#Ad: The Influence of Sponsorship Disclosure on
Responses to Sponsored Instagram Posts and the
Role of Parasocial Interaction
Daisy Veen – 11161876
Master Thesis’s Communication Science Graduate School of Communication Entertainment Communication University of Amsterdam Supervisor: Monique Timmers 2 February 2017
Abstract
The increasing growth of sponsored content created by online influencers on social media has been a cause of concern as it may violate the right of consumers to know whether a message has a commercial purpose. Therefore, the Federal Trade Commission (FTC) introduced new regulations that obligate online influencers to disclose sponsored content. In order to make consumers more aware of these hidden advertisements, a sponsorship disclosure must be effectively communicated. This study compares the effects of three types of sponsorship disclosure on brand attitude, purchase intention and source credibility, and examines whether this effect is mediated by persuasion knowledge and moderated by parasocial interaction. Results of this study’s online experiment show that only an implicit disclosure, and not an explicit disclosure, resulted in a decrease in attitude toward the brand, purchase intention, and source credibility, when compared to the condition without a disclosure. The relation between the implicit condition and the decrease in source credibility can be explained based on the activation of persuasion knowledge. Additionally, the research results show that parasocial interaction does not influence the relation between a disclosure and its responses. These results contribute to current literature on influencer marketing, which focuses on the effects of sponsorship disclosure on the social media platform Instagram. Furthermore, this study is one of the first that incorporates the role of a parasocial relationship within the field of influencer marketing. Based on the research results, this study suggests some important practical
implications for the FTC, marketers and online influencers.
Keywords: influencer marketing, sponsorship disclosure, Instagram, brand attitude,
Introduction
With the increasing growth of social media platforms, online influencers represent a new type of third party endorsers who shape consumers attitudes through their social media channels. Within this form of influencer marketing, brands identify influential individuals who reach a wide audience within a specific online community and offer them sponsorships or monetary rewards in exchange for promoting products in creative ways that takes the mind off regular advertising (Brown & Fiorella; Dimofte, Haugtvedt & Yalch, 2015). However, critics argue that this form of marketing is unethical and deceptive, because these messages are often presented as editorials rather than advertisements (Maheshwari, 2016). It is unclear for the audience that this is a form of advertising and that not all opinions, feelings and experiences of the sender are authentic and real (van Reijmersdal et al., 2016).
As a result, the Federal Trade Commission (FTC) introduced new guidelines to make it clear for consumers whether online influencers have been paid or given something of value in return for promoting a product through their online channels (Maheshwari, 2016). These guidelines state that online influencers are required to disclose a clear and conspicuous description about their relation with the brand in all sponsored posts (FTC, 2015a). However, the FTC has found itself struggling with how these paid endorsements should be
communicated on social media to ensure that they are identified as advertisements (Maheshwari, 2016). In their guidelines they advise that starting with #ad to social media posts could be effective, but other forms of disclosure are also approved (FTC, 2015a). It seems that there is a big grey area within these guidelines in which the distinction between a deceptive and a transparent disclosure is still unclear.
When disclosing information about a sponsored relation, persuasion knowledge in the consumers’ mind can be activated. Consumers are then able to recognize the persuasive intent of the message and as a result are more likely to counter argue and experience negative affect.
Consequently, this may lead to more negative attitudes toward the sponsored brand and lower purchase intentions (van Reijmersdal et al, 2016). In addition, disclosing sponsored content could also negatively affect the credibility of the online influencer (Hwang & Jeong, 2016). When online influencers are perceived as less credible, they will less likely shape audience attitudes, which makes them less appealing to collaborate with.
During this process the relationship between the online influencer and their followers is very important. Online influencers often develop strong relationships with their audiences, where followers begin to view the influencer as a ‘real friend’ (Labrecque, 2014). Due to this parasocial interaction, consumers are more likely to be influenced by a message received from an online influencer who is perceived as a trusted source, rather than a brands’ advertisement (Hwang & Jeong, 2016). The increasing usage of social media is stimulating these parasocial interactions, since these platforms allow its users to follow the influencer’s life through their daily picture and video stories. Recently, Instagram has been the main focus of most
marketers as this is currently the fastest growing mobile photo-sharing platform and it allows companies to reach consumers based on their interests (Miles, 2014).
This study examines whether different types of sponsorship disclosures influences brand attitude, purchase intention and source credibility on Instagram, and investigates whether such an effect is mediated by persuasion knowledge and moderated by parasocial interaction. While recent studies show an increased interest in disclosing sponsored content on online blogs, no previous study has investigated whether similar effects occur on social media, such as Instagram. Additionally, this is one of the first studies that aims to explain the relationship between sponsorship disclosure and brand responses by the role of parasocial interaction. Although, some research has focused on parasocial interaction, there is still very little scientific understanding of the influence of parasocial interaction within the field of influencer marketing.
Insights on the effects of different types of sponsorship disclosure on Instagram could also suggest several practical implications. Firstly, since the FTC is not prescriptive about how a disclosure should be articulated on social media, disclosures appear in many creative ways. This study aims to provide the FTC with a better understanding of the effects of different types of sponsorship disclosure, in order to improve their guidelines regarding the requirements of sponsorship disclosure on social media. Secondly, clear insights into the effects of different types of sponsorship disclosures are also helpful for marketers and
influencers who aim for an effective and persuasive communication of the sponsored content without being deceptive.
Therefore, by conducting an online experiment this study examines the effects of three types of sponsorship disclosure: an explicit condition that will explicitly state the relationship between the influencer and the brand using a hashtag, an implicit condition that will provide a description of the relationship between the online influencer and the brand, and a control condition that will not mention a relationship. The following part of this paper moves on to existing literature to describe the effects of sponsorship disclosure on brand responses and source credibility. Consequently, a detailed description of the conducted experiment is provided. Finally, the study reports the experiment results and reflects on its main findings in the conclusion and discussion.
Theoretical Background
Influencer Marketing
The term influencer marketing is a relatively new form of covert marketing, commonly referred to as ‘native’ or ‘hidden’ advertising. In this type of marketing, brands identify an opinion leader within a specific community and collaborate with this online influencer to avoid both consumer disinterest and scepticism of marketing messages by communicating in
ways that are perceived as non-marketing interactions (Petty & Andrews, 2008). The commercial source and the commercial message are being masked, because consumers are less sceptical toward messages received from independent sources than from marketers (Petty & Andrews, 2008; Hwang & Jeong, 2016). Because of this trusted relationship between the online influencer and consumers, brands are continuously looking for the right influencers who are willing to promote products and actively create content with the intention to generate awareness and positive attitudes toward the product or brand (Dimofte et al., 2016).
The promotion of products by online influencers is often referred to as sponsored content. In this study, sponsored content is defined as “the purposeful integration of brands or branded persuasive messages into editorial media content in exchange for compensation from a sponsor” (Boerman & van Reijmersdal, 2016). Online influencers often receive direct monetary compensation or indirect compensation such as free products in return for promoting products and to encourage positive electronic word-of-mouth (Lu, Chang & Chang, 2014). Examples of sponsored content are advertorials in magazines, product placement in television or movies, or product reviews on blogs. Product placement has also been adapted to social media, where a brand name, product, package or logo is carefully inserted in an influencer’s post (Kulin & Blomgren, 2016; Williams, Petrosky, Hernandeze & Page, 2011). Both brands and online influencers aim to make the product placement as integrated and genuine as possible, in order to remain credible and prevent sceptical attitudes.
