1
Effect of Disclosure Explicitness on Skepticism and Brand
Attitude under moderation of Persuasion Knowledge
Jingyue Liu
11647132
Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science
Supervior: Saar Mollen
28-06-20192 Abstract
Sponsorship disclosures on YouTube differ noticeably in terms of explicitness, which means that there is no uniform way in which YouTube users are informed of persuasive attempts by vloggers in their in videos. Past studies show conflicting findings regarding the direction of effects of disclosures that vary in explicitness on skepticism and brand attitude. According to persuasion knowledge model and reactance theory, YouTube users varying in high or low persuasion knowledge of YouTube sponsored content may interpret the persuasive attempt significantly different, resulting in more or less skepticism and positive brand attitude. This study thus aims at solving this puzzle by bringing YouTube users’ prior persuasion
knowledge into the relationship, as a moderator in the effect of disclosure explicitness on skepticism. A 3 (disclosure explicitness: explicit vs. implicit vs. no disclosure) x 2 (persuasion knowledge: low vs. high) between-subjects online experiment is conducted. Results show that there is no significant interaction between disclosure explicitness and persuasion knowledge on skepticism. However, more skepticism does result in less positive brand attitude. Additionally, this research finds a direct effect of disclosure explicitness on brand attitude, that more explicit disclosure will decrease YouTube users’ positive brand attitude.
In contrast to the expectations the findings do not show persuasion knowledge plays an important role on how disclosure explicitness affects skepticism, but this research shed a light on disclosure explicitness. And this result suggests marketing practitioners that using a more explicit disclosure will not increase YouTube users’ skepticism. Besides, the clearer insight on positive brand attitude decreasing for more explicit disclosures might encourage marketeers to use less explicit disclosures. More detailed standardization is thus needed.
3 Introduction
YouTube is the second most visited website in the world, with 5 billion videos watched and 30 million visitors per day (“MerchDope”, 2019). Via YouTube, companies can place ads directly in videos known as In-Stream ads, or they may sponsor YouTube content creators (“YouTubers”) to place and introduce products in their own videos. Also known as sponsored content (Wu, 2016). Although the sponsored content by YouTubers can raise more
engagement with audience compared to traditional marketing sponsorship, it raises ethical concerns that audience are not informed of the advertising nature when the message of sponsorship is not explicitly conveyed in the videos they consumed (Wojdynski and Evans, 2016). The only effective way for an audience to identify sponsored content blended in non-advertising content on YouTube is a disclosure (Amazeen & Wojdynski, 2018). Sponsorship disclosure refers to the explanation from content creators to inform consumers about the commercial intent of the content published and should aid consumers in distinguishing sponsored content from non-commercial editorial content (Boerman, Willemsen, & Van Der Aa, 2017).
Although Federal Trade Commission (FTC) has required a disclosure placed prominently in all kinds of sponsored content, the whole industry still lack a clear
standardization on the explicitness of disclosure (Amazeen & Wojdynski, 2018). While some YouTubers explicitly disclose by oral statement in beginning of their videos as a forewarning, others place more implicit disclosures by text information in the description box below their videos, and others do not even disclose sponsorships at all. Different ways of disclosing advertising content may be more or less successful in actually informing people that it is advertising and research shows that different levels of disclosure explicitness affect consumers' skepticism and brand attitudes significantly (e.g., Boerman, Reijmersdal, & Neijens, 2012; Tutaj & Van Reijmersdal, 2012; Van Reijmersdal, Tutaj, & Boerman, 2013;
4 Matthes, Schemer, and Wirth, 2007; Campbell & Marks, 2015; Han, Drumwright, & Goo, 2018; Carr & Hayes, 2014; Dekker & Van Reijmersdal, 2013). Disclosure explicitness refers to how clear a disclosure informs the persuasive nature to its audience. Understanding the effects of disclosure explicitness on YouTube is important because it will give both
practitioners and regulators an insight into practical application and improvement of relevant laws.
