Disclosing Influencer Sponsorships: The Effect of Disclosures on Adolescents’ Susceptibility to Sponsored Influencer Content via Persuasion Knowledge
Sophia van Dam 10200797
Master’s Thesis
Research Master’s programme Communication Science Graduate School of Communication, University of Amsterdam
Supervisor: Dr. Eva van Reijmersdal 30 June 2017
Abstract
Influencer marketing – the practice of sponsoring a person to make advertising to its network of followers – is increasing in popularity to target young consumers. Both the vulnerability of this group and the lack of sponsorship disclosures raises concerns about the ethics of
influencer marketing. There is still little information regarding adolescents’ persuasion knowledge and the specific disclosures that will help them deal with influencer marketing. This study aimed to address this gap. First, we investigated adolescents’ current persuasion knowledge within the influencer marketing context through focus group discussions (N = 20, aged 12 to 16). Next, we assessed the impact of different disclosures on activating
adolescents’ conceptual and attitudinal persuasion knowledge, and subsequent outcomes, i.e. brand attitude, purchase intention, and attitude towards the influencer in an experiment. We also tested whether an additional explanation of the sponsorship mechanism would moderate the effects of disclosure type. A 3 (disclosure: sponsorship vs persuasive intent vs no
disclosure) x 2 (explanation: with and without) experimental between-subject design was used (N = 412, aged 12 to 16). The findings of the focus group discussions indicated that adolescents are highly knowledgeable of influencer marketing practices, but do not apply their knowledge to the sponsored content. The outcomes of the experiment showed that a disclosure, compared to no disclosure, is effective in activating adolescents’ conceptual persuasion knowledge and attitudinal persuasion knowledge, which in turn decreases brand attitude and attitude toward the influencer. These effects were stronger for a disclosure of persuasive intent compared to a disclosure of sponsorship only. There were no effects on purchase intention and no moderating effect of the sponsorship mechanism explanation. The results of both studies contribute to the theoretical understanding of adolescents’ persuasion knowledge and provide practical guidelines for future legal and policy decisions concerning the protection of youth with respect to influencer marketing.
Disclosing Influencer Sponsorships: The Effect of Disclosures on Adolescents’ Susceptibility to Sponsored Influencer Content via Persuasion Knowledge
Advertisers are continuously looking for novel formats to reach and subsequently persuade their target audiences. Nowadays, marketers are sponsoring so-called online influencers on social media platforms – which is called online influencer marketing. Online influencers are “people who built a large network of followers, and are regarded as trusted tastemakers in one or several niches” (De Veirman, Cauberghe, & Hudders, 2016, p.1). Influencer marketing is a new way of practicing native advertising which entails embedding commercial messages into editorial content. Marketers are able to target adolescents through sponsoring the influencer by sending products, paying to mention or make content about the brand. Adolescents are an interesting target group for marketers as they still influence their parents’ buying behaviour while having considerable purchasing power themselves
(Livingstone & Helsper, 2006). Additionally, they are the consumers of the future and a bond between a consumer and a brand that started at a young age often lasts until adulthood
(McNeal, 1999). No wonder that 84 percent of marketers was interested in launching an influencer marketing campaign in 2016. Even 22 percent of marketers considered influencer marketing as ‘the fastest-growing online customer acquisition method’ (Launchmetrics, 2015).
Due to the embedded and subtle nature of influencer marketing, its targeted audiences are less aware of the commercial content and are less likely to use their persuasion
knowledge (i.e. knowledge of advertising and marketer’s motives) as critical defence compared to traditional advertising (Friestad & Wright, 1994; Moore & Rideout, 2007; Verhellen, Oates, & De Pelsmacker, 2014). Additionally, viewers generally do not regard influencers’ content as a source of advertisement (Wu, 2016). Adolescents in particular can be affected by hidden sponsorships as their persuasion knowledge has not reached adult
levels yet and they have not fully developed the cognitive and information processing skills to thoroughly comprehend commercial messages. This makes them more vulnerable for these persuasive attempts (Lawlor, Dunne, & Rowley, 2016; Livingstone & Helsper, 2006;
Rozendaal, Lapierre, Van Reijmersdal, & Buijzen, 2011; Verhellen et al., 2014) and raises considerable concerns among parents, educational professionals and policy makers (CvdM, 2017; Verhellen et al., 2014).
It is an established legal principle within communication law that advertising needs to be readily identifiable for its audience, in order to be considered fair (Middleton, Trager, & Chamberlin, 2001). A sponsorship disclosure is often proposed as tool to inform audiences about the hidden commercial intent of a message (Moore & Rideout, 2007; Rozendaal et al., 2011). International authorities (e.g. the Committee of Advertising Practice, the Federal Trade Commission) have already proposed guidelines for disclosing online sponsored
content, mostly stressing to disclose it fully by making the relationship between the brand and the influencer clear (CAP, 2017). Meanwhile, platforms have to deal with this issue too. Recently, YouTube has introduced a function for including a disclosure that will be shown the first 10 seconds of the video in the lower left corner, stating ‘paid promotion’ (YouTube, 2016).
However, empirical evidence of adolescents’ persuasion knowledge regarding
influencer marketing and the implementation and effects of disclosures on adolescents is still lacking. Therefore, the first aim of this study is to investigate adolescents’ current state of persuasion knowledge regarding online influencer marketing. Second, this study aims to assess the effect of disclosures within influencer content on adolescents’ activation of persuasion knowledge. Third, this study investigates whether this decreases adolescents’ susceptibility to advertising. More specific, it examines the effects of two disclosure types, relating to sponsorship and persuasive intent, on brand attitude, purchase intention and
attitude towards the influencer through their impact on conceptual and attitudinal persuasion knowledge. Besides, Tessitore and Geuens (2013) showed that awareness and comprehension of a disclosure is necessary, in order for it to be effective. Given their lower persuasion knowledge and underdeveloped cognitive abilities, adolescents likely benefit from an
additional explanation of the sponsorship mechanism. Therefore, the last aim of this study is to investigate the moderating effect of such an explanation on the effects of a disclosure.
This study answers the recent call for studies on disclosures of online native
advertising among a younger public made by different authors (Boerman & Van Reijmersdal, 2016; Hudders, De Pauw, Cauberghe, Panic, Zarouali, & Rozendaal, 2017; Livingstone & Helsper, 2006). Insights from this study contribute to our understanding of adolescents’ persuasion knowledge and the implementation of disclosures in influencer marketing. The outcomes of this study can be of guidance practically for policy makers. It will give insights on the question of the level of vulnerability of adolescents to influencer marketing and how to implement disclosures to empower young consumers effectively.
Theoretical framework
Sponsored content created by online influencers is generally not recognized as having commercial intent (Wu, 2016). By hiding the commercial intent of a message within online influencers’ content, marketers can avoid the resistance evoked by advertising, and increase effects of persuasive attempts.