There are potential ethical and legal implications associated with sponsored content and influencer marketing. With today’s blurred boundaries between editorial and commercial content, consumers may not recognize the sponsored content as persuasive, and thus may not carefully scrutinize the message and activate cognitive resistance strategies (Bhatnagar, Aksoy, & Malkoc, 2004; Nebenzahl & Jaffe, 1998). Sponsorship disclosure in television, movies and on radio is already obligatory in the European Union and in the United States
(Cain, 2011). Recently, the Federal Trade Commission in the United States introduced new guidelines to handle sponsored content on social media and online blogs, while such regulations in Europe are being developed. The aim of these guidelines is that sponsorship disclosure should “explicitly inform the audience when commercial content is integrated into editorial content to guarantee fair communication and avoid persuasion without audience awareness” (Boerman, van Reijmersdal & Neijens, 2012).
Since the increasing usage of these hidden marketing techniques and the development of guidelines regarding sponsorship regulations in the media, there has been a considerable amount of research done about the effects of sponsorship disclosure on consumer responses. These studies focused on different types of media, different types of disclosures, and
measured different mediating and moderating variables.
Effects of Sponsorship Disclosure on Processing
Earlier research shows that disclosing sponsored content can have important effects on the way consumers process sponsored messages. The main goal of a disclosure is to help consumers recognize the message as advertising and therefore, it functions as a warning to make consumers aware of the persuasive attempt of the message (Boerman & van
Reijmersdal, 2016). When consumers realize that the message is trying to persuade them, cognitive resistance strategies may be activated. One of these strategies is the activation of persuasion knowledge. According to the persuasion knowledge model, consumers have some knowledge regarding persuasive messages and they can activate this knowledge when they encounter persuasive attempts from advertisements (Friestad & Wright, 1994). This
knowledge is formed by experience and develops over time (Boermen, van Reijmersdal & Neijens, 2014).
When consumers recognize the commercial purpose of the message and activate their persuasion knowledge, they might respond different to the message than when this knowledge is not activated (van Reijmersdal et al., 2016). The persuasion knowledge model explains that people can use their knowledge about persuasion and decide to either be persuaded or to resist the persuasion attempt (Friestad & Wright, 1994). According to the reactance theory, people want to have a freedom of choice and do not want to be manipulated in this process (Brehm & Brehm, 1981). Therefore, when people recognize a message as advertisement, people tend to resist the persuasive attempts, because they might feel deceived (Petty & Cacioppo, 1977). When consumers realize that a message has a persuasive goal, they might experience this as a threat to their freedom of choice, which results in activating cognitive and affective resistance strategies to cope with the unwanted persuasive attempt (van Reijmersdal et al., 2016).
Numerous studies have demonstrated the effect of disclosing sponsored content on the activation of persuasion knowledge. Several studies by Boerman et al., (2012; 2014; 2015) showed that viewers recognized commercial product placement in a television program more often when they recalled seeing a disclosure, compared to viewers who did not recall a disclosure. In advergames, van Reijmerdal, Lammers, Rozendaal and Buijzen (2015) found a similar effect. They showed that users who viewed a disclosure were more aware of the game being advertised, compared to users who played the game that did not include a disclosure. Such effects were also found in online blogs by Carr and Hayes (2014) and van Reijmersdal et al., (2016), where participants noticed more often that a blog was influenced by a brand in the disclosure conditions, compared to in the condition that did not show a disclosure.
Based on the persuasion knowledge model and earlier research, this study proposes that when viewers are exposed to a sponsored Instagram post with a disclosure, their
the no disclosure condition; the explicit and the implicit disclosure conditions will both activate persuasion knowledge.
H1. Disclosing sponsored content on Instagram will activate persuasion knowledge.
Compared to the no disclosure condition, the explicit and the implicit disclosure conditions will both activate persuasion knowledge.
Effects of Sponsorship Disclosure on Brand Responses
As a result of the activation of persuasion knowledge, consumers are more likely to criticize the content (Boerman & van Rijemersdal, 2016). The awareness of the persuasive message and the feeling of deception will encourage more systematic and biased processing (Janssen, Fennis & Pruyn, 2010). Earlier studies show that the activation of persuasion knowledge can lead to different types of resistance, such as counter arguing, suspicion, and less favourable attitudes toward the message (Boerman, van Reijmersdal & Neijens, 2014). This indicates that when consumers recognize a message as advertising after seeing a disclosure, they activate knowledge about the persuasive attempt in their mind. As a result, this knowledge could trigger cognitive and affective resistance toward the sponsored brand (van Reijmersdal et al., 2016). This could lead to more negative attitudes toward the brand and lower purchase intentions (van Reijmersdal et al., 2016).
In line with this reasoning, earlier research found negative effects of disclosing sponsored content on brand attitudes on television (Boerman, van Reijmersdal & Neijens, 2012; 2015), and on radio (Wei, Fischer & Main, 2008). Recent studies have focussed on the online world and the effects of sponsorship disclosure on blogs and online reviews. Van Reijmersdal et al., (2016) found that disclosing a brand on a blog resulted in the activation of persuasion knowledge and higher cognitive and affective resistance, which in turn was
associated with more negative brand attitudes and lower purchase intention. Campbell, Mohr and Verlegh (2012) showed that product placement in blogs with no disclosure resulted in positive attitudes toward the brand. In the disclosure conditions they focused on when the disclosure was presented to its readers. Brand attitude was only corrected when the disclosure was presented after reading the blog, but not when presented before the placement.
While several studies found significant effects of sponsorship disclosure on attitudes toward the brand, others did not (Dekker & van Reijmersdal, 2013; van Reijmersdal et al., 2015; Hwang & Jeong, 2016). However, the results of these studies are inconclusive and can either be explained based on the absence of the persuasion knowledge model or the different conditions of sponsorship disclosure that were being used. When resistance of a persuasive message occurs as a result of the activation of persuasion knowledge, it is unlikely that attitudes become more favourable (Tormala & Petty, 2002). This study, therefore, proposes that Instagram posts with a sponsorship disclosure will induce more negative attitudes toward the advertised brand than posts with no sponsorship disclosure, and will examine whether this relation is mediated by persuasion knowledge.
H2a. Sponsorship disclosure has a negative effect on brand attitude. Compared to the no
disclosure condition, the explicit and the implicit disclosure condition will both induce negative brand attitudes.
H2b. The effect of sponsorship disclosure on brand attitude is mediated by persuasion
knowledge. This mediation will occur in the explicit and the implicit disclosure condition, and not in the no disclosure condition.
