positive outcome,
attractiveness. they expect the influencer’s endorsement to be beneficial. For instance,
The communicator’s physical attractiveness as perceived by the in a relationship with a salesperson, trust based on previous experiences
receiver affects the receiver’s initial judgment (Baker & Churchill, 1977; reduced uncertainty (Zeithaml, 1981) and enhanced sales effectiveness
Joseph, 1982; Snyder & Rothbart, 1971) and following opinion (Crosby, 1990). Thus, followers have a positive attitude and a high

225
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

purchase intention toward the product recommended by the influencer. literature: Homophily (e.g., “The source is similar to me in preferences
and values.”) (v. Wangenheim & Bayón, 2004), Trust (e.g., “The influ­
• H6: Trust positively influences product attitude. encer can be relied upon on his (her) content.”) (Kennedy et al., 2001),
• H7: Trust positively influences purchase intention. Loyalty to the Influencer (e.g., I intend to continue following the influ­
encer”) (Parasuraman et al., 2005), Product Attitude (e.g., “Likable/ Not
2.5. Moderating influence of relationship strength likable”) (Madden et al., 1988), and Purchase Intention, (e.g., “How
likely are you to purchase the product?”) (Taylor et al., 1975).
Social media users can develop different levels of relationship
strengths with influencers. Specifically, the frequency of visits and in­ 4. Results
vestment of time can drive the strength of the relationship with the
influencer. Relationship strength, also referred to as “tie strength”, is 4.1. Sample
defined by Granovetter (1973, p.1361) as a “combination of the amount
of time, the emotional intensity, the intimacy, and the reciprocal ser­ Data were collected from U.S. adults aged 18 or older (N = 384) who
vices.” When this variable was applied to the relationship between in­ had been following an influencer on social media using Amazing
dividuals and organizations, strong (vs. weak) relationships benefited M− Turk platform. Each participant received $ 0.50 as compensation.
individuals and organizations and created a better performance Approximately 48.7 percent of the sample was male (n=187) and 50.5
(Krackhardt & Stern, 1988; Schaefer et al., 1981; Uzzi, 1999). In addi­ percent was female (n=194). The majority of the sample was Caucasian
tion, in consumer-seller relationship, the relationship strength was ad­ (64.6%) followed by Asian (10.7%), African American (8.9%), Latino
vantageous by increasing the commitment to seller (Stanko et al., 2007) (7.3%), and others. The majority of participants (73.4%) reported that
and decreasing the complaint likelihood in a service failure (Mittal et al., they use social media daily (n = 282) and 35.2 percent checks the
2008; Yang & Mattila, 2012). postings of the influencer every day. The participants used social media
Relationship strength has been extended to computer-mediated platforms including YouTube (82%; n=315), Facebook (78.1%; n=300),
communication, especially in social media (Gilbert & Karahalios, Instagram (71.6%; n=275), Twitter (53.6%; n=206), and Pinterest
2009; Jones et al., 2013; Rapp et al., 2013; Wang et al., 2012). In social (29.2%; n=112), when they were able to choose multiple platforms. As
media, relationship strength is defined as the degree to which bonds the content of one influencer is not restricted to one topic, participants
among members of a social networking service are strong or weak were asked to choose multiple topics for the specific influencer that they
(Mittal et al., 2008). Relationship strength was found to affect the indicated. The following topics of influencer content were identified:
friendship on social networking sites as well as the social media fashion (35.7%; n=137), beauty (26.3%; n=101), travel (16.7%; n=64),
engagement (Chahal & Rani, 2017; Shan & King, 2015). Furthermore, technology and game (15.4%, n=59), and food (14.6; n=56).
the relationship strength intervenes the effect of social network char­
acteristics on user behavior. One study showed that relationship 4.2. Measurement model
strength moderated the effect of the social networking site (e.g., Face­
book, Twitter, Instagram, and Snapchat) on consumer engagement, The measurement model was tested using AMOS 27.0 and SPSS
commitment, and identification to the brand community (Phua et al., software. The reliability of each construct was assessed Cronbach’s
2017). Building on this research stream, the focal question explored in alpha. A value of 0.70 or above indicates an acceptable level of reli­
this study is: Does relationship strength provide a more nuanced view of ability (Nunnally, 1978). As the Cronbach’s value ranged between 0.82
influencer marketing? To address this question, the following hypothesis and 0.94, all measurements were evaluated as being acceptable;
was formulated: Expertise (α = 0.88), Authenticity (α = 0.91), Physical Attractiveness (α
= 0.89), Homophily (α = 0.82), Trust (α = 0.89), Loyalty to Influencer
• H8. Relationship strength plays a moderating role in H1-H7. That is, (α = 0.91), Product Attitude (α = 0.92), and Purchase Intention (α =
there will be a different pattern of paths between the strong vs. weak 0.94).
relationship groups. A confirmatory factor analysis (CFA) with maximum likelihood
estimation was then conducted (see Table 1). The value of factor load­
3. Methods ings, item error variances, and frequency of large modification indices
were collectively evaluated for deciding which items to delete. In this
An online self-administered, cross-sectional survey methodology was stage, one item of physical attractiveness was removed to improve the
employed to collect the data. Participants were asked to think about the measurement model (Bagozzi and Yi, 1998); Hair et al., 2006). The
social media influencer that they view the most frequently in the past measurement model exhibited an acceptable fit with the data: (χ 2 =
month and identify the name of the influencer. Then they answered the 684.18, df = 402, p < .000; χ 2/df = 1.70, CFI = 0.97, NFI = 0.93, IFI =
frequency of viewing, the average time spent watching the influencer’s 0.97, RMSEA = 0.04, SRMR = 0.05) (Hair et al., 1998).
content, and the duration of following the influencer. The viewing fre­ Then, convergent and discriminant validity were examined by
quency and average time spent were derived from the measurement of average variance extracted (AVE), composite reliability (CR), and cor­
relationship/tie strength in the previous study (Levin & Cross, 2004; Chu relation coefficients among the variables (see Table 2). The convergent
& Kim, 2011). Those who did not indicate the specific name of the validity was assessed by testing if the composite reliability for each
influencer were excluded. In the following section, they were asked to construct exceeds the recommended level of 0.70, and the average
complete the questionnaire concerning the influencer they had variance extracted (AVE) for each construct fulfills the benchmark of
identified. 0.50 (Hair et al., 1998). The results found that the composite reliability
The measurement items for each construct in the hypothesized fell between 0.82 and 0.94, exceeding the recommended value of 0.70
model were adopted and modified from the preexisting measurements (Bagozzi & Yi, 1988). The average variance extracted for all constructs
that were validated in the previous studies. Influencer characteristics ranged between 0.60 and 0.84, greater than the acceptable value of 0.50
were measured through 7-point semantic differential scales: Expertise (Fornell & Larcker, 1981). Therefore, the convergent validity was sup­
(e.g., “Experienced/ Not experienced”), Authenticity (e.g. “Dependable/ ported. The discriminant validity was verified by confirming if the
Not dependable”), and Physical Attractiveness (e.g., “Attractive/ Unat­ correlations between constructs are lower than 0.85 and if the AVE of
tractive”) adopted from Ohanian (1990). Also, scales for each of the each construct exceeds the squared inter-correlations between latent
following constructs were measured on a 7-point Likert rating scale (1 = constructs (Fornell & Larcker, 1981). The correlations between con­
Strongly Disagree, 7 = Strongly Agree) by adopting the preexisting structs were between 0.21 and 0.81 and the average variance extracted