Activation of Persuasion Knowledge
The Persuasion Knowledge Model (PKM) by Friestad and Wright (1994) explains how people cope with persuasive messages. When faced with persuasive messages (i.e. standard television commercials), people tend to activate persuasion knowledge, which is defined as “the personal knowledge about the tactics, intentions and strategies used in persuasion attempts. This knowledge helps consumers to identify how, when, and why
marketers try to influence them” (Friestad & Wright, 1994, p.1). Individuals can use their persuasion knowledge to cope with the persuasive attempt, and decide whether they want to be persuaded or whether they want to resist the persuasion. Rozendaal, Opree, and Buijzen (2016) differentiate two dimensions of persuasion knowledge. The first dimension is conceptual persuasion knowledge, and refers to the understanding of selling and persuasive intentions of advertising, such as the recognition of a message as being advertising. The second dimension is attitudinal persuasion knowledge, which entails the critical attitude towards advertising. This dimension involves critical beliefs about honesty, credibility and trustworthiness of the persuasive attempt. The hidden and integrated character of online influencer marketing complicates activation of persuasion knowledge (Owen, Lewis, Auty, & Buijzen, 2013; Verhellen et al., 2014), especially compared to traditional formats (i.e. TV commercials).
The hybrid and integrated character of the sponsored online influencer content is not the only factor that limits the activation of persuasion knowledge. Young adolescent’s cognitive abilities necessary for recognizing and processing sponsored online influencer content, such as emotion regulation and executive functioning, are not yet fully developed (Eagle, 2007; Moses & Baldwin, 2005; Rozendaal, Lapierre, Van Reijmersdal, & Buijzen, 2011; Rozendaal, 2016b). As a consequence, persuasion knowledge will not be triggered that easily, making adolescents more likely than adults to be influenced by advertising (Eagle, 2007).
Despite these notions, clear evidence is still missing on the current state of adolescents’ conceptual and attitudinal persuasion knowledge within today’s online
environment. Most studies on this topic have focused on children’s ability to discern between advertising and traditional media platforms (e.g. commercial breaks and television
of, for instance, product placement (Tutaj & Van Reijmersdal, 2012; Verhellen et al., 2014). The few studies on children’s persuasion knowledge in an online context mostly focused on advergames (Hudders, De Pauw, Cauberghe, Panic, Zarouali, & Rozenaal, 2017; Van Reijmersdal, Lammers, Rozendaal, & Buijzen, 2015; Vanwesenbeeck, Ponnet, & Walrave, 2017; Verhellen et al., 2014). To date, the only study exploring adolescents’ persuasion knowledge in an online media context, apart from advergames, is conducted by Lawlor, Dunne and Rowley (2016). They conducted a three staged study including focus group discussions, participant observation and in-depth interviews among adolescent girls, to investigate their knowledge of integrated and interactive advertising in social network sites (e.g. sponsored posts on Facebook, permission-based brand communication originating from the brand’s post on their own SNS page). Even though the girls purported to be fully aware of the integrated advertising, it turned out that they did not recognize the hidden commercial messages when they were exposed to them. The young adolescents tended to identify only banner, click through and pop up-advertisements as online advertising, while being unaware of the integrated ways of marketing practices that are used nowadays, such as sponsored posts that appear between posts of their friends (Lawlor et al., 2016). Next to their inability to recognize the sponsored posts, they were unaware of the persuasive intentions of these
practices.
Owing to this research gap, there is only limited information about adolescents’ understanding and knowledgeability of online influencer marketing. Before any steps can be taken to empower adolescents, it is essential to further explore the current level of
adolescents’ conceptual and attitudinal persuasion knowledge in the context of online influencer marketing. Therefore, the goal of this study is to investigate the nature and extent of adolescents’ persuasion knowledge in order to create a deeper understanding of
perceptions and impact of online influencer marketing. Therefore, the following research question is developed:
RQ1: What is the current level of adolescents’ a) conceptual persuasion knowledge and b) attitudinal persuasion knowledge with regard to online influencer marketing?
Furthermore, as disclosing sponsored influencer content is proposed as a solution in the debate on transparency and the ethics of this practice, it would be interesting to explore the viewpoint of today’s youth on this solution. Hence, the second research question is proposed:
RQ2: What are the perceptions of disclosures for online influencer marketing among adolescents?
Finally, it remains unclear how and which disclosures can help to infer persuasive attempts. Therefore, the current study will investigate the possible effects of a disclosure and the following research question is proposed:
RQ3: How do disclosures for influencer marketing affect a) brand attitude, b) purchase intention and c) attitude towards the influencer?
Disclosures
Recent studies show that the activation of adolescents’ persuasion knowledge can be stimulated by the presence of an external cue, such as a sponsorship disclosure (e.g.
Boerman, Van Reijmersdal, & Neijens, 2015; Van Reijmersdal et al., 2016, Vanwesenbeeck et al., 2017). A disclosure can serve as a forewarning of the upcoming persuasive attempt and thus can activate resistance to persuasion (Fransen & Fennis, 2014).
Guidelines have been proposed by different international authorities (e.g. the Dutch Advertising Committee, UK’s Internet Advertising Bureau and the Federal Trade
Commission) stating that consumers should be made aware of the content’s commercial nature so they can make informed decisions about their interaction with the advertising and
the weight they would give the information conveyed in the ad. Recently, the Committee of Advertising Practice (CAP) has issued new guidelines for brands working with influencers, urging both parties to protect consumers before they engage with content that is sponsored (CAP, 2017). These guidelines emphasize above all to fully disclose sponsored content.
Even though the labelling of sponsored content is often proposed as a remedy for a more transparent online media environment (Campbell, Mohr, & Verlegh, 2012), insights on whether these disclosures actually help young consumers to recognize native advertising and activate their persuasion knowledge are limited.
Effects on conceptual persuasion knowledge. Conceptual persuasion knowledge
entails different dimensions and people can acquire persuasion knowledge at different levels (Robertson & Rossiter, 1974).
A first level relates to the necessity to recognize the advertising and the source of the message. This could be reached by adding a third-party disclosure to the content, stating that the influencer advertises a certain brand within its content. This disclosure makes adolescents more likely to identify the persuasive messages within the content. By showing such a
disclosure at the beginning of the message, a priming effect may occur: as the disclosed information (i.e. that the content contains advertising) is fresh in memory, it is easy to access while processing the content (Van Reijmersdal, Tutaj, & Boerman, 2013). Subsequently, the commercial content will be more easily recognized. Disclosing sponsorship is therefore a first step in the process of activating adolescents’ persuasion knowledge.
The next level of conceptual persuasion knowledge entails the understanding of the purposes of advertising (Robertson & Rossiter, 1974; Rozendaal et al., 2011). Campbell and Kirmani (2000) showed that realizing the ulterior motive of a message increases activation of persuasion knowledge. In other words, revealing the actual goal of the sponsored content (i.e. to persuade) improves the activation of conceptual persuasion knowledge. It is even shown
that this is the case when cognitive capacity is low (Campbell & Kirmani, 2000). Therefore, it could be effective to disclose the persuasive intent of a message because using persuasion knowledge requires cognitive capacity and adolescents tend to process advertising at a less elaborate level than adults (Buijzen, Van Reijmersdal, & Owen, 2010).
Together, it is expected that adolescents will benefit from a disclosure of sponsorship as this improves the recognition of the commercial content. Moreover, disclosing more dimensions of conceptual persuasion knowledge is expected to enable higher levels of conceptual persuasion knowledge. Therefore, it is expected that not only disclosing
sponsorship (i.e. the first level) but also exhibit its persuasive intentions (i.e. the second level) will result in higher levels of conceptual persuasion knowledge. The following hypothesis is proposed:
Hypothesis 1: Disclosure type affects conceptual persuasion knowledge, such that disclosure of sponsorship and persuasive intent will lead to higher levels of conceptual persuasion knowledge than solely a disclosure of sponsorship, whereas a disclosure of sponsorship will lead to higher levels of conceptual persuasion knowledge than no disclosure.