The process of activating persuasion knowledge may not only influence cognitive and affective responses, but also attitudinal responses such as behavioural intentions (Boerman & van Reijmersdal, 2016). Personal experiences of online influencers shown on Instagram accounts are useful cues for consumers in evaluating products for purchase decisions (Lu, Chang, Chang, 2014). According to the theory or reasoned action and the theory of planned behaviour, consumers’ attitudes will directly affect their intention to perform a specific behaviour (Fishbein & Ajzen, 1975). Earlier research supports this finding and shows that consumers’ attitude toward a brand can indeed predict their intention to purchase the product (Homer, 1990; MacKenzie, Lutz, & Belch, 1986; Spears & Singh, 2004).
Liljander, Gummerus and Söderlund (2015) compared sponsorship disclosure in blogs to no disclosure and concluded that blog posts that reveal the sponsorship of the
recommended brand have a negative influence on young consumers’ purchase intentions. Based on this knowledge, this study proposes that if consumers have positive attitudes toward the brand, they will have a greater intention to purchase the product. Moreover, compared to the no disclosure condition, the explicit and the implicit condition will both induce lower purchase intention. The following hypothesis is proposed:
H3. Attitudes toward a brand positively influences purchase intention. Compared to the no
disclosure condition, the explicit and the implicit condition will both induce lower purchase intentions.
Effects of Sponsorship Disclosure on Source Credibility
Sponsorship disclosure will not only influence responses toward the brand, it could also have an effect on the credibility of the source. The term source credibility can be understood as the expertise and the trustworthiness of the source (Sternthal, Dholakia & Leavitt, 1978). When
an online influencer is perceived as credible, consumers will be more likely to accept that the arguments provided by the influencer are honest, than when the perceived trustworthiness of the influencer is low (Chu & Kamal, 2008). In addition, marketers collaborate with online influencers to promote their products, because consumers are more likely to positively engage in a message received from a trusted source, rather than from a company’s own advertisement (Hwang & Jeong, 2016). As online influencers increasingly collaborate with more brands and share more sponsored content, their messages are losing the authenticity that made influencer marketing successful. Especially due to the requirements of the new guidelines to disclose every sponsored post, consumers will realize that a lot of the content has commercial purposes and they will stop considering their recommendations as genuine and honest.
In addition, this relation can also be explained based on the activation of persuasion knowledge. Persuasion knowledge may attract consumers’ attention to the persuasive tactics that are used in the advertisement. For example, consumers will realize that heuristic cues, such as an endorser promoting the products, are used in an attempt to persuade them. After activating their persuasion knowledge, consumers will not only critically process the message, but also the source they receive the message from. This may result into resistance toward the persuasive intent of the message and a downgrade of the credibility of the influencer (Lee & Koo, 2012). Fransen and Fennis (2014) argue that these processes may subsequently affect the perceptions of the reliability and effectiveness of the message and the source and might lead to source derogation.
Earlier research also shows a negative effect of sponsorship disclosure on source credibility. Hwang and Jeong (2016) concluded that a simple disclosure on blogs resulted in more negative source credibility than when the sponsorship was not disclosed. These negative attitudes disappeared when it was stated that the content was based on honest opinions.
newspaper declaring that a blogger was sponsored, resulted in negative attitudes toward the blog and caused a decrease in credibility, compared to the group who did not have a
revelation of the sponsorship. Finally, Carr and Hayes (2014) argued that an online product reviewer was perceived as the most credible in the control and explicit condition, less credible when disclosing the absence of third-party influence (impartiality), and least credible when the brand was implicitly mentioned in the disclosure.
Thus, sponsorship disclosure functions as a warning to alert consumers of the persuasive intent of the message, which results in lower perceived source credibility. Therefore, this study proposes that sponsorship disclosure will result in lower perceived source credibility, and examines the mediating role of persuasion knowledge. The following hypotheses have been formed:
H4a. Sponsorship disclosure will influence source credibility. Compared to the no disclosure
condition, the explicit and implicit condition will result in lower perceived credibility.
H4b. The effect of sponsorship disclosure on source credibility is mediated by persuasion
knowledge. Compared to the no disclosure condition, this mediation will occur in the explicit and implicit disclosure condition.
The Moderating Role of Parasocial Interaction
A unique characteristic of influencer marketing is that users establish deep online
relationships with people they have not met in the physical world. This is called parasocial interaction (PSI), which is the “illusion of a face-to-face relationship with a media performer” (Horton & Wohl, 1956). The parasocial interaction theory emerged from a psychological topic and was later on developed in the field of communication science to explain consumer
relationships with mass media, such as radio and television (Giles, 2002; Horton & Wohl, 1956). The parasocial relationship is in many ways similar to interpersonal relations, because parasocial relationships are also voluntarily, provide companionship, and are predicted by social attraction (Perse & Rubin, 1989). Overall, parasocial relations resemble social relationships, and although they may consist of a weaker bond and may be less intense than close social relationships, for many media users, relationships with media performers and other celebrities take an important part in their social life (Ballantine & Martin, 2005; Skumanich & Kintsfather, 1998). Mass mediated communication that closely resembles interpersonal communication, may increase the credibility and the persuasiveness of the message (Beninger, 1987).
Compared to traditional PSI environments (e.g., radio and television), where people believed they are engaged in a direct two-way conversation, creating the feeling that the media performer is talking directly to them (Rubin, Perse & Powell, 1985), online media, especially blogs and social media platforms, may actually make this direct conversation possible, because of new technological improvements of the Internet (Labreque, 2014). This direct communication may even increase the online engagement and strengthen the parasocial relationship. Online media gives users the opportunity to gain insights into the daily life of influencers through their personal pictures and video stories (Colliander & Dahlén, 2011). PSI is developed when users often engage with content from the influencer, which results in a strong relationship after numerous online encounters (Auter, 1992). If this engagement continues, users can start considering the social media performer as a personal friend and he or she becomes a part of their everyday life (Ballentine & Martin, 2005; Labrecque, 2014).
Over time, if this relationship intensifies, online engagement may increase in order to retain the friendship (Rubin, Perse & Powell, 1985). To show their loyalty to the media performer, users may start sending messages and purchasing products recommended by the
performer (Ballantine & Martin, 2005; Horton & Wohl, 1956). With this increasing
commitment it is likely to expect that users who have a strong parasocial relationship with an online performer are more likely to support their opinions about a certain brand or product. Earlier research on the influence of a parasocial relationship on brand responses also
concluded that users who had a strong parasocial relationship with a blogger and consider the blogger as a friend, show higher attitudes toward the brand and higher purchase intentions (Colliander & Dahlén, 2011).
Based on these findings, this study suggests that parasocial interaction has a
moderating role in the relation between sponsorship disclosure, brand attitude and purchase intention, and in the relation between sponsorship disclosure and source credibility. When consumers have a strong parasocial relationship with an online influencer, they are more likely to accept recommendations about a specific product or brand, and are more likely to consider the influencer as trustworthy, even though they know the influencer has been sponsored. The following hypotheses are created:
H5a. The negative impact of sponsorship disclosure on brand attitude and purchase intention
is positively moderated by parasocial interaction.
H5b. The negative impact of sponsorship disclosure on source credibility is positively
moderated by parasocial interaction.