226
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

Table 1
Measurement validity and reliability.
Factor Items FLa αb CRc AVEd

Expertise Experts 0.83 0.88 0.90 0.68

Knowledgeable 0.82
Qualified to offer 0.87
Skilled 0.781
Authenticity Dependable 0.83 0.91 0.92 0.69
Honest 0.83
Reliable 0.85
Sincere 0.81
Trustworthy .851
Physical Attractiveness Attractive 0.88 0.89 0.89 0.66
Beautiful 0.90
Sexy 0.77
Elegant 0.701
Homophily Have similar interests 0.71 0.82 0.82 0.60
Have similar enthusiasm 0.75
Have similar values 0.861
Trust The influencer can be relied upon on his (her) content. 0.85 0.89 0.88 0.64
I believe what this influencer says and that he/she would not try to take advantage of the followers. 0.86
The influencer is straightforward and honest even though his/her self-interests are involved 0.77
The influencer would not tell a lie even if he/she could gain by it. 0.701
Loyalty to the Influencer I would recommend this influencer to someone who seeks my advice 0.831 0.91 0.90 0.69
I say positive things about this influencer 0.89
I intend to continue following this influencer 0.79
I will continue to watch the posting of this influencer 0.81
Product Attitude Likable 0.851 0.92 0.92 0.74
Desirable 0.84
Interesting 0.88
Favorable 0.88
Purchase Intention How likely are you to purchase the product? 0.931 0.94 0.94 0.84
How inclined are you to purchase the product? 0.91
How willing are you to purchase the product? 0.90

Notes: 1 Loadings fixed to 1 in unstandardized solution.

a. Factor loading b. Cronbach’s α c. Composite reliability d. Average variance extracted.

Table 2
Discriminant and convergent validity of constructs.
1 2 3 4 5 6 7 8

1. Expertise 0.68
2. Authenticity 0.66 0.69
3. Physical Attractiveness 0.10 0.15 0.66
4. Homophily 0.30 0.44 0.05 0.60
5. Trust 0.43 0.58 0.07 0.42 0.64
6. Loyalty 0.52 0.60 0.08 0.36 0.61 0.69
7. Product Attitude 0.38 0.36 0.11 0.33 0.41 0.46 0.74
8. Purchase Intention 0.19 0.18 0.05 0.21 0.24 0.27 0.43 0.84

Notes: Numerical value of diagonal: AVE, Numerical value of bottom of diagonal: squared correlation coefficient between constructs.

(AVE) for all constructs exceeded the squared correlation coefficients using PROCESS SPSS macro (Model 4, n = 5000 resamples; Hayes,
between variables, providing evidence of discriminant validity. 2013). The result found that trust mediated the effects of expertise,
authenticity, and homophily on marketing outcomes. Specifically, trust
4.3. Hypothesis testing mediated the effect of expertise on loyalty (indirect effect: 0.26, 95% CI
= [0.1866, 0.3564]), product attitude (indirect effect: 0.24, 95% CI =
The structural equation model (SEM) was estimated to test the pro­ [0.1518, 0.3391]), and purchase intention (indirect effect: 0.26, 95% CI
posed relationships (see Fig. 1). The fit statistics indicated an acceptable = [0.1578, 0.3749]). Trust also mediated the effect of authenticity on
model fit (χ 2 = 833.09, df = 417, p < .000; χ 2/df = 1.99, CFI = 0.96, NFI loyalty (indirect effect: 0.26, 95% CI = [0.1711, 0.3538]), product atti­
= 0.92, IFI = 0.96, RMSEA = 0.05, SRMR = 0.80). The results revealed tude (indirect effect: 0.27, 95% CI = [0.1666, 0.3974]), and purchase
that expertise (β = 0.20, p < .05), authenticity (β = 0.45, p < .000) and intention (indirect effect: 0.33, 95% CI = [0.1963, 0.4734]). For the effect
homophily (β = 0.29, p < .000) were positively related to trust. There­ of homophily, the mediation effect of trust was significant on loyalty
fore, H1, H2, and H4 were supported. However, physical attractiveness (indirect effect: 0.28, 95% CI = [0.2062, 0.3654]), product attitude (in­
was insignificant (β = 0.10, p = .80), rejecting H3. Furthermore, trust direct effect: 0.23, 95% CI = [0.1551, 0.3089]), and purchase intention
had positive influences on loyalty to the influencer (β = 0.87, p < .000), (indirect effect: 0.22, 95% CI = [0.1417, 0.3122]). Therefore, it was
product attitude (β = 0.74, p < .000), purchase intention (β = 0.58, p < confirmed the mediation of trust between influencer characteristics (e.g.
.000). Thus, H5, H6, and H7 were supported. expertise, authenticity, homophily) on marketing outcomes (e.g. influ­
encer loyalty, product attitude, and purchase intention).
4.4. Mediation analysis of trust

The mediation effect of trust was examined by bootstrapping analysis

227
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

Fig. 1. Structural equation model and standardized coefficients for testing hypotheses.