Effect of Conceptual Persuasion Knowledge on Attitudinal Persuasion Knowledge
Activated conceptual persuasion knowledge increases the motivation to resist the message by promoting feelings of psychological reactance (Brehm, 1966). That is, being aware of sponsorship and/or its persuasive intent causes people to feel restricted in their freedom to think and feel how they want, which motivates them to actively restore this freedom. Reactance produces resistance to persuasion through, for instance, critical
evaluation of the message (Fransen & Fennis, 2014). This critical attitude towards advertising (i.e. attitudinal persuasion knowledge) can serve as an attitudinal defence to help adolescents
question and discount advertising claims the way adults generally do (Knowles & Linn, 2004; Rozendaal, Buijs, & Van Reijmersdal, 2016a).
Research has also shown that people’s recognition of persuasive attempts (i.e.
conceptual persuasion knowledge) leads to more sceptical attitudes (i.e. attitudinal persuasion knowledge) (An, Jin, & Park, 2014; Boerman, Willemsen, & Van Der Aa, 2017; Boerman, Van Reijmersdal, & Neijens, 2012; Campbell & Kirmani, 2000; Opree & Rozendaal, 2015). An and colleagues (2014) showed in their study on children’s persuasion knowledge in advergames that recognition of advertising was a precursor of children’s critical attitude towards advertising. This aligns with the findings of Boerman et al. (2017), who found in their study on sponsored Facebook posts that the activation of conceptual persuasion knowledge caused higher levels of attitudinal persuasion knowledge. When adolescents recognize a sponsored message in the influencers’ content (e.g. by a disclosure), they will be more aware of the sponsorship and be more critical towards the content (Boerman et al., 2012). Hence, it is proposed that the activation of conceptual persuasion knowledge is followed by the activation of attitudinal persuasion knowledge:
Hypothesis 2: Conceptual persuasion knowledge affects attitudinal persuasion
knowledge, such that higher levels of conceptual persuasion knowledge lead to higher levels of attitudinal persuasion knowledge.
Effects of Disclosures through Persuasion Knowledge
From a theoretical viewpoint (e.g. Friestad & Wright, 1994), it is likely that
disclosures activate adolescents’ existing persuasion knowledge which they can use to cope with the persuasive attempt. This could be by processing the content in a more critical way and choose to resist the persuasive attempt or to be persuaded (Friestad &Wright, 1994). When adolescents realize that the content is not just entertainment but has a persuasive intent the change of meaning principle (Friestad & Wright, 1994) will occur: when someone
realizes the actual intent of the content, the meaning of the content is redefined which may impact their response to the content, such as brand attitude, purchase intention, and attitude towards the online influencer. Change of meaning can result in ‘detachment’ (Friestad & Wright, 1994), implying that adolescents disconnect from the content and are conscious that it is used to persuade. When adolescents realize that the content they were watching is not meant to entertain them, but actually to sell them certain products, they can feel fooled and this realization will motivate them to process the content more critically. Attitudinal
persuasion knowledge is shown effective in generating resistance (An & Stern, 2011; Opree & Rozendaal, 2015; Vanwesenbeeck et al., 2017; Zuwerink & Devine, 1996). Subsequently, this could lead to less favourable advertising effects, such as brand attitude and purchase intention.
Effects of disclosures on brand attitude via persuasion knowledge. Several studies
on disclosures, conducted among adults, found that a disclosure leads to less favourable brand attitudes (Campbell et al., 2012; Wojdynski & Evans, 2015; Wei, Fischer, & Main, 2008; Boerman et al., 2012; 2015). Moreover, findings indicate that persuasion knowledge is more easily activated after the exposure of a disclosure, and resulted in less positive brand attitudes (Boerman et al., 2012; 2015; Campbell et al., 2012). Other studies, however, did not find significant effects of disclosures on brand attitude via persuasion knowledge (Dekker & Van Reijmersdal, 2013; Van Reijmersdal, 2015).
Research on brand effects on adolescents is scant (i.e. Livingstone & Helsper, 2016, Rozendaal et al., 2011). Van Reijmersdal et al. (2016) replicated Boerman et al.’s (2012) study on brand placement disclosures on television, using adolescents between the age of 13 and 17, and observed that a disclosure has limited effects on adolescents’ brand attitude via persuasion knowledge. Even though adolescents’ conceptual persuasion knowledge was partially activated (i.e. understanding persuasive attempt, but not recognition of advertising),
there was no significant effect found on brand attitude via attitudinal persuasion knowledge. This indicates that while a disclosure may activate adolescents’ conceptual and attitudinal persuasion knowledge, this is not a certainty for processing the content more critically, which has been attributed to adolescents’ underdeveloped cognitive abilities and information
processing skills (Livingstone & Helsper, 2006; Rozendaal et al., 2011).
Studies on the relationship between persuasion knowledge and brand attitude among children showed that activated persuasion knowledge leads to a less favourable brand attitude within traditional advertising formats (Rozendaal, Buijzen, & Valkenburg, 2012). Regarding online embedded advertising, Rozendaal, Slot, Van Reijmersdal, and Buijzen (2013) found that attitudinal persuasion knowledge was an effective defence, such that higher attitudinal persuasion knowledge led to less favourable brand attitudes.
In sum, it is unclear what the effects of conceptual and attitudinal persuasion
knowledge will be on adolescents’ brand attitude with regard to online influencer marketing. Based on findings of studies conducted among adults, it is expected that activated attitudinal persuasion knowledge, due to a disclosure that triggers conceptual persuasion knowledge, will lead to less favourable brand attitude. Therefore, the following hypotheses are proposed:
Hypothesis 3a: Attitudinal persuasion knowledge affects brand attitude, such that higher levels of attitudinal persuasion knowledge result in a less favourable brand attitude.
Hypothesis 3b: Disclosure type affects brand attitude via the activation of conceptual and subsequently attitudinal persuasion knowledge, such that disclosure of
sponsorship and persuasive intent will lead to more activation of persuasion
knowledge and subsequently a less positive brand attitude than solely a disclosure of sponsorship, whereas a disclosure of sponsorship will lead to more activation of
persuasion knowledge and subsequently less favourable brand attitude than no disclosure.
Effects of disclosures on purchase intention via persuasion knowledge. Another
and important goal of advertising is that consumers will buy the advertised product. Research on the effects of disclosures on purchase intention again showed mixed findings (Rozendaal et al., 2016a; Tessitore & Geuens, 2013; Vanwesenbeeck et al., 2017). On the one hand, Rozendaal et al. (2016a) found in their study on disclosures on television among children, that a disclosure activated attitudinal persuasion knowledge which subsequently lowered product desire. Attitudinal persuasion knowledge was increased by disclosing the
manipulative intent of the message. This aligns with the study of An and Stern (2011) that showed a direct negative effect of a textual and graphic disclosure on children’s (8 - 11 years old) product desire. A study with adults showed similar results; a disclosure reduced movie watchers’ intention to purchase the advertised brand (Tessitore & Geuens, 2013).
On the other hand, Vanwesenbeeck et al. (2017) found a reversed effect in their study on advergames. They noticed that adolescents with an increased understanding of the
persuasive intention of advertising were more likely to buy the advertised brand, which could be due to the specific and entertaining context of advergames. According to the authors, the integrated character of advertising in advergames could mean that a player first needs to understand why a brand is integrated before considering buying the product.