Previous studies has shown the effectiveness of disclosure on brand attitude can differ in formats of disclosures, namely text or logo placement (Boerman, van Reijmersdal & Neijens, 2015) and disclosure duration (Boerman et al., 2012). These characteristics of disclosure are more or less related to the explicit level of a disclosure and the results demonstrate a more explicit disclosure will lead to less positive brand attitude. However, there are also opposite findings that when people are provided a more explicit disclosure, their attitude toward the sponsored brand will be more positive via a decreased skepticism level due to their appreciation to transparency (Han et al.,2018; Carr & Hayes, 2014; Campbell & Marks, 2015).
Whether explicit versus more implicit disclosures may result in more or less skepticism and more negative or positive brand attitudes may depend on the personal prior persuasion knowledge. Persuasion knowledge refers to an individual capability which is established by consistent accumulation of recognizing persuasive commercial content (Friestad & Wright, 1994). The persuasion knowledge model (PKM) describes how individual’s persuasion knowledge develops and how people use that knowledge to refine their attitudes toward products and brands. The developing process of persuasion knowledge and using that knowledge to form attitude is contingent upon prior experience with similar content (Friestad & Wright, 1994). As the marketeers are consistently working on mix sponsored content into non-commercial content to make the commercial nature less
5 noticeable, some unexperienced consumers might be less familiar with certain new format such as on YouTube and thus fail in recognize the cues signifying sponsorships and the commercial intent behind (Evans & Park, 2015; Wojdynski, 2016). Differing levels of persuasion knowledge may therefore explain when explicit disclosure results in higher skepticism level and less positive brand attitude than implicit disclosure or no disclosure, and when it generates lower skepticism level and a more positive brand attitude.
According to the PKM, people increase the use of their persuasion knowledge to cope with the persuasion when noticing something unexpected in the sponsorship format (Friestad & Wright, 1994). YouTube users with low persuasion knowledge may not expect a
sponsorship in the YouTube video they consume as they are less familiar with these persuasion tactics. When an explicit disclosure successfully informs them this unexpected sponsorship format, the explicit disclosure, as a forewarning, will therefore generate reactance to persuasion as a persuasion coping strategy (Zuwerink & Devine, 2000). The reactance of YouTube users with low persuasion knowledge may not be activated when an implicit disclosure or no disclosure is provided as YouTube users may be not aware of the sponsorship in sponsored content. Therefore, for YouTube user with low persuasion knowledge, the skepticism level as a result will probably be higher and attitude towards sponsored brand will be less positive when they are exposed to an explicitly stated disclosure than implicit disclosure or no disclosure.
However, for YouTube users with high persuasion knowledge, their prior knowledge allows them to more easily recognize the persuasive intent in a sponsored content (Amazeen & Wojdynski, 2018), so they would expect an explicit disclosure of the sponsorship in the video. When an explicit sponsorship disclosure is provided, which is within their
expectations, they will not cope with the persuasive intent and instead they might appreciate the transparently stated sponsorship (Carr & Hayes, 2014). In this case, an explicit disclosure
6 is likely to lower the skepticism level compared with implicit disclosure and no disclosure, resulting in more positive brand attitude.
The assumptions above predict completely opposite directions of how disclosure explicitness affects skepticism for people with high or low persuasion knowledge. This research therefore will investigate that to which direction disclosures varying in explicitness on YouTube will affect skepticism level and brand attitude, for both YouTube users with high persuasion knowledge and YouTube users with low persuasion knowledge.
Theoretical Framework
Effect of Disclosure Explicitness on Skepticism and Brand Attitude
Sponsored content is a trendy new marketing strategy and the spending on it is expected to grow (Adyoulike, 2015). An integration of sponsored content and editorial content on YouTube leaves users a problem of identifying the existence of a commercial intent, hence sponsorship disclosure is necessary to help YouTube users differentiate
sponsored content from editorial content. It aims at guaranteeing all the users understand the persuasive attempt in a sponsored content, no matter people with high or low persuasion knowledge. However, even if a disclosure is noticed, many people do not understand that ‘sponsored’ indicates the content is paid advertising (Austin and Newman, 2015; Gilley, 2013; Lazauskas, 2014; Wojdynski, 2016). This is because the level of disclosure explicitness can differ greatly, which affect skepticism and subsequently brand attitude differently.