4.5. Moderation effect of relationship strength unconstrained model was tested (see Table 4). The influence of
authenticity on trust between strong relationship (β = 0.31, t = 2.07, p
As hypothesized in H8, the influences of influencer characteristics < .05) vs. weak relationship (β = 0.60, t = 4.72, p < .000) was signifi­
are expected to vary by relationship strength. To test the differential cantly varied. Also, the influence of trust on the loyalty to the influencer
influences by relationship strength, a multi-group SEM analysis was had a significant difference between strong relationship (β = 0.87, t =
conducted between strong (n = 206) and weak (n = 178) relationship 10.50, p < .000) vs. weak relationship (β = 0.80, t = 7.41, p < .000). The
groups. The participants were split by the median value of relationship other paths did not reveal significant differences between two groups.
strength measured with a composite variable of the frequency of views
and average time spent (Levin & Cross, 2004; Chu & Kim, 2011). To do 5. Discussion
so, the model of the hypothesized relationships (i.e., base model) was
tested and the standardized coefficient values between the two groups This study presented and tested a trust-based model of influencer
were compared (Dabholkar & Bagozzi, 2002) marketing consisting of source credibility and attractiveness constructs
First, the model fit of the unconstrained baseline model in which with relationship strength as a moderating variable. In the context of
parameters were freely estimated for both groups exhibited a good fit
(χ 2 = 1585.42, df = 836, p < .000; χ 2/df = 1.90, CFI = 0.92; NFI = 0.85;
Table 4
TLI = 0.91; IFI = 0.92; RMSEA = 0.48) by confirming the configural
Comparison of hypothesized paths between strong and weak relationship
invariance. Second, the metric invariance was assessed by constraining groups.
all path parameters between the two groups to examine if the con­
Hypothesized Strong relationship Weak relationship χ2
strained model was invariant between the groups. The chi-square dif­
paths difference
ference between the unconstrained model and measurement model was Std. T-value Std. T-value
(t-statistic)
Coeff. Coeff.
insignificant (Δχ 2 = 23, Δdf = 25.94, p = .30) and confirmed the metrics
invariance. Then, the model fit difference between the two groups was H1 Expertise → 0.36 2.97** 0.03 0.26 ns 2.49 (1.59)
ns
examined between the unconstrained model and the structural weight Trust
H2 Authenticity → 0.31 2.07* 0.60 4.72*** 5.13 (2.31)
model. It exhibited a significant change in model fit (Δχ 2 = 61, Δdf =
Trust *
122.21, p = .000) and proved the moderating influence of relationship H3 Attractiveness 0.03 0.56 ns − 0.03 − 0.44 0.41 (0.64)
strength in supportive of H8 (see Table 3). Specifically, each path was → Trust ns ns

constrained respectively and the chi-square difference with the H4 Homophily → 0.29 2.98** 0.26 3.48*** 0.14 (0.38)
ns
Trust
H5 Trust → Loyalty 0.87 10.50*** 0.80 7.41*** 5.41 (2.15)
Table 3 to the *
Multiple-group structural model invariance test. Influencer
H6 Trust → Product 0.69 6.50*** 0.70 9.76*** 1.33 (1.12)
df χ2 Δdf Δχ 2 p Model ns
Attitude
invariance
H7 Trust → 0.46 5.30*** 0.55 7.48*** 0.80 (0.88)
ns
Base model 218 1585.42 61 122.21 0.000 No Purchase
Constrained 157 1707.63 Intention
model
Notes: *** p < .001, ** p < .01, * p < .05, ns- Not significant.