This study examines the model represented in Figure 1. This model proposes that a disclosure of the sponsored content negatively influences brand attitude, source credibility and purchase intention. In addition, this study expects that this relation is mediated by the activation of persuasion knowledge and moderated by parasocial interaction.
Figure 1. Proposed conceptual model explaining the relations between the variables.
Methods
Sample and Procedure
A total of 172 respondents participated in the online experiment. This study only recruited women between 18 and 28 years old, because influencer marketing is largely focused on a female audience. Moreover, this study chose for this age group, because emerging adults spend more time online than doing any other activity (Coyne, Padilla-Walker & Howard, 2013). After excluding three men, one responded and who did not agree the consent form, and 38 respondents who did not finish the survey, 130 (Mage = 22.18, SD = 2.53) female
emerging adults between 18 and 28 years old were used for analysis. Participants were recruited via Facebook and Instagram, and were asked to fill in a short online survey about Instagram that would take around 6-8 minutes to complete. 89% of the participants have an Instagram account, and 72% uses Instagram daily or multiple times a day.
Participants were randomly assigned to one of the three conditions; control condition (n = 41), explicit condition (n = 46), implicit condition (n = 43). After giving informed consent, the questionnaire started with control variables (demographics and Instagram use),
Sponsorship Disclosure Persuasion Knowledge Source Credibility Brand attitude Purchase Intention Parasocial Interaction
and then showed participants one of the three conditions. Next, the questionnaire asked about source credibility, participants’ persuasion knowledge, parasocial interaction, and brand responses (i.e. brand attitudes and purchase intentions). The questionnaire ended with control variables (product interest and advertisement interest). By measuring source credibility before persuasion knowledge, this study made sure that attitudes toward the credibility of the source were the result of the stimulus and not a result of the questions about advertising.
Stimuli
Instead of using a hypothetical influencer, this study used the online influencer Yara Michels to enhance ecological validity and measure parasocial interaction, since this is developed over time. Yara Michels is one of the biggest Dutch online influencer with around 135.000
followers on Instagram, 13.000 subscribers on her YouTube channel and who has her own blog called Chapter Friday. In the Instagram post that was used for in this study, Yara Michels was clearly visible and she casually showed a watch from the brand Calvin Klein. This picture was chosen because it could be expected that both the influencer and the brand were familiar to a large audience.
The captions were manipulated in this study to measure the influence of sponsorship disclosure on Instagram (see Appendix 1). For the explicit condition, a disclosure stating “#ad In love with my new @calvinklein watch” was inserted in the caption. For the implicit
condition, the disclosure “Thanks @calvinklein for the gift of this lovely new watch” was inserted. Finally, the control condition did not disclose that the content was sponsored: “In love with my new watch #calvinklein”. The disclosure is based on the new regulations by the Federal Trade Commission and on earlier research on disclosure effects (FTC, 2015). Within these guidelines the FTC states that the disclosure should be clear and conspicuous and that #ad is likely to be effective. That is why this disclosure is chosen for the explicit condition
because this type of disclosure explicitly states that the message is advertising. The implicit condition thanks the brand and explains the way the product was obtained, specifically that the product was received as a gift. However, the consumer has to make the connection that this is a form of advertisement. Finally, the control condition does not include any
commercial relation between the brand and the influencer. The content is then presented as if the online influencer bought the product herself. In order to enhance ecological validity three Instagram pictures were shown to create the feeling of scrolling through an actual Instagram feed. In all conditions the picture of Yara Michels was shown the last.
Measures
Persuasion knowledge. To identify to what extent participants recognized the sponsored
Instagram post as advertising, the activation of the persuasion knowledge was measured using four items (see Appendix 2). Participants were asked to indicate on a 7-pount scale
(1=definitely no, 7=definitely yes) to what extent they agreed with statements such as “the Instagram post of [ONLINE INFLUENCERS] is advertising” (Boerman et al., 2012; van Reijmersdal, Neijens & Smit, 2010; van Reijmersdal et al., 2016). Factor analysis revealed that the four items loaded on one factor (Eigenvalue = 2.93; Cronbach’s alpha = .87). The variance explained by this factor was 73.24%.
Brand attitude. Participants’ attitude toward the brand was measured to indicate the
persuasive effect of the sponsored content. The dependent variable was measured using a six item (see Appendix 2) 7-point scale (1=strongly disagree, 7=strongly agree) in which participants could indicate what their opinions were about the brand (Gordon & Bruner, 2009). Factor analysis revealed that the items loaded on one factor (Eigenvalue = 3.99; Cronbach’s alpha = .9). The variance explained by this factor was 66.55%.
Purchase intention. Participants’ intention to purchase a product from the brand was
measured using a four item (see Appendix 2) 7-point scale (1=strongly disagree, 7=strongly
agree) in which participants could indicate to what extent they agreed with statements such as
“I will buy [BRAND]” (Spears & Singh, 2004). Factor analysis revealed that the items loaded on one factor (Eigenvalue = 3,48; Cronbach’s alpha = .95). The variance explained by this factor was 86.99%.
Source credibility. To identify to what extent sponsorship disclosure influences the
credibility of the source, source credibility was measured by asking participants whether they on a 7-point scale agreed or disagreed (1=strongly disagree, 7=strongly agree) with five items (see Appendix 2) measuring source credibility (Ohanian, 1990). Factor analysis
revealed that the items loaded on two factors. Since only ‘biased’ loaded on a different factor, this item was deleted, because non-native speakers might have misunderstood this term. All other items loaded on one factor (Eigenvalue = 2.59; Cronbach’s alpha = .82). The variance explained by this factor was 64.7%.
Parasocial interaction. Parasocial interaction indicated to what extent participants
perceived the online influencer as a friend. The first three questions were asked to indicate whether participants were familiar with the online influencer (1=none at all, 5=a great deal), whether participants currently follow the online influencer online (1=no, 2=yes), and how often they view online content of the online influencer (1=never, 7=daily). Secondly, a seven items measure (see Appendix 2) asked participants to indicate on a 7-point scale (1=strongly disagree, 7=strongly agree) to what extent they agreed with the statements such as
“[ONLINE INFLUENCER] makes me feel comfortable, as if I were with a friend” (Rubin, Perse & Powell, 1985). Factor analysis revealed that the items loaded on one factor
(Eigenvalue = 4.05; Cronbach’s alpha = .87). The variance explained by this factor was 57.78%.
Control variables. Furthermore, participants were asked whether they owned a Calvin
Klein watch (97% said no). Based on a scale by van Reijmersdal, Neijens, and Smit (2007), product interest was measured by asking participant to what extent they agreed with three items (1=strongly disagree, 7= strongly agree), such as “I like buying watches” (see Appendix 2) (Eigenvalue= 2.48; Explained variance= 82.64%; Cronbach’s alpha= .89, M= 3.66, SD= 1.68). Moreover, to indicate whether the advertisement corresponded to
participants’ interest, advertisement interest was measured by asking to what extent on a 7-point scale (1=strongly disagree, 7= strongly agree) what they thought of the advertising (see Appendix 2). (Eigenvalue= 2.3; Explained variance= 57.54%; Cronbach’s alpha= .73, M= 3.97, SD= 1.02).