In sum, empirical evidence whether a disclosure will alter adolescents’ purchase intention is needed. To date, it remains unclear whether the activation of persuasion knowledge will mitigate advertising effects in the context of sponsored online influencer content (e.g. purchase intention). In line with theoretical notions (i.e. PKM) and empirical evidence (Rozendaal et al., 2016a; Tessitore & Geuens, 2013), is expected that activated conceptual and attitudinal persuasion knowledge, through the help of a disclosure, can offset
the changing of meaning principle and a detachment effect, which will have a negative impact on adolescents’ purchase intention, compared to content without a disclosure.
Hypothesis 4a: Attitudinal persuasion knowledge affects purchase intention, such that higher levels of attitudinal persuasion knowledge result in a less positive purchase intention.
Hypothesis 4b: Disclosure type affects purchase intention via the activation of
conceptual and consequently attitudinal persuasion knowledge, such that disclosure of sponsorship and persuasive intent will lead to more activation of persuasion
knowledge and subsequently a less positive purchase intention than disclosure of the sponsorship, whereas a disclosure of sponsorship will lead to more activation of persuasion knowledge and subsequently less positive purchase intention than no disclosure.
Effects of disclosures on attitude towards online influencer via persuasion knowledge. While a disclosure can impact brand attitude and purchase intention, it can also
affect adolescents’ attitude towards the influencer. Previous studies (e.g. Amos, Holmes, & Strutton, 2008; Dekker & Van Reijmersdal, 2013) mostly focused on the attitude towards the endorser as a moderator, establishing whether this attitude influences the effects of a
disclosure. Yet, it is equally important to investigate this attitude as an outcome, and establish whether the attitude changes through the use of a disclosure. Information about such an attitude change might affect how marketers and online influencers themselves will address the current issue of transparency.
According to Kozinets, De Valck, Wojnicki, and Wilner (2010), online influencers who engage in word-of-mouth marketing campaigns often find themselves in a tense moral conflict. They have difficulties in balancing the editorial and commercial content, as strict, clear and general guidelines on sponsorship disclosures are missing. This tension can
influence how online influencers disclose the sponsorship (Kozinets et al., 2010). In addition, it is not clear how their audience respond to the disclosure of the sponsorship – ranging from appreciations to antipathy.
In line with the PKM (Friestad & Wright, 1994), it is expected that adolescents’ perspective of an influencer will change when activating their conceptual persuasion
knowledge and realizing that the created content is not just for entertainment, but also serves commercial purposes. They can feel fooled by the influencer, potentially leading to
detachment of the content and presumably leading to a less positive attitude towards the influencer. Additionally, literature on embedded advertising in television programs found a spill-over effect (i.e. when attitudes towards one object influence the attitudes towards another) when persuasion knowledge was activated (Van Reijmersdal, Smit, & Neijens, 2010). Critical attitudes that are a result of activated attitudinal persuasion knowledge could spill over to the online influencer. Furthermore, Colliander and Erlandsson (2015) showed that a third-party disclosure of a sponsored blog results in a decreased perceived credibility of and attitude toward the blog, compared to a sponsored blog without a disclosure.
Based on these arguments, it is expected that a disclosure affects online influencer attitude as it makes adolescents aware of the commercial intention (i.e. increased conceptual persuasion knowledge) and subsequent more critical evaluation of the influencer (i.e.
increased attitudinal persuasion knowledge). The following hypotheses are proposed: Hypothesis 5a: Attitudinal persuasion knowledge affects attitude towards the
influencer, such that higher levels of attitudinal persuasion knowledge result in a less favourable attitude towards the influencer.
Hypothesis 5b: Disclosure type affects online influencer attitude via conceptual and subsequently attitudinal persuasion knowledge, such that disclosure of sponsorship and persuasive intent will lead to higher activation of persuasion knowledge and
subsequently less positive influencer attitude than disclosure of sponsorship, whereas a disclosure of sponsorship will lead to higher activation of persuasion knowledge and subsequently less favourable influencer attitude than no disclosure.
The Role of Explanation of the Sponsorship Mechanism
Awareness and comprehension of a disclosure is necessary, in order to communicate its message effectively (Tessitore & Geuens, 2013). A lack of awareness could be the reason why many studies on disclosures and its ability to alter persuasion effects have not shown significant results (e.g. Van Reijmersdal et al., 2016). It may be necessary for the disclosure’s effectiveness to provide extensive information of the sponsorship mechanism and use of disclosures. The more someone elaborates on the disclosure and its meaning, the better this information is accessible when exposed to it (Tessitore & Geuens, 2013). Additional information about the sponsorship mechanism and disclosures will therefore improve the noticeability and impact of the disclosure.
Furthermore, the cognitive defence view states that once consumers have knowledge about the persuasive intent of advertising, they become more critical about the advertiser’s motives in general (Hudders, Cauberghe, Panic, & De Vos, 2016). By providing an
explanation of the sponsorship mechanism behind online influencer marketing, the information regarding advertising and how it works will be cognitively better available. When adolescents’ understanding and comprehensibility of the disclosure increase, disclosures will activate conceptual persuasion knowledge more easily.
These assumptions are supported in studies on advertising literacy training and interventions, which demonstrate that even a brief intervention can significantly impact children’s recognition of advertising in advergame (An et al., 2014) and that a training session can improve conceptual and attitudinal persuasion knowledge (Hudders et al., 2016).
Based on these outcomes, it is expected that providing information about the online influencer marketing mechanism will create more awareness when watching a sponsored video, and strengthen the effects of disclosures. The following hypothesis is proposed:
Hypothesis 6: The effect of disclosures on conceptual persuasion knowledge is moderated by sponsorship explanation such that the effect of disclosures leads to higher levels of persuasion knowledge when there is a sponsorship explanation available.
Altogether, it is expected that the indirect effect of a disclosure through conceptual and attitudinal persuasion knowledge on brand attitude, purchase intention and attitude towards the influencer is moderated by an additional explanation of the sponsorship mechanism. Such that an additional explanation strengthens the effect of a disclosure on brand attitude, purchase intention and attitude towards the influencer through conceptual and attitudinal persuasion knowledge. Therefore, the final hypothesis is proposed:
Hypothesis 7: Sponsorship explanation moderates the indirect effect of disclosure type on a) brand attitude, b) purchase intention and c) attitude towards the influencer via conceptual and attitudinal persuasion knowledge, such that an explanation strengthens the indirect disclosure effects.
Figure 1. Conceptual model
+! +! +! #! #! #! Disclosure Type Conceptual Persuasion Knowledge Attitudinal Persuasion Knowledge Brand Attitude Purchase Intention
Attitude Towards the Influencer
Sponsorship Explanation
Study 1: Insights in Adolescents’ Persuasion Knowledge Method
Participants and procedure. To answer the research questions, four qualitative focus
group discussions were conducted with adolescents between 12 and 16 years old (N = 20, 40% female). Participants were accessed via a secondary school in a medium-sized city in the south of the Netherlands. Each discussion was with randomly chosen participants from the same grade (i.e. year 1, 2 and 3).