228
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

social media, this study provides a social exchange paradigm to predict consistent emphasis on trust for influencer marketing (Falls, 2021).
trust in influencer-follower relationships. Specifically, the results sug­ Specifically, mediation analysis indicated that trust mediated the impact
gest that influencers’ expertise, authenticity, and homophily play a of expertise, authenticity, and homophily on marketing outcomes.
significant role in building follower trust. These findings are consistent Therefore, trust mediates the social exchange process of influencer re­
with the source credibility model (Hovland et al., 1953) and the source sources to persuade followers.
attractiveness model (McGuire, 1968; Mills & Aronson, 1965). When
influencers are perceived as reliable and trustworthy, they can achieve 6. Theoretical and managerial implications
relational trust (Giffin, 1967; Ohanian, 1990; (Sternthal, Phillips, &
Dholakia, 1978) Homophily is also crucial to building trust, extending This study contributes to social media marketing literature by
the research findings derived from interpersonal relationships applying the social exchange theory (Homans, 1961); Thibaut & Kelley,
(McPherson et al., 2001) to influencer marketing on social media. 1959) and the principle of reciprocity (Malinowski, 1922)to
Surprisingly, there was no significant relationship between physical influencer-follower relationships. In doing so, this study provides
attractiveness and trust. This contradicts the source attractiveness beneficial insights into the persuasion mechanism of influencer mar­
model, in which physical attractiveness creates positive inference of keting. Building on the social exchange theory highlighting that the
positive source personality traits such as intelligent, sincere, and successful exchange of resources reinforces actors to continue the
genuine (Dipboye et al., 1977; Joseph, 1982). Thus, this empirical study interaction (Gouldner, 1960) and create psychological ties between
does not support the assumption that physical attractiveness is an them (Blau, 1964), this study presents a novel view of influencer mar­
effective social exchange driver for building trust in influencer-follower keting as multi-layered relationship marketing on social media.
relationships. This finding indicates that the nature of influencer- This study also provides empirical evidence of relationship strength
follower relationships is not based on instant interaction or communi­ as a critical variable in the influencer marketing process. While previous
cation. As followers continuously observe the influencer over time and studies focused on identifying the necessary conditions of influencer
across multiple postings, the effect of physical attractiveness on initial credibility (Breves et al., 2019; Lou & Yuan, 2019; Sokolova & Kefi,
judgment (Dipboye et al., 1977) was not extended to trust in the long- 2020), they did not fully capture when and how credible characteristics
term relationship. This assertion is further supported by participants’ could enhance the effectiveness of influencer marketing. Besides, mod­
responses to a survey question of how long they had subscribed to the erators such as product-endorser fit, self-discrepancy, perceived motive,
influencer. Interestingly, 55.3% of participants had followed the influ­ and sponsorship memory have been limitedly applied to the self-
encer for more than a year and 37.4% for 1–12 months. Only 5.5% identification process with the influencer (Jin et al., 2019; Schouten
indicated having followed the influencer for 1–4 weeks and 1.8% for less et al., 2020; (Shan et al., 2020) and the sponsorship disclosure context
than a week. Thus, we can interpret that the long duration of relation­ (Boerman & Van Reijmersdal, 2020); Evans et al., 2017), but not to the
ship resulted in the ineffectiveness of physical attractiveness as a social source characteristic models. Therefore, this study contributes to liter­
exchange driver in developing followers’ trust. The findings of this study ature on influencer marketing by identifying the moderating role of
suggest that the effect of physical attractiveness on relational trust does relationship strength on influencer credibility and attractiveness. Spe­
not persist beyond initial judgment. cifically, relationship strength was incorporated from previous social
The multigroup analysis revealed that the impact of authenticity on network studies (Gilbert & Karahalios, 2009; Jones et al., 2013; Rapp
trust and the effect of trust on loyalty varied according to relationship et al., 2013; Wang et al., 2012) into the context of influencer marketing.
strength. First, the impact of authenticity on trust in the influencer was As a result, the issue of how influencer marketing can be tailored and
significantly stronger for weak relationships than for strong relation­ fine-tuned by the intensity of the influencer-follower relationship is
ships. This implies that authenticity is more crucial when followers do addressed.
not have a strong relationship with the influencer. When followers have This study contributes to the advancement of models of source
frequently visited the influencer and spent a significant amount of time credibility (Hovland et al., 1953) and attractiveness (Giffin, 1967;
viewing the influencer’s content, authenticity becomes less critical to McGuire, 1968; Mills & Aronson, 1965). Specifically, this study found
followers’ trust in the influencer. The impact of authenticity can be that the influencer’s physical attractiveness did not exert an impact on
diminished through interactions and mitigated as the relationship trust in the influencer. This finding implies that physical attractiveness
deepens. Furthermore, the impact of trust on loyalty to the influencer can be effective for positive initial judgment in instant communication
was more evident in strong rather than weak relationship. As a rela­ (Dipboye et al., 1977), but that it is ineffective to building trust in long-
tionship develops, trust results in greater follower loyalty. term interactions between influencers and followers on social media.
The effect of expertise also varied by relationship strength, even This finding supports the idea that physical attractiveness can be
though the statistical difference was marginal. The impact of expertise conditionally effective (Maddux & Rogers, 1980) and should be inter­
on trust was significant in strong relationships, but insignificant in weak preted in the context of communication (Baker & Churchill, 1977; Till &
relationships. It can be inferred that greater depth and frequency of Busler, 2000). Furthermore, this study found that the effect of expertise
interaction are required for expertise to build trust. In other words, more can be restricted by relationship strength. While the literature on source
time and interactions are needed for followers to recognize the influ­ credibility model (Giffin, 1967; (McGuire, 1968) has considered the
encer’s expertise and become reliant on the relationship. While the effect of expertise to be substantial, this study found that expertise was
traditional literature on source credibility posited expertise as a solid insignificant for the weak relationship group.
feature of speaker’s credibility (Giffin, 1967; (McGuire, 1968), this Furthermore, this study extends the outcomes of source character­
study suggests that its effect can be contingent on the relationship istic models to loyalty. Previous studies on the source credibility model
strength on social media. This nuanced view is aligned with Lou and examined attitude change and persuasion effectiveness as consequences
Yuan (2019) in which the expertise of the influencer did not lead to trust of source characteristics (Horai et al., 1974; Hovland & Weiss, 1951;
in the branded posting. The difference of this study from Lou and Yuan Maddux & Rogers, 1980). The source attractiveness model also tested
(2019) is that the interpersonal trust in the influencer was examined and the effect of attractiveness on initial judgment (Baker & Churchill, 1977;
found that expertise was impactful in a strong relationship. Dion et al., 1972; Dipboye et al., 1977; Joseph, 1982), opinion change
Furthermore, trust enhanced followers’ acceptance of endorsements, (Chaiken, 1979; Horai et al., 1974), and behavioral influence (Byrne,
regardless of relationship strength. Trust caused followers to perceive 1961; Byrne & Wong, 1962; Leventhal & Perloe, 1962). However, this
the recommended product/service more positively and increased pur­ study suggests that source characteristics can alter followers’ thoughts
chase intent. Therefore, trust in the influencer was powerful enough to and behaviors through relationship development. In so doing, this study
affect the followers’ opinions and behaviors. This demonstrates the presents the possibility that source characteristics models can go beyond