Results
Randomization
ANOVA an chi-square analyses showed there were no differences between the experimental group with respect to whether the participant had an Instagram account, χ2 (2) = .39, p =.824,
frequency of Instagram use F(2, 127) = 1.27, p =.285, familiarity of the influencer, F(2,127) = 1.95, p = .146, product interest F(2, 127) = .09, p =.917, and advertisement interest F(2, 127) = .32 , p =.728.
Analyses
Table 1 compares the mean and standard deviation of the three conditions on the dependent variables. The first hypothesis predicted that compared to the no disclosure condition, the explicit and the implicit condition would both activate persuasion knowledge. To assess hypothesis 1, a one-way ANOVA compared the control condition (n = 41, M = 5.51, SD =
Table 1. Descriptive statistics of the three conditions on dependent variables.
1.04), to the explicit condition (n = 46, M = 5.71, SD = .99), and the implicit condition (n = 43, M = 6.03, SD = .92). The difference was not significant F(2,127) = 3.04, p = .052, indicating that there was no statistical difference in persuasion knowledge between the three conditions. Since equal variances could be assumed, a LSD post-hoc test revealed that the difference between the control condition and the implicit condition was significant (Mdifference
= -.52, p = .016). There was no significant difference between the control group and the explicit group (p = .345), and between the explicit and the implicit group (p = .125). These results show that hypothesis 1 is partially supported.
The second hypothesis predicted that compared to the control condition, both the explicit and implicit condition would induce negative brand attitudes (H2a), and expected that this relation was mediated by persuasion knowledge (H2b). To assess hypothesis 2a, a one-way ANOVA compared the control condition (n = 41, M = 5.2, SD = .77), to the explicit condition (n = 46, M = 4.95, SD = .9), and the implicit condition (n = 43, M = 4.79, SD = .91). The difference was not significant F(2,127) = 2.29, p = .106, indicating that there was no statistical difference in brand attitudes between the three condition. Since equal variances could be assumed, a LSD post-hoc test revealed that the difference between the control condition and the implicit condition was significant (Mdifference = -.40, p = .036). There was no
significant difference between the control group and the explicit group (p = .182), and between the explicit and the implicit group (p = .412). These results show that hypothesis 2a is partially supported, and that only compared to the implicit condition, the control condition resulted in higher attitudes towards the brand.
In testing H2b, that persuasion knowledge mediated the attitude toward the blog, this study used Baron and Kenny’s (1986) frequently used procedure for testing mediation.
Dummy variables were created for both the explicit and the implicit condition with the control condition as the reference category in order to test the effects of each disclosure. In brief, Baron and Kenny’s method (1986) comprises of a series of regression analyses. First, the dependent variable sponsorship disclosure is regressed on the independent variable (Step 1); subsequently the dependent variable sponsorship disclosure is regressed on the potential mediator persuasion knowledge (Step 2); the potential mediator persuasion knowledge is regressed on the dependent variable (Step 3); and lastly, the dependent variable is regressed on both the independent variable sponsorship disclosure and the potential mediator persuasion knowledge. The results of these regression analyses on the dependent variable brand attitude are presented in Table 2 and Appendix 3. The tests showed no significant results at all stages for the explicit condition, Sobel’s z = .39, p = .700. The implicit condition only showed a significant effect in Step 1 and Step 2, stating that sponsorship disclosure significantly predicts brand attitude and persuasion knowledge. There was no interaction effect of the mediator persuasion knowledge, Sobel’s z = 1.10, p = .271. Thus, hypothesis 2b can be rejected.
The third hypothesis predicted that brand attitude influences purchase intention, and that compared to the no disclosure condition, the explicit and the implicit condition both induce lower purchase intentions. First a regression analysis showed that brand attitude has a
Table 2. Mediation analyses of differences in brand attitude and source credibility between
conditions.
Note. Control condition as reference category.
was conducted to compare the control condition (n = 41, M = 4.03, SD = 1.70), to the explicit condition (n = 46, M = 3.55, SD = 1.61), and the implicit condition (n = 43, M = 2.99, SD = 1.53). The differences between the groups were significant F(2, 127), p = .015. Since equal variances could be assumed, a LSD post-hoc test revealed that the difference between the control condition and the implicit condition was significant (Mdifference = -.104, p = .004).
There was no significant difference between the control group and the explicit group (p = .171), and between the explicit and the implicit group (p = .103). These results show that compared to the implicit condition, purchase intentions in the control condition were significantly higher, thus hypothesis 3 is partially supported.
The fourth hypothesis predicted that compared to the control condition, both the explicit and implicit condition would induce lower source credibility (H4a), and expected that this relation was mediated by persuasion knowledge (H4b). To test hypothesis 4a, a one-way ANOVA compared the control condition (n = 41, M = 3.80, SD = .92), to the explicit
condition (n = 46, M = 3.81, SD = 1.00), and the implicit condition (n = 43, M = 3.78, SD = 1.15). The differences between the groups were not significant F(2,127) = .01 , p = .989). Since equal variances could be assumed, a LSD post-hoc test revealed that there was no difference between the conditions; therefore hypothesis 4a can be rejected.
In testing hypothesis 4b, mediation analyses were conducted to examine the mediating role of persuasion knowledge between sponsorship disclosure and source credibility (Baron & Kenny, 1986). These analyses (see Table 2 and Appendix 3) showed that there was no
mediation of persuasion knowledge between the explicit condition and source credibility, Sobel’s z = -.88, p = .376. In addition, the effect of an implicit disclosure on source credibility showed to be fully mediated by the activation of persuasion knowledge. This finding was supported by the Sobel test, z = -1.99, p = .046. These results show that hypothesis 4b was partially supported, and that there was only a mediating effect of persuasion knowledge on source credibility in the implicit condition, and not in the explicit condition, with reference to the control condition.
Hypothesis 5a an 5b expected that parasocial interaction moderate the relation between sponsorship disclosure, brand attitude, purchase intention and source credibility. 78,5% of the participants were omitted from questions about parasocial interaction, since they responded that they were not familiar with the online influencer and that familiarity is one of the requirements for a parasocial relationship. Only 21.5% of the participants were familiar with the online influencer, 8,5% followed her on social media, and 2,3% viewed daily content of the online influencer. A moderation analysis was conducted within ANOVA, and found
that the moderating effect of parasocial interaction on brand attitude was not significant, F(2) = .09, p = .915. No moderating interaction effect of parasocial interaction was found in the relation between sponsorship disclosure and source credibility, F(2) = .51, p = .607. However, when looking at the graphs in Figure 2, it shows that in all conditions, people with a high parasocial relationship scored higher on brand attitude and source credibility, than people with a low parasocial relationship. Since the moderation was not significant, hypothesis 5a and 5b can be rejected.
Figure 2. Parasocial interaction on brand attitude and source credibility for three conditions.