The focus group discussions were conducted within the school setting, in a separate room. At the start of the discussions, the interviewer gave a short introduction. An interview guide was used, leading to semi-structured discussions that ensured consistency in topics and questions across groups, yet also allowed for some flexibility within the topics. Questions were primarily aimed at assessing adolescents’ perceptions and knowledge of sponsored YouTube videos. All ethical guidelines concerning the conduct of research with children were strictly adhered to. Before the study started, informed consent was obtained from the schools, parents and participants.
Focus group discussions reflect the free-flowing, social ambience of online media, offering the participants an opportunity to discuss their use of these platforms and their awareness of commercial messages within the content. Participants were asked about their familiarity with YouTube videos and whether they recalled seeing brands on this platform or in videos. After this discussion, they were exposed to a video by Dutch YouTube influencers Ponckers that is sponsored by crisps brand Doritos. The influencers in the video try a new flavour of Doritos within a game. The brand and the product is clearly visible throughout the video. The participants were asked to write down individually their thoughts on yellow notes. This way, independent thinking was encouraged before the group discussion started. When the video was finished, participants were asked to share what they had written. The yellow
notes were discussed and categorised, leading to a big mind map in the middle of the table. The discussion leader focused primarily on anything written down that had to do with a brand, sponsorship or advertising. Additionally, questions about participants’ opinion o the sponsored content were asked. In this way, not only the conceptual knowledge of advertising could be examined, but also their attitudinal persuasion knowledge.
Finally, the discussion leader inquired what participants thought about the fairness of this practice and the use of disclosures. Again, participants were asked to write down
individually what they thought as a proper implementation of a sponsorship disclosure. Afterwards, these ideas were shared and discussed within the group. There were two written disclosures shown and participants were asked for their opinion about these disclosures. The focus group discussions lasted for approximately 50 minutes, which equals the duration of their classes. The discussions were audio recorded, transcribed and analysed.
Analysis. After transcribing, the focus groups were analysed in two steps. First, the
transcripts were read and codes were assigned to words and sentences. This coding was already focused as the codes were structured by the elements of conceptual and attitudinal persuasion knowledge (Rozendaal et al., 2016b). During this phase of focused coding, primary determinants of the different elements were defined and written down as memo’s. Moreover, codes were assigned to signs of adolescents’ perceptions of online influencer marketing, the use of disclosures and their responses to disclosures. The second step in the process consisted of specifying relationships between different findings within the elements. To understand the level of persuasion knowledge in the context of online influencer
marketing, an overview was created consisting of theme-categorized codes, memos and quotes. Additionally, the material created during the focus group discussions (i.e. notes with thoughts and remarks of the video) were used to complement the existing themes.
Conceptual persuasion knowledge. With respect to RQ1, the level of conceptual
persuasion knowledge was reviewed. Participants reported to be aware of the sponsored content at a certain level. Moreover, they showed knowledge of the sponsorship mechanism behind influencer marketing. When investigating whether the participants have ever noticed brands on YouTube, they all immediately acknowledged this and came with examples from their experience. After seeing the sponsored video, they also all mentioned the sponsored brand: “This whole video, which was like 10 minutes, is actually secretly only about
Doritos”. Remarkably, all participants were confident of their ability to recognize sponsored content, while indicating that it could be difficult for others to recognize the hidden
advertising. “Some people will know this [the hidden advertising], and some people won’t notice, but I do notice it immediately”. They showed to be aware of the sponsorship
mechanism as a revenue model behind influencer marketing. They were aware of the fact that the influencer received money from a brand to make advertising in their videos and that this is mostly integrated in the editorial content. “For example Center Parks. If they sponsor, they give money so EnzoKnol visits them and makes a video. Many people will watch this video and thereafter want to visit Center Parks”.
They were knowledgeable of the persuasive tactics that are used within influencer marketing and accentuate its integrated and hidden character: “Of course they will not say that [that it’s advertising] the whole time. It is not that you’ve been watching a true advertisement for ten minutes, but it is the whole time focused on a product which they
actually want to sell to you.” They even made comparisons of advertising in YouTube videos and television advertising, underlying the integrated and hidden character of the first: “I think that [advertising] within YouTube is inconspicuously integrated [the video]. With [television] advertising you just know the purpose”. Moreover, participants showed to be familiar with
concepts within the current advertising practice, as they used the words sponsoring and sponsorships throughout the discussions.
The intention to sell was understood really well, as the participants pointed this out as main goal of sponsored content: “They do this with the Doritos challenge so the people who watch this want to buy Doritos to do the challenge too”. However, participants did not demonstrate any understanding of persuasive intentions of the sponsored content.
Attitudinal persuasion knowledge. While conceptual persuasion knowledge seemed
to reach high levels among the participants, attitudinal persuasion knowledge was rather scant. Whereas Rozendaal et al. (2016b) described scepticism towards advertising and a general disliking of advertising as two important factors within attitudinal persuasion knowledge, the data of the current study showed a lack of scepticism and rather a high tolerance of advertising.
Lack of scepticism The respondents mainly reported the medium to be genuine and
the influencers to be honest. They seemed to value these characteristics a lot and showed that those are mainly the reasons to watch YouTube videos instead of television: “I have the idea that YouTube is honest, or at least more honest than television. And this makes you more attracted to it”. This perceived honesty appeared the most important reason for the adolescents to be low in scepticism towards the influencers’ content.
Only a few adolescents showed that they sometimes doubt the truthfulness of a video. It seemed that experience with videos which they thought to be only for entertainment, but turned out to be commercial, caused their scepticism. This refers to the change of meaning principle, whereby the realization of the real intention of the sender causes a change in
perception of the message. These adolescents also showed that they appreciate videos that are true and consist of honest information about a product. One participant explicitly said that she does not mind if the content is sponsored, but she just wants to have clarity: “I don’t care
whether it is sponsored or not, but sometimes it is annoying if you’re doubting whether it is sponsored or not. Then it would be nice to know whether this is the case”.
Furthermore, their confidence in recognizing advertising contributes to the nonchalant attitude towards sponsored content. “Most of the times you just notice whether a person [an influencer] is sincere or not, because when I notice that someone is not honest I like it a lot less”.
Tolerance of advertising. The adolescents showed a high level of tolerance of
influencer marketing. This tolerance can be divided into three aspects.
First, the adolescents accepted the integrated way of marketing. They did not mind these practices and perceive it as a part of YouTube and online influencers’ content.
Their acceptance is related to their understanding of the revenue model of influencer marketing. The adolescents argued that the online influencers have to engage in advertising, otherwise they would not be able to spend so much time on making videos. They argued that it is in benefit of the influencer as they have to make a living. “I don’t really care, I mean, they also have to make money and this is their way of doing so”. Moreover, the sponsorship is in their eyes not only in benefit of the influencer but also beneficial for them, as they argued that the money earned by the influencer will be used to make better videos. “I actually do like it [the sponsorships], because most of the times when they [the influencers] are sponsored this means they have more money so they can do more fun stuff or they make qualitatively better videos. Thanks to the sponsors”.
Finally, their relationship with the influencer appeared to be important. Experience with videos of a certain influencer make that the adolescents tolerate the influencers’ actions, also when this is advertising. This results in a certain compassion. Their connection with the influencer was explicitly pointed out by one participant: “I think that if you’re watching YouTube [videos], you’re feeling a certain connection. For example the thought that you
want to buy something when a famous YouTuber bought it, because that makes you feel more attracted to that product.”