229
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

the celebrity endorsement context and can be extended to influencer- social media platform (e.g., YouTube, Instagram, Facebook, Twitter,
follower relationship development. Pinterest, Tiktok) can be tested in the communication process. More­
This study also suggests practical implications, especially for social over, our study was conducted in the context of U.S. social media users,
media marketers who have been increasing their budget for influencer and thus the results may not be generalizable to other cultures or na­
marketing and exploring approaches for selecting the right influencer as tions. Therefore, comparative cross-national and cross-cultural studies
a partner (Ward, 2018). While influencers create appeal through various could provide an interesting avenue for future research. Finally, this
characteristics, marketing practitioners have struggled with a lack of study concentrated on the role of trust in leveraging the social exchange
information for selecting the right influencer for their brand/product process between influencers and followers. By extending the social ex­
(Eyal, 2018). While social media platforms such as Instagram and change mechanism to persuading followers, other relationship traits can
Facebook offer user information and assist the search for influencers, be examined in future research.
data are limited to demographic information and quantitative analyses
(McDowell, 2019). In this regard, this study provides nuanced criteria to References
assess an influencer’s characteristics such as expertise, authenticity,
physical attractiveness, and homophily to leverage relational trust. Andersen, K., & Clevenger, T., Jr (1963). A summary of experimental research in ethos.
Communications Monographs, 30(2), 59–78.
Specifically concerning attractive characteristics, perceived similar­ Audrezet, A., de Kerviler, G., & Moulard, J. G. (2018). Authenticity under threat: When
ity is significant to building trust while physical attractiveness is not. It social media influencers need to go beyond self-presentation. Journal of Business
explains the recent phenomenon in which ordinary people impact mil­ Research, 117, 557–569.
Bagozzi, R. P., & Yi, Y. (1998). On the evaluation of structure equation models. Journal of
lions of followers with friendly image and approachable lifestyles as the Academy of Marketing Science, 16(1), 76–94.
influencers (Godwin, 2018; O’Leary, 2019). Therefore, marketing Baker, M. J., & Churchill, G. A., Jr (1977). The impact of physically attractive models on
practitioners should interpret the similarity between the influencer and advertising evaluations. Journal of Marketing Research, 14(4), 538–555.
Baker, S., & Martinson, D. L. (2002). Out of the red-light district: Five principles for
followers and incorporate it into their marketing strategy or message to ethically proactive public relations. Public Relations Quarterly, 47(3), 15.
leverage relational trust. Moreover, the result implies that influencer Blau, P. (1964). Exchange and power in social life. New York: Wiley.
marketing is different from the traditional celebrity endorsement where Boerman, S. C., & Van Reijmersdal, E. A. (2020). Disclosing influencer marketing on
YouTube to children: The moderating role of para-social relationship. Frontiers in
physical attractiveness is a crucial quality. Even for beauty products,
Psychology, 10(3042).
those relatable influencers such as Patrick Starrr and Amber Wagner are Boerman, S. C., Willemsen, L. M., & Van Der Aa, E. P. (2017). “This post is sponsored”:
successful though they are not considered as typical beauty icons Effects of sponsorship disclosure on persuasion knowledge and electronic word of
(Gerard, 2019; Capadona, 2019) In contrast to the celebrity advertising mouth in the context of Facebook. Journal of Interactive Marketing, 38, 82–92.
Breves, P. L., Liebers, N., Abt, M., & Kunze, A. (2019). The perceived fit between
that instantly captures viewers’ attention with the external beauty, instagram influencers and the endorsed brand: How influencer–brand fit affects
influencer marketing should devise a way to use physical attractiveness source credibility and persuasive effectiveness. Journal of Advertising Research, 59(4),
as an interpersonal resource for cultivating influencer-consumer, influ­ 440–454.
Brooks, G., & Piskorski, M. (2018). The Trusted Influencer: How They Do It and How
encer-brand, and brand-consumer relationships.Figure 1 Brands Can Benefit. ACR North American Advances.
Interestingly, relationship strength, determined by the frequency of Brown, D., & Hayes, N. (2008). Influencer marketing. Routledge.
visits and time spent alters the role of trust. Authenticity was more Byrne, D. (1961). Interpersonal attraction and attitude similarity. The Journal of
Abnormal and Social Psychology, 62(3), 713.
important for the weak relationship, suggesting significant effect of Byrne, D., & Wong, T. J. (1962). Racial prejudice, interpersonal attraction, and assumed
authenticity in the early stage of an influencer-follower relationship. dissimilarity of attitudes. The Journal of Abnormal and Social Psychology, 65(4), 246.
While authenticity has been emphasized tremendously across the in­ Capadona, G. (2019, Jul, 11). Beauty YouTuber Patrick Starrr: ‘Being an influencer is this
religion of being brand agnostic’. Glossy. Retrieved from https://www.glossy.co/
dustry (Fou, 2021; Launchmetrics, 2019; Rakuten, 2019) and academia podcasts/beautyyoutuber- patrick-starrr-being-an-influencer-is-this-religion-of-
(Audrezet et al., 2018; Pöyry et al., 2019), its specific role has not been being-brand-agnostic.
identified. This study suggests that the perception of genuine intention Chahal, H., & Rani, A. (2017). How trust moderates social media engagement and brand
equity. Journal of Research in Interactive Marketing, 11(3), 312–335.
can be more impactful to followers who have a weak relationship, or
Chaiken, S. (1979). Communicator physical attractiveness and persuasion. Journal of
have not yet developed an intense relationship, with the influencer. This Personality and Social Psychology, 37(8), 1387–1397.
indicates that authenticity is less compelling to those who view the Chu, S. C., & Kim, Y. (2011). Determinants of consumer engagement in electronic word-
influencer’s posts frequently and spend more time, and thus became of-mouth (eWOM) in social networking sites. International Journal of Advertising, 30
(1), 47–75.
familiar with the influencer. Therefore, brand managers and marketing Cook, K. S., & Emerson, R. M. (1978). Power, equity and commitment in exchange
practitioners can leverage authenticity as a social exchange driver for networks. American Sociological Review, 721–739.
influencers with fewer views and less time watched by followers. They Cook, K. S., & Yamagishi, T. (1992). Power in exchange networks: A power-dependence
formulation. Social Networks, 14(3–4), 245–265.
should also keep in mind that authenticity alone may not be sufficient Crosby, L. A., Evans, K. R., & Cowles, D. (1990). Relationship quality in services selling:
for maintaining trust if the influencer is heavily consumed by followers. An interpersonal influence perspective. Journal of Marketing, 54(3), 68–81.
Furthermore, expertise was critical to trust in a strong relationship, not Dabholkar, P. A., & Bagozzi, R. P. (2002). An attitudinal model of technology-based self-
service: moderating effects of consumer traits and situational factors. Journal of the
in a weak relationship. To utilize the influencers’ expertise effectively, it Academy of Marketing Science, 30(3), 184–201.
requires an intense relationship where followers invested sufficient time Dion, K., Berscheid, E., & Walster, E. (1972). What is beautiful is good. Journal of
and frequency of viewing the influencer. Lastly, the impact of trust on Personality and Social Psychology, 24(3), 285.
Dipboye, R. L., Arvey, R. D., & Terpstra, D. E. (1977). Sex and physical attractiveness of
loyalty was stronger in a deeper relationship. As followers experience raters and applicants as determinants of resume evaluations. Journal of Applied
more time with the influencer, their trust enhances their loyalty to the Psychology, 62(3), 288.
influencer. Thus, trust is more impactful to influencers who build in- Djafarova, E., & Rushworth, C. (2017). Exploring the credibility of online celebrities’
Instagram profiles in influencing the purchase decisions of young female users.
depth relationships with followers.
Computers in Human Behavior, 68, 1–7.
Djafarova, E., & Trofimenko, O. (2018). ‘Instafamous’–credibility and self-presentation of
7. Limitations and suggestions for future research micro-celebrities on social media. Information, Communication & Society, 1–15.
Fornell, C., & Larcker, D. F. (1981). Structural equation models with unobservable variables
and measurement error. Algebra and Statistics.
Since our study is cross-sectional in nature, study outcomes were Fou, A. (2021, Feb 3). The disconnect between brand advertisers and consumers has
limited to a defined point in time. Therefore, a longitudinal approach never been wider. Forbes. https://www.forbes.com/sites/augustinefou/2021/02/
03/the-disconnect-between-brand-advertisers-and-consumers-has-never-been-
could be applied in future research to explore the influencer-follower
wider/?sh=683080652ab9.
relationships over time. Specifically, this study identifies the moder­ Edelman. (2019). 2019 Edelman trust barometer. Edelman. https://www.edelman.com/
ating role of relationship strength. Future studies can explore other sites/g/files/aatuss191/files/2019-02/2019_Edelman_Trust_Barometer_Global_
potential moderators that could affect the direction of marketing stra­ Report.pdf.