The findings of previous studies regarding the effects of disclosure explicitness on skepticism and brand attitude were contradictory. On the one hand, some studies found a more explicit disclosure would develop distrusting feeling, leading to less positive brand attitude (Boerman et al, 2017; Sweetser et al, 2016; Tessitore and Geuens, 2013). For instance, in a research (Boerman et al., 2015) which explored the effect of disclosure
7 placement type on brand attitude, the results showed that a combination of disclosure text and a product placement (“PP”) logo was more effective in generating a negative brand attitude compared to mere logo exposure or a mere disclosure text. Taken together these findings suggest that the more sufficient and more clear disclosure will generate more negative brand attitude. To summarize these findings, the characteristics of disclosures examined in the three studies can be categorized as low versus high in explicitness and the results reveal that more explicit disclosures will lead to less positive brand attitudes.
On the other hand, several studies found that a more explicit disclosure leaded to less skepticism and subsequently more positive brand attitudes. Han et al. (2018) found that a more explicit disclosure of sponsored content actually resulted in less skepticism towards advertisers by increasing consumers’ appreciation on commercial’s transparency. In line with Han et al. (2018), by examining different disclosures on blogs, Carr and Hayes (2014) found a significant positive effect of disclosure explicitness on credibility of influencer’s opinion and in turn on the attitude towards the sponsored product.
However, there is also a study which indicates that disclosure explicitness had no effect on brand attitudes. Dekker and Van Reijmersdal (2013) examined three types of disclosure content, which differed in explicitness regarding the language used to help participants understand the persuasive attempt. It was expected that the more explicit disclosure would result in less acceptance of the sponsorship claims and negative attitudes. Nevertheless, no significant difference in brand attitude was found among these three disclosures.
To sum up, all the different findings above indicate that the direction of how explicit disclosure affects skepticism and brand attitude is not clear yet. And this is what this study aims to look into.
8 Effect of Disclosure Explicitness on Skepticism and Brand Attitude, via Persuasion knowledge model and Reactance theory
As the existing studies provided contradictory findings, I assume that the conflicting findings of to which direction disclosure explicitness affects skepticism and brand attitude may be explained by persuasion knowledge based on the persuasion model and reactance theory. Subsequently, more skepticism results in less positive brand attitude. To clarify, skepticism in the present study is defined as the tendency towards disbelief of advertising claims
( Obermiller & Spangenberg, 1998).
How disclosure explicitness affect skepticism, explained by Persuasion knowledge model?
The persuasion knowledge model (Friestad and Wright 1994) explains that possessing
persuasion knowledge of persuasive commercial tactics help people to identify the persuasive attempt and further analyze brand as a coping behavior. Persuasion knowledge could be divided into two types: conceptual and attitudinal persuasional knowledge. Conceptual persuasion knowledge could be seen as the recognition of sponsorship. Attitudinal persuasion knowledge refers to general critical feeling of skepticism and distrust towards a sponsored content (Boerman et al., 2017).
Previous studies has found that disclosure can activate both types of knowledge, in the order of the conceptual persuasion knowledge being firstly activated (Boerman et al., 2012; Boerman et al., 2017). More specifically, a more explicit disclosure will activate more attitudinal persuasion knowledge, which is supported by the study of Boerman et al. (2017). In this study, celebrity-sponsored ads and ads posted by brand itself were compared, with the former being more explicitly informing persuasive attempt. The results showed that the disclosure of celebrity-sponsored ads activated more attitudinal persuasion knowledge than disclosure of ads posted by brand.
9 Moreover, majority of research on persuasion knowledge suggested that persuasion knowledge and skepticism invariably go hand in hand (Isaac & Grayson, 2017; Yoo & Macinnis, 2005) Therefore, an explicit disclosure may activate more attitudinal persuasion knowledge than implicit disclosure and no disclosure, because it informs consumers of the persuasive attempt more clearly. And the attitudinal persuasion knowledge further motivate consumers to be skeptical and analyze the brand unfavorably. It is thus reasonable to assume that the more explicit disclosure can increase skepticism via the activation of persuasion knowledge.