Conclusion
The present study examined the effects of different types of sponsorship disclosure on brand attitude, purchase intention, and source credibility, and investigated whether this effect was mediated by persuasion knowledge and moderated by parasocial interaction. First, this study shows that an implicit disclosure could induce higher activation of persuasion knowledge compared to no sponsorship disclosure. Surprisingly, compared to no sponsorship disclosure, an explicit disclosure does not induce higher activation of persuasion knowledge. This finding
is partially in contrast to the persuasion knowledge model, which states that a sponsorship disclosure helps consumers recognize sponsored content as advertising which results in the activation of knowledge in consumers’ minds (Friestad & Wright, 1994). This study shows that not all types of disclosure activate persuasion knowledge on Instagram.
Second, regarding the effects of sponsorship disclosure on brand responses, this study showed that an implicit disclosure could induce more negative attitudes toward the brand and lower purchase intentions compared to no sponsorship disclosure. When sponsored content was explicitly disclosed, brand attitudes and purchase intention did not differ significantly from a condition in which a disclosure was not present. Additionally, results showed that the relation between sponsorship disclosure and brand attitude was not mediated by the activation of persuasion knowledge. This is in contrast to the existing literature stating that when
persuasion knowledge is activated, consumers may critically process the persuasive attempt, which can result in resistance toward the brand that is being promoted (Boerman et al., 2014). This resistance could lead to more negative attitudes and lower purchase intention (van Reijmersdal et al., 2016). Furthermore, results showed that brand attitudes were positively related to purchase intention. This finding is in line with the theory of reasoned action and the theory of planned behaviour, stating that consumers’ attitude toward a brand predicts their intention to purchase a product (Fishbein & Ajzen, 1975).
Third, this study also examined the effects of sponsorship disclosure on source
credibility. Results showed that an implicit disclosure could result in lower source credibility, when compared to no sponsorship disclosure, and that this relation was fully mediated by persuasion knowledge. An explicit disclosure did not show any significant difference in source credibility, when compared to no sponsorship disclosure. The relation between the implicit disclosure and source credibility can be explained by the activation of persuasion knowledge. This finding is in line with earlier research showing that a disclosure activates
persuasion knowledge, which may attract consumers’ attention to the persuasive tactics that have been used in the persuasive attempt. Consumers could then lose trust in the source because of their attempt to deceive the consumer, which consequently could result in source derogation (Hwang & Jeong, 2016). The fact that no similar effect occurred when comparing an explicit disclosure to no sponsorship disclosure, is in contrast to existing literature.
Finally, this study also examined the moderating role of parasocial interaction. No significant results were found that parasocial interaction moderates the relation between sponsorship disclosure, brand attitude, purchase intention, and source credibility. This finding is in contrast to earlier research that showed that participants with a strong parasocial
relationship with a media performer showed more positive attitudes toward the brand and higher purchase intentions, compared to participants with a weak relation with the media performer (Colliander & Dahlén, 2011). When looking at the graphs it can be noticed that in all three conditions participants with a strong parasocial relation showed more positive
attitudes toward the brand and scored higher on source credibility, than participants who had a weak parasocial relation with the influencer. However, because this relation was not
significant, this data must be interpreted with caution.
Discussion
The general aim of this study was to increase a better understanding of the effects of different types of sponsorship disclosure on brand attitude, purchase intention and source credibility, and to find out how these effects are explained by the activation of persuasion knowledge as a resistance strategy. Moreover, to contribute to the current gap in the literature, this study also aimed to explain the concept of parasocial interaction in the field of influencer marketing. The assumptions based on theories and existing literature, were tested in an online experiment using three conditions. In doing so, this study found some interesting results regarding the
effects of different types of sponsorship disclosure on the social media platform Instagram. However, this study also showed some inconsistent findings that can either be explained by the research design or by earlier research. Based on these findings, recommendation for future research and theoretical and practical implications can be suggested.
This study showed that in comparison to no sponsorship disclosure, only an implicit sponsorship disclosure, and not a an explicit disclosure, had a positive effect on the activation of persuasion knowledge, and a negative effect on attitudes toward the brand, purchase intention and source credibility. This rather contradictory result may be due to the fact that participants were either not familiar with the hashtag in the explicit condition, or that the hashtag did not attract participants’ attention. The hashtag that was used in the explicit condition in this study was based on the FTC guidelines (FTC, 2015a). However, it might be possible that another hashtag would lead to different results. This shows that familiarity and recognition of the disclosure is a very important step within the activation of persuasion knowledge. This was also shown by Tessitore and Geuens (2013) who found that a product placement logo only resulted in activating resistance strategies, when the meaning of the label was explained in combination of a verbal label and information training. Such training helps to increase recognition and comprehension of the symbol as a disclosure.
With regard to the mediating role of persuasion knowledge, this study found some inconsistent results. While persuasion knowledge did not mediate the relation between sponsorship disclosure and brand attitude, a full mediation was found in the relation between implicit sponsorship disclosure and source credibility. It is difficult to explain this result, but it might be related to the perceived appropriateness of the tactics that have been used in the persuasive attempt (Wei et al., 2008). When consumers feel that the online influencer used an inappropriate tactic, they might feel deceived and will translate these negative feelings only to the source and not to the brand. Another possible explanation might be that when consumers
recognize sponsored content as advertising, they only remember the source, but do not pay further attention to the content of the message.
Finally, the relation between sponsorship disclosure, attitudes toward the brand, purchase intention and source credibility was, in all three conditions, not moderated by parasocial interaction. A possible explanation for the lack of significant result on parasocial interaction might be explained by the minor amount of participants in this study who had actually developed a parasocial relation with the online influencer. This resulted in a small sample that was used in the analysis measuring the influence of parasocial interaction, and this might explain the lack of significance.
This study also contains some limitations that must be taken into account and could have affected the results. Firstly, a convenience sample was used to recruit the participants. However, a convenience sample is vulnerable to selection bias and a high level of sampling error. Due to these reasons, this study might not represent the population of women between 18 and 28 years old. Secondly, questions about persuasion knowledge might reveal the commercial nature of the Instagram post. Since brand responses were measured after these questions, participants might have been biased and their answers could have been the result of the questions about advertising rather than the actual stimulus. Thirdly, the online influencer chosen in this study is one of the biggest online influencers in The Netherlands, and therefore participants might have already known that most of her posts are sponsored. Another online influencer, picture or brand might lead to different results. Finally, this study aimed to enhance ecological validity by showing multiple Instagram pictures, however, since participants did not actually open the app, the survey might have created a different experience than in a natural setting.
Future research needs to examine more closely the influence of different types of sponsorship disclosure in the field of influencer marketing on social media. An interesting
question further studies should focus on is the different types of explicit and implicit disclosures, for example different types of hashtags, but also on sponsorship disclosures on different social media platforms, such as vlogs and videos on YouTube or Snapchat. Moreover, the role of a parasocial relation between the influencer and their followers must also be examined in further detail using a bigger sample size. In addition, it is also interesting to examine the specific factors that influence such parasocial relationship. This has not been studied and could be useful for marketers to understand this process and help them identify convincing online influencers social media.