tegies. For instance, consumer age, influencer topic, and the type of

230
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

Emerson, R. M. (1976). Social exchange theory. Annual Review of Sociology, 2(1), Levin, D. Z., & Cross, R. (2004). The strength of weak ties you can trust: The mediating
335–362. role of trust in effective knowledge transfer. Management Science, 50(11),
Etzkorn, K. (2021, Apr 9). How digital shopping will evolve: Three trends to watch. 1477–1490.
Forbes. https://www.forbes.com/sites/forbestechcouncil/2021/04/09/how-digital- Lou, C., & Yuan, S. (2019). Influencer marketing: How message value and credibility
shopping-will-evolve-three-trends-to-watch/?sh=6d43c2e73773. affect consumer trust of branded content on social media. Journal of Interactive
Eyal, G. (2018, Jul 9). Why influencers fail to disclose commercial relationships and the Advertising, 19(1), 58–73.
brands that enable them. Adweek, https://www.adweek.com/digital/why- Maddux, J. E., & Rogers, R. W. (1980). Effects of source expertness, physical
influencers-fail-to-disclose-commercialrelationships-and-the-brands-that-enable- attractiveness, and supporting arguments on persuasion: A case of brains over
them/. beauty. Journal of Personality and Social Psychology, 39(2), 235–244.
Evans, N. J., Phua, J., Lim, J., & Jun, H. (2017). Disclosing instagram influencer Malinowski, B. (1922). Argonauts of the Western Pacific. New York: Holt, Reinhart and
advertising: The effects of disclosure language on advertising recognition, attitudes, Winston.
and behavioral intent. Journal of Interactive Advertising, 17(2), 138–149. Madden, T. J., Allen, C. T., & Twible, J. L. (1988). Attitude toward the ad: An assessment
Falls, J. (2021, Feb 10). Why consumers care about influencers, and why you should too. of diverse measurement indices under different processing “sets”. Journal of
Entrepreneur. https://www.entrepreneur.com/article/364993. Marketing Research, 242–252.
Gassenheimer, J. B., Houston, F. S., & Davis, J. C. (1998). The role of economic value, Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of
social value, and perceptions of fairness in interorganizational relationship retention organizational trust. Academy of Management Review, 20(3), 709–734.
decisions. Journal of the Academy of Marketing Science, 26(4), 322–337. McDowell, M. (2019, Dec, 18). Instagram wants to be a matchmaker between influencers
Gatignon, H., & Robertson, T. S. (1986). An exchange theory model of interpersonal and brands, Voguebusiness, https://www.voguebusiness.com/companies/
communication. in NA - Advances in Consumer Research, 13, (eds. Richard J. Lutz, instagram-influencer-marketing-facebook-brand-collabs-manager.
Provo). UT: Association for Consumer Research. McNutt, L. (2021, Feb 1). 5 influencer marketing predictions for 2021. PR Daily. https://
Gerard, B. (2019, Jul, 2). The MAC cosmetics love me lipstick collection has the shade www.prdaily.com/5-influencer-marketing-predictions-for-2021/.
you need to make a statement this summer. Elite Daily. Retrieved from https://www. McGuire, W. J. (1968). The Nature of Attitudes and Attitude Change. In G. Lindzey, &
elitedaily.com/p/the-mac-cosmetics-love-me-lipstick-collection-has-the-shade-you- E. Aronson (Eds.), Handbook of Social Psychology (pp. 233–346). Reading, MA:
need-to-make-a-statement-this-summer-18164820. Addison-Wesley.
Gerard, H. B., & Greenbaum, C. W. (1962). Attitudes toward an agent of uncertainty McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in
reduction. Journal of Personality, 30(3), 485–495. social networks. Annual Review of Sociology, 27(1), 415–444.
Giffin, K. (1967). The contribution of studies of source credibility to a theory of Mills, J., & Aronson, E. (1965). Opinion change as a function of the communicator’s
interpersonal trust in the communication process. Psychological Bulletin, 68(2), 104. attractiveness and desire to influence. Journal of Personality and Social Psychology, 1
Gilbert, E., & Karahalios, K. (2009). April). Predicting tie strength with social media. In In (2), 173–177.
Proceedings of the SIGCHI conference on human factors in computing systems (pp. Mittal, V., Huppertz, J. W., & Khare, A. (2008). Customer complaining: The role of tie
211–220). strength and information control. Journal of Retailing, 84(2), 195–204.
Gilly, M. C., Graham, J. L., Wolfinbarger, M. F., & Yale, L. J. (1998). A dyadic study of Molleda, J. C. (2010). Authenticity and the construct’s dimensions in public relations and
interpersonal information search. Journal of the Academy of Marketing Science, 26(2), communication research. Journal of Communication Management, 14(3), 223–236.
83–100. Molm, L. D. (1990). Structure, action, and outcomes: The dynamics of power in social
Godwin, R. (2018, Nov 14). The rise of the nano-influencer: how brands are turning to exchange. American Sociological Review, 427–447.
common people. The Guardian, https://www.theguardian.com/commentisfree/ Molm, L. D., Collett, J. L., & Schaefer, D. R. (2007). Building solidarity through
2018/nov/14/rise-nano-influencer-brands-celebrities-youtube-instagram. generalized exchange: A theory of reciprocity. American Journal of Sociology, 113(1),
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American 205–242.
Sociological Review, 161–178. Molm, L. D., Takahashi, N., & Peterson, G. (2000). Risk and trust in social exchange: An
Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 78 experimental test of a classical proposition. American Journal of Sociology, 105(5),
(6), 1360–1380. 1396–1427.
Grönroos, C. (1990). Service management and marketing: Managing the moments of truth in Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship
service competition. Lexington Books. marketing. Journal of Marketing, 58(3), 20–38.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York, NY: McGraw-Hill.