How disclosure explicitness affect skepticism, explained by Reactance theory? According to the reactance theory (Brehm & Brehm, 1981), reactance is a motivational state that occurs when freedom is threatened. Being exposed to explicitly stated disclosure may cause people to more strongly feel the threat of limiting their experienced freedom to make their own decision (Brehm, 1966). As a result, reactance towards the ad claims and the resistance to the persuasive attempt will be activated (Fransen et al., 2015). This could mean that people’s concerns of deception and skepticism towards the advertising message will increase, or they will start to contest persuasive messages by counterarguing and distrusting (Wood and Quinn, 2003). Therefore, overall, disclosure explicitness will have a positive effect on skepticism via activating people’s reactance to persuasion.
How skepticism affects brand attitude?
A higher level of skepticism will lead to a more negative attitude towards the brand.
Skepticism shows a tendency of not believing advertising, the disbelief in specific ad claims follows directly. Normally speaking, ad claims are positive, less belief in those claims should result in less positive attitudes towards the product or the brand (Obermiller &
Spangenberg,1998; Tutaj & Van Reijmersdal, 2012; Yoo & Macinnis, 2005). Also, from the perspective of brand attitude formation process, it is higher credibility of a sponsored content
10 that will increase audience’s positive feelings and results in more favorable brand attitudes (Yoo & Macinnis, 2005).
As disclosure explicitness affects skepticism and skepticism influences brand attitude, skepticism is thus highly possible to be a mediator for the effect of disclosure explicitness on brand attitude. This study is going to investigate whether such a mediation exists.
Effect of Persuasion Knowledge and Disclosure Explicitness on Skepticism via PKM As discussed above, the findings of to which direction disclosure explicitness affects skepticism and brand attitude were conflicting, and the lack of the consideration of individual differences in the possession of persuasion knowledge might be an important reason for the controversy.
John (1999) pointed out that persuasion knowledge refers to an ability of
distinguishing persuasive message from editorial message and this kind of ability varies from person to person (Amazeen & Muddiman, 2017; Campbell et al., 2013; Iversen & Knudsen, 2017). Therefore, even if facing the same disclosure, people’s level of recognition and
understanding could be very different. In addition, as discussed before, persuasion knowledge and skepticism go in hand in hand (Isaac & Grayson, 2017; Friestad and Wright, 1994), so it is reasonable to argue that the degree of skepticism caused by a certain disclosure
explicitness could be different among individuals, due to the diverse level of the personal possession of persuasion knowledge. In other word, this study assumes that persuasion knowledge is a moderator of the effect of disclosure explicitness on skepticism.
As for how persuasion knowledge affects people processing the advertising
message and analyzing the sponsored brand, Friestad and Wright (1994) indicated that people interpreted the persuasive attempt based on their knowledge-based expectations. To cope with persuasion, consumers will anticipant a sponsorship format in the content they consume.
11 In other words, the interpretations of persuasive attempt and message processing can differ greatly when the sponsorship format is beyond YouTube users’ expectations or not, resulting in high or low skepticism level. As people with high or low persuasion knowledge vary in knowing the persuasion tactics, they are likely to have different expectations in the existence and explicitness of disclosure.
On the one hand, for people with little knowledge of sponsored content on YouTube, they are likely less able to detect sponsored content without an explicit disclosure. Without knowing there is a sponsorship, they would thus not expect a disclosure in the sponsored content they consume. When an explicit disclosure is provided, the persuasive attempt, which is out of their expectation, will generate reactance to persuasion as a coping behavior (Brehm, 1966; Hass & Grady, 1975). The reactance aroused by a forewarning will further motivate people to counterargue the message (Petty & Cacioppo, 1979) and distrust the message as audience felt that they were misled, resulting in an increase in skepticism and negative evaluation in sponsored brand (Darke et al., 2018). However, when there is no disclosure provided to YouTube users with low persuasion knowledge, their reactance to persuasion and skepticism will not be activated because they are likely to fail in knowing the existence of sponsorship in sponsored content. In short, for YouTube users with low persuasion
knowledge, an explicitly stated disclosure will probably make them being more skeptical due to the activation of their reactance to persuasion, resulting in evaluating the brand less
positively.