This research extends our knowledge of sponsorship disclosure in influencer
marketing. While earlier research shows the effects of a sponsorship disclosure on television, radio and online blogs (Boerman et al., 2012; 2014; 2015, Carr & Hayes, 2014; van
Reijmersdal et al., 2015; 2016; Wei et al., 2008), this is, in the author’s knowledge, the first study that found support that an implicit sponsorship disclosure is also related to effects on persuasion knowledge, brand attitude, purchase intention and source credibility on the social media platform Instagram. In addition, this study shows that different types of sponsorship disclosure also could have different effects, and therefore highlights the importance of research on sponsorship disclosure to determine the most effective way a disclosure can be communicated on different types of media. Finally, this is the first time that the concept of parasocial interaction has been used to explore the relation between sponsorship disclosure, brand responses and source credibility on social media.
This study suggests some important practical implications for the FTC, marketers and online influencers. Based on the findings in this study, it can be advised that the FTC changes its guidelines and requires that online influencers disclose their relation with a brand in a transparent way by thanking the brand and mentioning how the product was obtained, for example, stating the product was a gift. This study showed that a hashtag with the
abbreviation #ad was not effective in communicating the commercial purpose. If the FTC does want to introduce a hashtag for social media that only allows its users to share a short amount of text (e.g. Twitter and Instagram), then it can be advised that consumers are given training about this hashtag in order to create awareness and to make sure the hashtag will be recognized as advertising.
In addition, marketers and online influencers can still use the hashtag #ad in order to disclose their sponsored content. In this way, they do not violate the FTC guidelines, but also do not experience the negative effects of an effective disclosure. However, it is important for marketers and online influencers to understand the potential long-term consequences of deceiving audiences, which can hurt their reputation and credibility. Consumers desire to have a fair chance in making their own decisions about a product or a brand without dealing with deception. Consequently, there is no doubt that we can expect even stricter disclosure requirements from the FTC in the future.
References
Auter, P. (1992). TV that talks back: An experimental validation of a parasocial interaction scale. Journal Of Broadcasting & Electronic Media, 36(2), 173-181.
Ballantine, P. W., & Martin, B. A. (2005). Forming parasocial relationships in online communities. NA-Advances in Consumer Research Volume 32.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of
personality and social psychology, 51(6), 1173.
Bhatnagar, N., Aksoy, L., & Malkoc, S. A. (2004). Embedding brands within media content: The impact of message, media, and consumer characteristics on placement efficacy. The
Boerman, S. C., & van Reijmersdal, E. A. (2016). Informing Consumers about “Hidden” Advertising: A Literature Review of the Effects of Disclosing Sponsored Content. In
Advertising in new formats and media: Current research and implications for marketers
(pp. 115-146). Emerald Group Publishing Limited.
Boerman, S. C., van Reijmersdal, E. A., & Neijens, P. C. (2012). Sponsorship disclosure: Effects of duration on persuasion knowledge and brand responses. Journal of
Communication, 62(6), 1047-1064.
Boerman, S. C., van Reijmersdal, E. A., & Neijens, P. C. (2014). Effects of sponsorship disclosure timing on the processing of sponsored content: A study on the effectiveness of European disclosure regulations. Psychology & Marketing, 31(3), 214-224.
Boerman, S. C., Van Reijmersdal, E. A., & Neijens, P. C. (2015). How audience and disclosure characteristics influence memory of sponsorship disclosure. International
Journal of Advertising.
Brown, D., & Fiorella, S. (2013). Influence marketing: How to Create, manage, and measure
Brand Influencers in Social media marketing. Que Publishing.
Carr, C. T., & Hayes, R. A. (2014). The effect of disclosure of third-party influence on an opinion leader's credibility and electronic word of mouth in two-step flow. Journal of
Interactive Advertising, 14(1), 38-50.
Chu, S. C., & Kamal, S. (2008). The effect of perceived blogger credibility and argument quality on message elaboration and brand attitudes: An exploratory study. Journal of
Interactive Advertising, 8(2), 26-37.
Colliander, J., & Dahlén, M. (2011). Following the fashionable friend: The power of social media. Journal of Advertising Research, 51(1), 313-320.
Colliander, J., & Erlandsson, S. (2015). The blog and the bountiful: Exploring the effects of disguised product placement on blogs that are revealed by a third party. Journal of
Marketing Communications, 21(2), 110-124.
Coyne, S. M., Padilla-Walker, L. M., & Howard, E. (2013). Emerging in a digital world a decade review of media use, effects, and gratifications in emerging adulthood.
Emerging Adulthood, 1(2), 125-137.
Dekker, K., & Van Reijmersdal, E. A. (2013). Disclosing celebrity endorsement in a
television program to mitigate persuasion: How disclosure type and celebrity credibility interact. Journal of Promotion Management, 19(2), 224-240.
Dimofte, C. V., Haugtvedt, C. P., & Yalch, R. F. (2015). Consumer Psychology in a Social
Media World. Routledge.
Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An Introduction to
Theory and Research. Reading, MA: Addison-Wesley Publishing Co.
Fransen, M. L., & Fennis, B. M. (2014). Comparing the impact of explicit and implicit resistance induction strategies on message persuasiveness. Journal of Communication,
64(5), 915-934.
Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with persuasion attempts. Journal of consumer research, 21(1), 1-31.
FTC, (2015a). The FTC’s Endorsement Guides: What People Are Asking | Federal Trade
Commission. Ftc.gov. Retrieved 8 November 2016, from
https://www.ftc.gov/tips- advice/business-center/guidance/ftcs-endorsement-guides-what-people-are-asking#social
FTC, (2015b). Native Advertising: A Guide for Businesses | Federal Trade Commission.
Ftc.gov. Retrieved 15 January 2017, from
Giles, D. C. (2002). Parasocial interaction: A review of the literature and a model for future research. Media psychology, 4(3), 279-305.
Gordon, C., & Bruner, I. I. (2009). Marketing Scales Handbook: A Compilation of Multi-Item
Measures for Consumer Behavior & Advertising Research. GCBII Productions.
Homer, P. M. (1990). The mediating role of attitude toward the ad: Some additional evidence.
Journal of Marketing Research, 78-86.
Horton, D., & Wohl, R. (1956). Mass communication and para-social interaction: Observations on intimacy at a distance. Psychiatry, 19(3), 215-229.
Hwang, Y. & Jeong, S. (2016). “This is a sponsored blog post, but all opinions are my own”: The effects of sponsorship disclosure on responses to sponsored blog posts. Computers
In Human Behavior, 62, 528-535.
Kulin, E., & Blomgren, L. (2016). Credible or not: A study on the factors influencing consumers' credibility assessment of product placements on Instagram.
Labrecque, L. I. (2014). Fostering consumer–brand relationships in social media
environments: The role of parasocial interaction. Journal of Interactive Marketing,
28(2), 134-148.
Lee, K. T., & Koo, D. M. (2012). Effects of attribute and valence of e-WOM on message adoption: Moderating roles of subjective knowledge and regulatory focus. Computers in
Human Behavior, 28(5), 1974-1984.