analysis. Upper Saddle River, NJ: Prentice Hall. O’ Donell, E (2018, Aug 7). Instagram influencers: when a special relationship with fans
Hair, J. F., Black, W., Babin, B., Anderson, R., & Tatham, R. (2006). Multivariate data turns dark. The Conversation. https://theconversation.com/instagram-influencers-
analysis (6th ed.). Upper Saddle River, NJ: Prentice Hall. when-a-special-relationship-with-fans-turns-dark-100543.
Harmon, R. R., & Coney, K. A. (1982). The persuasive effects of source credibility in buy Ohanian, R. (1990). Construction and validation of a scale to measure celebrity
and lease situations. Journal of Marketing Research, 19(2), 255–260. endorsers’ perceived expertise, trustworthiness, and attractiveness. Journal of
Hayes, A. F. (2013). Introduction to mediation: A regression-based approach. Guilford Press. Advertising, 19(3), 39–52.
Homans, G. C. (1961). Social behavior: Its elementary forms. New York: Harcourt, Brace & O’ Leary, R. (2019, April 25). Common People: The Rise of the Nano-Influencer, Sprinklr,
World. Retrieved from https://blog.sprinklr.com/rise-of-nano-influencers/.
Homer, P. M., & Kahle, L. R. (1990). Source expertise, time of source identification, and Parasuraman, A., Zeithaml, V. A., & Malhotra, A. (2005). ES-QUAL: A multiple-item scale
involvement in persuasion: An elaborative processing perspective. Journal of for assessing electronic service quality. Journal of Service Research, 7(3), 213–233.
Advertising, 19(1), 30–39. Phua, J., Jin, S. V., & Kim, J. J. (2017). Gratifications of using Facebook, Twitter,
Horai, J., Naccari, N., & Fatoullah, E. (1974). The effects of expertise and physical Instagram, or Snapchat to follow brands: The moderating effect of social comparison,
attractiveness upon opinion agreement and liking. Sociometry, 601–606. trust, tie strength, and network homophily on brand identification, brand
Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Persuasion and communication. New engagement, brand commitment, and membership intention. Telematics and
Haven, CT: Yale University Press. Informatics, 34(1), 412–424.
Hovland, C. I., & Weiss, W. (1951). The influence of source credibility on communication Pornpitakpan, C. (2004). The persuasiveness of source credibility: A critical review of
effectiveness. Public Opinion Quarterly, 15(4), 635–650. five decades’ evidence. Journal of Applied Social Psychology, 34(2), 243–281.
Ipsos. (2019). Trust in the media. Ipsos. https://www.ipsos.com/sites/default/files/ct/ Pöyry, E., Pelkonen, M., Naumanen, E., & Laaksonen, S. M. (2019). A Call for
news/documents/2019-06/global-advisor-trust-in-media-2019.pdf. authenticity: Audience responses to social media influencer endorsements in
Interactive Advertising Bureau. (2018). Why Publishers are Increasingly Turning to strategic communication. International Journal of Strategic Communication, 13(4),
Influencer Marketing – and What That Means for Marketers. IAB. https://www.iab. 336–351.
com/wp-content/uploads/2018/01/IAB_Influencer_Marketing_for_Publishers_2018- Rakuten (2019). Influencer marketing global survey report, Rakuten Advertising,
01-25.pdf. https://rakutenadvertising.com/en-uk/resources/influencer-marketing-global-
Jin, S. V., Muqaddam, A., & Ryu, E. (2019). Instafamous and social media influencer survey-report/.
marketing. Marketing Intelligence & Planning, 37(5), 567–579. Rapp, A., Beitelspacher, L. S., Grewal, D., & Hughes, D. E. (2013). Understanding social
Jones, J. J., Settle, J. E., Bond, R. M., Fariss, C. J., Marlow, C., & Fowler, J. H. (2013). media effects seller, retailer, and consumer interactions. Journal of the Academy of
Inferring tie strength from online directed behavior. PLoS ONE, 8(1), Article e52168. Marketing Science, 41(5), 547–566.
Joseph, W. B. (1982). The credibility of physically attractive communicators: A review. Robertson, L. (2020, Jan 9). Who will win the influencer marketing game in 2020?, The
Journal of Advertising, 11(3), 15–24. drum, https://www.thedrum.com/opinion/2020/01/09/who-will-win-the-
Kennedy, M. S., Ferrell, L. K., & LeClair, D. T. (2001). Consumers’ trust of salesperson influencer-marketing-game-2020.
and manufacturer: An empirical study. Journal of Business Research, 51(1), 73–86. Schaefer, C., Coyne, J. C., & Lazarus, R. S. (1981). The health-related functions of social
Kenton, S. B. (1989). Speaker credibility in persuasive business communication: A model support. Journal of Behavioral Medicine, 4(4), 381–406.
which explains gender differences 1. The Journal of Business Communication Schouten, A. P., Janssen, L., & Verspaget, M. (2020). Celebrity vs. Influencer
(1973), 26(2), 143-157. endorsements in advertising: The role of identification, credibility, and Product-
Krackhardt, D., & Stern, R. N. (1988). Informal networks and organizational crises: An Endorser fit. International Journal of Advertising, 39(2), 258–281.
experimental simulation. Social Psychology Quarterly, 123–140. Shan, Y., Chen, K. J., & Lin, J. S. (2020). When social media influencers endorse brands:
Launchmetrics (2019) The state of influencer marketing, Launchmetrics, Retrieved from The effects of self-influencer congruence, parasocial identification, and perceived
https://www.launchmetrics.com/landing/influencer-marketing-report-2019. endorser motive. International Journal of Advertising, 39(5), 590–610.
Lazarsfeld, P. F., & Merton, R. K. (1954). Friendship as a social process: A substantive and Shan, Y., & King, K. W. (2015). The effects of interpersonal tie strength and subjective
methodological analysis. Freedom and Control in Modern Society, 18(1), 18–66. norms on consumers’ brand-related eWOM referral intentions. Journal of Interactive
Leventhal, H., & Perloe, S. I. (1962). A relationship between self-esteem and Advertising, 15(1), 16–27.
persuasibility. The Journal of Abnormal and Social Psychology, 64(5), 385–388.