Perception on disclosures. With respect to RQ2, the perception of adolescents on
disclosing the sponsored content is explored and seems to be determined by the perceived balance of editorial and commercial content. Within this dimension it is important to realize that it is not about the actual balance – as most of the videos are created just as a result of a sponsorship – but about the balance as perceived by the adolescents. Influencer marketing is found to be appropriate until a certain level. This level is exceeded when the commercial content is perceived as predominant. Two factors appear to affect this perception, namely the level of integration and the way of disclosure.
The participants accepted commercial messages, but they need to be integrated into the content. They appreciated the hidden character of influencer marketing. When it is too obvious and the balance is disrupted, for example when a brand is mentioned constantly, the participants get annoyed and show resistance. “In my opinion it’s not that bad, but in certain vlogs or videos the advertising is really disturbing as they overdo it a lot. I think that
everyone will notice the advertising and click away the video, just like I do most of those times. But in this case I wouldn’t mind. Yes, I know that it contains advertising, but it’s not that clear that it ruins the entertainment”. They preferred to be left in the dark with regard to the sponsorship behind the content than that they are aware of it.
Additionally, all the participants agreed in their opinion on videos that do contain a disclosure. Although they are aware of the advertising, the adolescents do not appreciate disclosures that too clearly convey that the video is advertising. They prefer a disclosure of the influencers themselves with an indirect choice of words (e.g. collaboration with a brand instead of advertising for a brand), a written disclosure hidden in the description underneath the video, a disclosure that is integrated within the video (e.g. a logo in the video that does
not distract) or, preferably, no disclosure at all: “But if you say this [a written disclosure of the sponsorship] in the beginning, then the whole video will be less amusing, as you will notice all things sponsored”. In their eyes, a clear disclosure emphasizes the commercial message too much and disrupts the perceived balance. They argued that a clear disclosure would overrule the entertaining element of the videos and therefore makes it less enjoyable to watch: “In my opinion it should not be too pushy, like, this is advertising, because then you will keep this in mind all the time. Just indicating is fine, but it should not be visible all the time”.
The effects of a disclosure. With regard to RQ3, it is explored how adolescents are
affected by a disclosure in terms of their brand attitude, purchase intention and attitude towards the influencer. As previously mentioned, a clear disclosure is perceived as a disturbance of the balance between the editorial and commercial content. As a result of this imbalance due to a disclosure, several forms of resistance occur, such as reactance, irritations, and avoidance.
Reactance appears when the adolescents are exposed to a suggested disclosure. They reported that the disclosure makes them feel restricted in their freedom, “Like I said, I don’t really like it [a written disclosure of sponsorship and persuasive intent] as I think you should be allowed to decide yourself what to like”.
The change of meaning principle was clearly visible, as the adolescents indicated that a clear written disclosure alters their perception of the video and the intentions of the
influencer: “First I thought it would be nice to do it [the Doritos challenge]. But now that I see the disclosure it’s like the whole video is only about Doritos and it’s not for the fun but like Doritos wants money, so you have to buy it”. Their principal attitude towards the brand and their desire to have the product is negatively changed as they are now aware of the ulterior intention of the video.
This process also affects the adolescents’ attitude towards the influencer. As one participant indicated after seeing a written disclosure of the selling intent and the source of the sponsorship: “If you put it [the disclosure] at the top [of the video] it is like: we only make these videos because we are getting paid by those brands. That’s kind of an asshole thing to do”.
It is clear that these forms or resistance are the result of the perceived predominance of the commercial message: “It seems very negative. In the way that they only make the vlog not because they like it, but because they get paid for it”.
Conclusion
The aim of the first study was to provide insights into adolescents’ understanding of influencer marketing and their level of conceptual and attitudinal persuasion knowledge and an exploration of their perceptions and the effects of sponsorship disclosures. Focus group discussions among adolescents showed that their conceptual persuasion knowledge with regard to online influencer marketing reach high levels. While previous research on
adolescents’ persuasion knowledge indicated mixed findings, this study demonstrated a high level of awareness of the sponsored content and its selling intentions. Only a thorough understanding of the persuasive intent is missing.
At the same time, the critical attitude proposed as a defence for children is lacking; instead a high level of tolerance of sponsored content is visible. This lack of scepticism is due to a perceived honesty of both the medium and the online influencers. Adolescents accepted the sponsorships as part of the influencer’s content and took the perspective of the influencer as they understand the motivations of the influencer to engage in these practices.
Considering the debate on transparency, the adolescents took a remarkable perspective. The perceived balance of the editorial and commercial content was very important. This balance is disrupted when the commercial messages are in their eyes
overshadowing the entertainment. This could happen by a disclosure of the influencer personally or a third-party disclosure. While they were aware of the sponsored content, the adolescents preferred to be ignorant so they can still enjoy the content.
Disclosures that cause an imbalance seem to activate reactance (i.e. a threat to one’s freedom to act) and the changing meaning principle. These evoke several resistance
strategies, such as avoidance and negative affect, which subsequently cause less favourable brand attitudes, purchase intentions and influencer attitudes.
Study 2: Testing the Effects of Disclosures Method
Design. To test the hypotheses, a 3 (disclosure: sponsorship, persuasive intent and no
disclosure) x 2 (explanation: with and without) experimental between-subject design was used. The participants were randomly assigned to the conditions. The experiment took place at the participants’ schools and was conducted through an online survey tool (Qualtrics).
Participants procedure. In total, 412 Dutch high school students of 12-to 16- years
old participated in the experiment. The average age was 14 (SD = .95), 56.6% were female. The participants differed in grade, whereby most were in second grade, namely 45.6% (first grade: 26,5%; third grade: 27.9%). Adolescents were recruited from three high schools in different urban and suburban areas that provided different educational levels. Thus, participants were 28.4% from vmbo (i.e. LGSE), 50% from havo (i.e. GSCE) and 21.6% VWO (i.e. A-levels). Prior to participating, institutional approval, parental approval and adolescent’s informed consent were obtained. The entire class, consisting of approximately 30 students, participated at the same time in the experiment using available computers, laptops or tablets of their high school. Depending on experimental condition, participants saw an explanation of the sponsorship mechanism or not. Next, participants were randomly assigned to one of the three disclosure conditions, which differed in disclosure type (i.e.
sponsorship, persuasive intent, no disclosure) that was shown during a 10-minute YouTube video. Participants were asked to watch the video just like they would do at home. Once they had watched the whole video, they could directly proceed with the questionnaire (see
Appendix C). Control variables regarding the YouTube video (i.e. video familiarity, attitude towards video, influencer familiarity) and attitude towards the influencer were measured. Next, questions regarding the mediating variables (i.e. conceptual and attitudinal persuasion knowledge) were asked, followed by the two other dependent variables brand attitude and purchase intention.
The questionnaire ended with a manipulation check of the disclosure type and additional control variables (i.e. product category liking, product category use, advertised brand use, medium use and demographic variables). By measuring influencer attitude before persuasion knowledge, it is ensured that it is the result of the stimulus and not influenced by possible critical attitudes towards advertising evoked by the persuasion knowledge items. Furthermore, the brand attitude and purchase intention items followed the persuasion knowledge items, as these could prime the brand in the participants’ mind when answering the latter. The experiment took about 20 minutes in total.