Liljander, V., Gummerus, J., & Söderlund, M. (2015). Young consumers’ responses to suspected covert and overt blog marketing. Internet Research, 25(4), 610-632. Lu, L. C., Chang, W. P., & Chang, H. H. (2014). Consumer attitudes toward blogger’s
sponsored recommendations and purchase intention: The effect of sponsorship type, product type, and brand awareness. Computers in Human Behavior, 34, 258-266.
Miles, J. (2013). Instagram power: Build your brand and reach more customers with the
power of pictures. McGraw Hill Professional.
MacKenzie, S., Lutz, R., & Belch, G. (1986). The Role of Attitude toward the Ad as a Mediator of Advertising Effectiveness: A Test of Competing Explanations. Journal Of
Marketing Research, 23(2), 130.
Maheshwari, S. (2016). Endorsed on Instagram by a Kardashian, but Is It Love or Just an
Ad?. Nytimes.com. Retrieved 2 November 2016, from
http://www.nytimes.com/2016/08/30/business/media/instagram-ads-marketing-kardashian.html?_r=1
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers' perceived expertise, trustworthiness, and attractiveness. Journal of advertising, 19(3), 39-52.
Perse, E. M., & Rubin, R. B. (1989). Attribution in social and parasocial relationships.
Communication Research, 16(1), 59-77.
Petty, R. D., & Andrews, J. C. (2008). Covert marketing unmasked: A legal and regulatory guide for practices that mask marketing messages. Journal of Public Policy &
Marketing, 27(1), 7-18.
Petty, R. E., & Cacioppo, J. T. (1977). Forewarning, cognitive responding, and resistance to persuasion. Journal of Personality and Social Psychology, 35(9), 645.
Rubin, A. M., Perse, E. M., & Powell, R. A. (1985). Loneliness, parasocial interaction, and local television news viewing. Human Communication Research, 12(2), 155-180. Skumanich, S. A., & Kintsfather, D. P. (1998). Individual media dependency relations within
television shopping programming a causal model reviewed and revised. Communication
Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase intentions. Journal of Current Issues & Research in Advertising, 26(2), 53-66.
Sternthal, B., Dholakia, R., & Leavitt, C. (1978). The persuasive effect of source credibility: Tests of cognitive response. Journal of Consumer research, 4(4), 252-260.
Tessitore, T., & Geuens, M. (2013). PP for ‘product placement’or ‘puzzled public’? The effectiveness of symbols as warnings of product placement and the moderating role of brand recall. International Journal of Advertising, 32(3), 419-442.
Tormala, Z. L., & Petty, R. E. (2002). What doesn't kill me makes me stronger: the effects of resisting persuasion on attitude certainty. Journal of personality and social psychology,
83(6), 1298.
Van Reijmersdal, E. A., Fransen, M. L., van Noort, G., Opree, S. J., Vandeberg, L., Reusch, S., van Lieshout, F., & Boerman, S. C. (2016). Effects of Disclosing Sponsored Content in Blogs How the Use of Resistance Strategies Mediates Effects on Persuasion.
American Behavioral Scientist, 0002764216660141.
Van Reijmersdal, E. A., Lammers, N., Rozendaal, E., & Buijzen, M. (2015). Disclosing the persuasive nature of advergames: Moderation effects of mood on brand responses via persuasion knowledge. International Journal of Advertising, 34(1), 70-84.
Van Reijmersdal, E. A., Neijens, P. C., & Smit, E. G. (2007). Effects of television brand placement on brand image. Psychology & Marketing, 24(5), 403-420.
Van Reijmersdal, E. A., Neijens, P. C., & Smit, E. G. (2010). Customer magazines: Effects of commerciality on readers' reactions. Journal of Current Issues & Research in
Advertising, 32(1), 59-67.
Wei, M. L., Fischer, E., & Main, K. J. (2008). An examination of the effects of activating persuasion knowledge on consumer response to brands engaging in covert marketing.
Appendix 1: Conditions
Appendix 2: Measures of the scales
Persuasion Knowledge (Boerman et al., 2012; van Reijmersdal, Neijens & Smit, 2010; van
Reijmersdal et al., 2016)
1. The Instagram post of [ONLINE INFLUENCER] is advertising. 2. The Instagram post of [ONLINE INFLUENCER] is commercial. 3. The Instagram post of [ONLINE INFLUENCER] contains advertising.
4. [ONLINE INFLUENCER] received anything of value in return of the Instagram post.
Brand Attitude (Gordon & Bruner, 2009)
1. I think [BRAND] is good. 2. I think [BRAND] is pleasant. 3. I think [BRAND] is favourable. 4. I think [BRAND] is positive. 5. I think [BRAND] is likeable. 6. I think [BRAND] is of high quality.
Purchase Intention (Spears & Singh, 2004)
1. I will buy [BRAND].
2. I have the intention to buy a product of [BRAND]. 3. I am interested in buying a product of [BRAND].
4. It is likely that I will buy a product of [BRAND] in the future.
Source Credibility (Ohanian, 1990)
1. I think the Instagram post of [ONLINE INFLUENCER] is honest. 2. I think the Instagram post of [ONLINE INFLUENCER] is trustworthy.
3. I think the Instagram post of [ONLINE INFLUENCER] is convincing. 4. I think the Instagram post of [ONLINE INFLUENCER] is biased. (reversed) 5. I think the Instagram post of [ONLINE INFLUENCER] is not credible. (reversed)
Parasocial Interaction (Rubin, Perse & Powell, 1985)
1. [ONLINE INFLUENCER] makes me feel comfortable, as if I were with a friend. 2. When I interact with [ONLINE INFLUENCER], I feel included.
3. I can relate to [ONLINE INFLUENCER].
4. I like hearing what [ONLINE INFLUENCER] has to say. 5. I care about what [ONLINE INFLUENCER] does in her life. 6. I hope [ONLINE INFLUENCER] can achieve her goals.
7. I find myself comparing my ideas with what [ONLINE INFLUENCER].
Product Interest Reijmersdal, Neijens, and Smit (2007) 1. I like buying watches.
2. I like to view watches on social media. 3. I am interest in watches.
Advertising Interest
1. I think the Instagram post is entertaining. 2. I think the Instragm post is appealing. 3. I think the Instagram post is interesting.
4. I think the Instagram post is irritating. (reversed)
Appendix 3: Mediation Analyses
Explicit Disclosure Brand Attitude Persuasion Knowledge -.25 (.18) -.07 (.09) -.24 (.18) .20 (.22) Implicit Disclosure Brand Attitude Persuasion Knowledge Explicit Disclosure Source Credibility Persuasion Knowledge Implicit Disclosure Source Credibility Persuasion KnowledgeFigure 2. Mediation analyses of differences in brand attitude and source credibility.
Note. a = unstandardized regression coefficient; b = standard error. *p < .05. In accordance
with Baron and Kenny (1986), Step 1: text above the horizontal line; Step 2: text to the left of the diagram; Step 3: text to the right of the diagram; Step 4: text under the horizontal line.
-.40 (.18)* -.12 (.10) -.34 (.19) .52 (.21)* .01 (.21) -.32 (.10)* .07 (.20) .20 (.22) -.03 (.23) -.38 (.11)* .17 (.22) .52 (.21)*