231
D.Y. Kim and H.-Y. Kim Journal of Business Research 134 (2021) 223–232

Simons, H. W., Berkowitz, N. N., & Moyer, R. J. (1970). Similarity, credibility, and Wang, X., Yu, C., & Wei, Y. (2012). Social media peer communication and impacts on
attitude change: A review and a theory. Psychological Bulletin, 73(1), 1–16. purchase intentions: A consumer socialization framework. Journal of Interactive
Sirdeshmukh, D., Singh, J., & Sabol, B. (2002). Consumer trust, value, and loyalty in Marketing, 26(4), 198–208.
relational exchanges. Journal of Marketing, 66(1), 15–37. Ward, T (2018, Dec 8) The Influencer Marketing Trends That Will Explode In 2019.
Skvoretz, J., & Willer, D. (1993). Exclusion and power: A test of four theories of power in Forbes, https://www.forbes.com/sites/tomward/2018/12/18/the-influencer-
exchange networks. American Sociological Review, 801–818. marketing-trends-that-will-explode-in-2019/#4f6860552786.
Snyder, M., & Rothbart, M. (1971). Communicator attractiveness and opinion change. Xiao, M., Wang, R., & Chan-Olmsted, S. (2018). Factors affecting YouTube influencer
Canadian Journal of Behavioural Science, 3(4), 377–387. marketing credibility: A heuristic-systematic model. Journal of Media Business
Sokolova, K., & Kefi, H. (2020). Instagram and YouTube bloggers promote it, why should Studies, 15(3), 188–213.
I buy? How credibility and parasocial interaction influence purchase intentions. Yang, W., & Mattila, A. S. (2012). The role of tie strength on consumer dissatisfaction
Journal of Retailing and Consumer Services, 53. responses. International Journal of Hospitality Management, 31(2), 399–404.
Stanko, M. A., Bonner, J. M., & Calantone, R. J. (2007). Building commitment in v. Wangenheim, F., & Bayón, T. (2004). The effect of word of mouth on services
buyer–seller relationships: A tie strength perspective. Industrial Marketing switching: Measurement and moderating variables. European Journal of
Management, 36(8), 1094–1103. Marketing, 38(9/10), 1173-1185.
Statistica. (2021). Influencer marketing market size worldwide from 2016 to 2021. Zickmund, S. (2007). Deliberation, phronesis, and authenticity: Heidegger’s early
Statistica. https://www.statista.com/statistics/1092819/global-influencer-market- conception of rhetoric. Philosophy & Rhetoric, 40(4), 406–415.
size/. Zeithaml, V. A. (1981). How consumer evaluation processes differ between goods and
Sternthal, B., Phillips, L. W., & Dholakia, R. (1978). The persuasive effect of source services. In Conference: American Marketing Association First Services Marketing
credibility: a situational analysis. Public Opinion Quarterly, 42(3), 285–314. Conference.
Sternthal, B., & Samuel, C. C. (1982). Consumer behavior: An information process
perspective. Englewood Cliffs, NJ: Prentice-Hall.
Do Yuon Kim is an assistant professor in the Department of Consumer and Design Sciences
Taylor, J. W., Houlahan, J. J., & Gabriel, A. C. (1975). The purchase intention question in
at Auburn University. Her research interest is centered on the cognitive and emotional
new product development: A field test. Journal of Marketing, 39(1), 90–92.
processing of consumers when evaluating information and making decisions in digital
Thibaut, J. W., & Kelley, H. H. (1959). The social psychology of groups. New York: Wiley.
environment. In this regard, (a) the mechanism how limited information is being pro­
Till, B. D., & Busler, M. (2000). The match-up hypothesis: Physical attractiveness,
cessed in technology-mediated communication and (b) the relationship development in
expertise, and the role of fit on brand attitude, purchase intent and brand beliefs.
the digital environment and its impact on marketing communication, have been central to
Journal of Advertising, 29(3), 1–13.
her research.
Uzzi, B. (1999). Embeddedness in the making of financial capital: How social relations
and networks benefit firms seeking financing. American Sociological Review, 481–505.
Vannini, P., & Franzese, A. (2008). The authenticity of self: Conceptualization, personal Hye-Young Kim is a professor at the University of Minnesota. Kim’s research interests
experience, and practice. Sociology Compass, 2(5), 1621–1637. revolve around consumer behavior and strategic retail management. Her scholarship in
Venkateshi, C. R. (2020, Jan 14). Influencer Marketing 2020 Infographic, Business 2 these areas has appeared in leading academic journals, including Journal of Business
community, https://www.business2community.com/infographics/influencer- Research and International Journal of Advertising.
marketing-2020-infographic-02275293.

232