Stimulus materials. The stimulus material consisted of an existing YouTube video of
the well-known Dutch YouTuber Dylan Haegens. Dylan Haegens makes videos about his life and interests, which are sometimes sponsored or contain sponsored products. The video used for this study was sponsored by soft drink brand Fanta. Dylan Haegens provides its viewers with ten tips to make life easier whereby three of the ten tips involved Fanta, whereby the product and the logo are clearly visible.
In the experimental disclosure conditions, during the first 10 seconds of the video a disclosure was shown at the top of the video. The disclosure of sponsorship consisted of the sentence: “Dylan Haegens is paid to make advertising for Fanta during this video”, while the
disclosure of persuasive intent entailed: “Dylan Haegens is paid to make advertising for Fanta during this video, to make you like Fanta” (see Appendix B). The participants within the control conditions were exposed to the video without any disclosure.
Participants in the experimental explanation condition were exposed to an explanation of the sponsorship mechanism on YouTube, prior to watching the video. This explanation was based on social media guidelines propose by different international directives (i.e. Federal Trade Commission, the Dutch Advertising Committee, Interne Advertising Bureau) and made clear what the possible relations between a brand and a YouTuber could be in the context of advertising (see Appendix A).
Measures. Almost all items were measured with a 6-point scale ranging from 1 (No,
definitely not) to 6 (Yes, for sure). Children are known to have a tendency to choose a neutral mid-point when this is offered (Borgers, Sikkel, & Hox, 2004). Therefore, a 6-point scale is used. Only some control variables (i.e. video and influencer familiarity, attitude towards the video and demographics) were measured differently and will therefore be described
separately.
Conceptual persuasion knowledge. Conceptual persuasion knowledge is measured
using the scale of Rozendaal et al. (2016b) and consists of recognition of advertising, understanding selling intent and understanding persuasive intent. Recognition of advertising means that the participant was aware of advertising during the video. This was measured through three items, e.g. “Did this video contain advertising for a brand?” and “Is this video advertising?”. Understanding selling intent had one item, namely “Is the video made to make people buy Fanta?” and understanding persuasive intent three, such as “Is this video made to make people want Fanta?” and “Is this video made to make people like Fanta?”. Mean scores were calculated to create one conceptual persuasion knowledge scale (Cronbach's α = .89, M = 3.46, SD = 1.34).
Attitudinal persuasion knowledge. Attitudinal persuasion knowledge is also
measured with the scale of Rozendaal et al. (2016b) and entails the level of critical attitude towards advertising and is measured by four items inquiring about participants’ opinion about showing Fanta in the video, e.g. “Do you think that’s wrong?”, “Do you think that’s honest?” (reversed) (Cronbach’s α = .72, M = 2.88, SD = 1.02).
Brand attitude. The participants had to rate whether they agreed with six items on
their opinion of Fanta to measure their attitude towards the brand (e.g. “Do you think Fanta is tasty?”, “Do you think Fanta is nice?”) (Rozendaal et al., 2012). The mean score of the six items is used as a measurement of brand attitude (Cronbach’s α = .79, M = 4.61, SD = 0.94).
Purchase Intention. Purchase intention was measured with two items, namely
“Would you like to buy Fanta?” and “Are you going to buy Fanta?”. The mean score of the two items is used as a measurement of purchase intention (rSB = .68, M = 2.48, SD = 1.09). Higher scores indicate higher intention to buy the product.
Attitude towards the influencer. The six-item scale of Batra and Stayman (1990) is
used to ask the participants about their opinion of the influencer. For instance, “Do you think Dylan Haegens is interesting?” and “Do you think Dylan Haegens is stupid?” (reversed). Mean scores were calculated to measure attitude towards the influencer (Cronbach’s α = .92, M = 4.23, SD = 1.31).
Control variables. There were several control variables included in this study to
assure that the effects of the manipulation were not caused by differences between the experimental groups.
Considering the fact that an existing video was used and Dylan Haegens is a very popular YouTuber, we measured video and influencer familiarity by asking whether
(0 = No, 1 = Yes) and how often they watch videos of Dylan Haegens with a response scale ranging from 1 (Never) to 6 (Every day).
Furthermore, attitude towards the video was measured by asking to rate the video on a scale of 1 to 10 (which is also used with the Dutch grading system), whereby the higher the grade, the more positive someone thinks about the video.
Next, product category liking, product category use, advertised brand use and medium use were measured as control variables. Product category liking is measured by one item asking whether the participants like soft drinks. The questions to measure product category use, advertised brand use and medium use were all posed in the same way, namely how often the participants drink soft drinks, drink Fanta and watch videos on YouTube. Responses ranged from 1 (Never) to 6 (Every day).
Finally, demographic variables grade, school level, sex and age were measured.
Results
Manipulation checks. The manipulation checks for disclosure type showed that there
was a significant difference between the three conditions (i.e. control, sponsorship,
persuasive intent), χ2(10) = 338.00, p < .001. Most of the adolescents in the control condition (79.7%) reported correctly that they had not seen any of the disclosures provided in the answers. However, the disclosures shown in the experimental conditions were not that often noticed or remembered. 54.5% of the participants in the sponsorship disclosure condition recognized the disclosure, while 34.1% of the same group proclaimed not to have seen any disclosure. Within the persuasive intent disclosure condition, only 48.4% of the participants recognized the disclosure, whereas again a substantial part of these participants (34.3%) reported that they had not seen any disclosure. In sum, those exposed to a disclosure reported a significantly higher recognition of the disclosure than those who did not see a disclosure, indicating a successful manipulation.
Randomization checks. The experimental conditions (3x2) did not differ with respect
to gender, χ2(5) = 6.56, p = .256, grade, χ2(10) = 8.45, p = .585, age, F(5, 406) = 2.07, p = .068, educational level, F(5, 406) = 1.07, p = .376, influencer familiarity, F(5, 406) = .35, p = .885, medium familiarity, F(5, 406) = .78, p = .568, attitude towards the video, F(5, 406) = 1.78, p = .117, brand familiarity, F(5, 406) = 1.05, p = .388, brand use, F(5, 406) = 1.42, p = .216, product category use, F(5, 406) = 1.93, p = .089 and product category attitude, F(5, 406) = .55, p = .742. This indicates successful randomization.
Hypotheses test. To test H1 to H5, serial mediation analyses were conducted using
PROCESS Model 6 (Hayes 2013). Disclosure conditions were recoded into dummy variables as there are three conditions. The dummy variables were used as the independent variable while the other conditions were chosen as covariate. Conceptual persuasion knowledge is used as the first mediator and attitudinal persuasion knowledge as the second. Brand attitude, purchase intention and attitude towards the influencer were taken as dependent variables in separate analyses. Disclosure of sponsorship is compared with the control condition, whereas disclosure of persuasive intent is compared with disclosure of sponsorship. Finally,
PROCESS Model 1 is used to test H6 and H7 for moderation between disclosure type and sponsorship explanation on conceptual persuasion knowledge. Means and standard deviations of conceptual persuasion knowledge, attitudinal persuasion knowledge and the dependent variables for the different disclosure conditions are shown in Table 1.
Effects of disclosures on conceptual persuasion knowledge and attitudinal persuasion knowledge. With respect to H1, the analyses showed significant differences in
conceptual persuasion knowledge between a disclosure of sponsorship and no disclosure. Adolescents exposed to a disclosure of sponsorship activated significantly more conceptual persuasion knowledge compared to adolescents who were not exposed to a disclosure at all, b = 3.0, SE = 0.10, t(409) = 29.24, p < .001. Moreover, a disclosure of persuasive intent evoked
higher levels of conceptual persuasion knowledge than a disclosure of sponsorship, b = 0.37, SE = 0.16, t(409) = 2.25, p = .025. Logically, a disclosure of persuasive intent did trigger more conceptual persuasion knowledge than no disclosure at all, b = .93, SE = 0.16, t(409) = 5.97, p < .001. Results thus provide support for H1, as exposure to a disclosure of
sponsorship results in higher scores of conceptual persuasion knowledge compared to no disclosure. Additionally, a disclosure of persuasive intent does activate higher levels of conceptual persuasion knowledge than a disclosure of sponsorship. Additionally, descriptive analyses indicate that only adolescents exposed to a disclosure of persuasive intent activated conceptual persuasion knowledge above the mid-point, whereas the conceptual persuasion knowledge of adolescents exposed to a disclosure of sponsorship remained around the mid-point (see Table 1).
With respect to H2, the analyses showed that conceptual persuasion knowledge has a significant effect on attitudinal persuasion knowledge, b = .13, SE = 0.04, t(409) = 3.43, p < .001. This indicates that when someone activates higher levels of conceptual persuasion knowledge, this results in higher levels of attitudinal persuasion knowledge. The data thus support H2.
Table 1
Mean Scores and Standard Deviations for Disclosure Conditions
Control Disclosure Sponsorship Disclosure Persuasive Intent Conceptual PK 3.00a (1.29) 3.57b (1.37) 3.93c (0.19) Attitudinal PK 2.87 (1.00) 2.95 (1.06) 2.84 (1.00) Brand Attitude 4.59 (0.94) 4.67 (0.87) 4.56 (1.02) Purchase Intention 2.52 (1.14) 2.51 (1.11) 2.39 (1.04) Attitude Towards the
Influencer
4.16 (1.34) 4.16 (1.38) 4.41 (1.17)
Effects of disclosures on brand attitude, purchase intention and attitude towards the influencer via persuasion knowledge.
Brand attitude. With respect to H3, the analyses showed that there is a significant negative effect of attitudinal persuasion knowledge on brand attitude, b = -.32, SE = 0.04, t(407) = -7.29, p < .001. This indicates that higher levels of attitudinal persuasion knowledge result in a less favourable brand attitude. H3a can thus be accepted.
Furthermore, disclosure type affected brand attitude indirectly via conceptual and attitudinal persuasion knowledge. Adolescents exposed to a disclosure of sponsorship showed higher levels of conceptual persuasion knowledge which led to higher levels of attitudinal persuasion knowledge, which in turn led to a less favourable brand attitude, compared to adolescents exposed to no disclosure (b = -.02, SE = .01, [-.05; -.01]). Moreover, a disclosure of persuasive intent resulted in a less favourable brand attitude via conceptual and attitudinal persuasion knowledge, than solely a disclosure of sponsorship (b = .02, SE = .01, [.05; -.00]). This also means that a disclosure of persuasive intent evokes a less positive brand attitude than no disclosure (b = -.04, SE = .02, [-.08; -.02]). The data thus supports H3b.
Purchase intention. With respect to H4, the analyses showed that there is no direct significant effect of attitudinal persuasion knowledge on purchase intention (b = .08, SE = .05, t(407) = 1.43, p = .153). This indicates that level of activated attitudinal persuasion knowledge does not influence adolescents’ desire to buy a product. H4a will thus be rejected.
The analyses also showed no significant indirect effect of either a disclosure of
sponsorship compared with no disclosure on purchase intention via conceptual and attitudinal persuasion knowledge (b = .01, SE = .01, [-.00; .02]) nor of a disclosure of persuasive intent compared to a disclosure of the sponsorship (b = .00, SE = .04, [-.00; .02]). H4b is therefore rejected.
Attitude towards the influencer. Concerning H5, the analyses showed that attitudinal persuasion knowledge significantly affected adolescents’ attitude towards the influencer, b = -.30, SE = .06, t(407) = .39, p < .001. In other words, higher levels of attitudinal persuasion knowledge mean a less positive attitude towards the influencer. H5 can therefore be accepted.
Moreover, there is a significant indirect effect of disclosure type on attitude towards the influencer. A disclosure of sponsorship led to higher levels of conceptual persuasion knowledge which led to higher levels of attitudinal persuasion knowledge, which in turn led to a less favourable attitude towards the influencer, compared to adolescents exposed to no disclosure (b = -.02, SE = .01, [-.05; -.01]). Adolescents exposed to a disclosure of persuasive intent showed an even less positive attitude towards the influencer via conceptual and
attitudinal persuasion knowledge, compared to adolescents exposed to a disclosure of sponsorship (b = -.01, SE = .01, [-.04; -.00]). Logically, they also showed a lower attitude towards the influencer compared to adolescents who were not exposed to a disclosure (b = -.04, SE = .02, [-.08; -.01]). H5c is thus supported by the data.
Interaction effect of sponsorship explanation. To test H6 and H7, a moderation
analysis (Hayes, 2013; Model 1) was conducted, showing that explanation did not moderate the effect of a disclosure of sponsorship compared to no disclosure on conceptual persuasion knowledge, b = -.12, SE = .28, t(407) = -.44, p = .659. Neither was there a significant
interaction effect between sponsorship explanation and a disclosure of persuasive intent compared to a disclosure of sponsorship, b = -.15, SE = .29, t(407) = -.51, p = .612.
Apparently, an additional explanation of the sponsorship mechanism did not strengthen the effects of either a disclosure of sponsorship nor a disclosure of persuasive intent. H6 is therefore rejected. As there is no moderation effect, the proposed moderated indirect effect cannot be present. Therefore, H7a, H7b and H7c are rejected too.
Conclusion
The first objective of the second study was to test the effect of different disclosure types on brand attitude, purchase intention and attitude towards the influencer via conceptual and attitudinal persuasion knowledge. The second objective was to examine whether an additional explanation of the sponsorship mechanism would moderate this effect.
First, this study demonstrated that an adolescent exposed to a disclosure of sponsorship showed higher levels of conceptual persuasion knowledge than adolescents exposed to sponsored content without a disclosure. However, it should be noted that while conceptual persuasion knowledge increased after exposure to a disclosure of sponsorship, it is still around the mid-point of the scale. This indicates that a disclosure of sponsorship did not make adolescents aware of the sponsorship, but just a little less unaware than adolescents who were not exposed to a disclosure. Disclosing persuasive intent results in a higher level of conceptual persuasion knowledge than solely a disclosure of sponsorship, which does rise above the mid-point. This indicates that disclosing the persuasive intention is necessary in order to make adolescents really aware of the sponsored content and its intentions.
Furthermore, higher levels of conceptual persuasion knowledge did subsequently evoke higher levels of attitudinal persuasion knowledge. This indicates that when an adolescent’s understanding of the sponsorship is triggered, a more critical attitude towards the sponsored content will follow. In turn, this turned out to limit the effects of the sponsored content on brand attitude and attitude towards the influencer. Remarkably, this is not the case for the effects of purchase intention. Attitudinal persuasion knowledge does not predict whether adolescents want to buy or have the sponsored product.
Finally, an additional sponsorship explanation does not alter the effects of a sponsorship disclosure on activation of conceptual persuasion knowledge.