On the other hand, people with a high level of PK tend to easily detect whether
YouTubers who discuss certain products in their videos are rewarded by a brand or not. Their high level of knowledge in persuasion tactics will help them to some degree figure out the potential persuasive attempt in a YouTube video, even when the disclosure is not provided. Hence, they would expect an explicitly stated disclosure about the sponsorship they assumed.
12 When an explicitly stated disclosure is not provided, which is out of their expectation, a feeling of deception can be raised by the hidden persuasive attempt. As a result, the high persuasion knowledge will increase the likelihood of being skeptical towards the sponsored brand (Boush, Friesstad, & Rose, 1994). However, on the other hand, people with high persuasion knowledge will appreciate the explicit disclosure of sponsorship and therefore decrease the level of skepticism because of the appreciation of transparency. Hence, I assume that explicit disclosure will generate less skepticism among YouTube users with high
persuasion knowledge, with comparison to implicit disclosure or no disclosure, resulting in less skepticism and a more positive brand attitude.
To conclude, the present research aims to shed more light on the YouTube users’ persuasion knowledge of sponsored content, as a moderator for the effect of disclosure explicitness on skepticism. More specifically, I will examine how the effect of disclosure explicitness on YouTuber users’ s level of skepticism differs between people knowing sponsored content on YouTube well (i.e.: people with high persuasion knowledge) and people knowing little (i.e.: people with high persuasion knowledge), and whether skepticism will lead to a negative effect on brand attitude.
H1: The effect of explicitness of sponsorship disclosure on skepticism will be moderated by consumers’ persuasion knowledge. And skepticism mediates the effect of sponsorship
disclosure on brand attitude. For YouTube users with high persuasion knowledge, the explicit sponsorship disclosure will result in lower skepticism than an implicit disclosure or no disclosure, resulting in a more positive brand attitude. Whereas for YouTube users with low persuasion knowledge, the explicit sponsorship will generate higher skepticism compared to an implicit disclosure or no disclosure.
13 The whole moderated mediation model illustrating all the hypotheses is presented in the Fig. 1 below.
Fig. 1 Proposed moderated mediation model: Effect of disclosure explicitness on brand attitude via the mediating effect of skepticism, moderated by persuasion knowledge
Methods
Design and participants
A 3 (disclosure explicitness: explicit vs. implicit vs. no disclosure) x 2 ( persuasion
knowledge: low vs. high) between-subjects online experiment was conducted. A total of 271 participants started the online experiment and 196 participants who fully completed the experiment was kept after the exclusion after the exclusion of missing value. Then based on the manipulation check results, near half of participants who were exposed a disclosure thought they didn’t see any disclosure when they recalled if they noticed a sponsorship statement. This percentage was aligned with previous study (e.g., Boerman et al., 2017; Boerman et al., 2012; Campbell, Mohr, and Verlegh, 2013). Meanwhile, a higher percentage (63%, 44 out of 70) of participants who were not exposed to a disclosure recalled they saw a disclosure or they were not sure about it. In line with prior study by Boerman et al. (2017), I Explicitness of Sponsorship
disclosure: explicit disclosure, implicit disclosure, no disclosure Brand Attitude Skepticism Persuasion knowledge H1 - -
14 excluded participants who thought they saw a disclosure when they were not exposed to any disclosure and who failed in recognizing the disclosure when they were exposed to a
disclosure. In total, 72 participants were excluded and the remaining 124 participants are the valid dataset for all the following analyses (no disclosure n = 26, implicit disclosure n = 48, explicit disclosure n = 50). The mean age of participants was 27.42 (SD = 10.09), over half of participants were female (64.5 %), with 79.1% of the participants having a Bachelor degree or higher. Most of participants watched YouTube videos every day or at least once a week (72.5%).
Procedure
Participants were asked to take part in an online study on YouTube video research. They were told that this study was interested in their opinions about the healthy lifestyle presented in a YouTube video to prevent them not behaving as naturally as in real life after knowing the real purpose.
Firstly, demographic and background variables were assessed (i.e., age, education level, YouTube watching frequency and preferred video topic) were measured in the beginning of the experiment Then, participants were randomly assigned to one of three conditions of sponsorship disclosure. The chosen YouTube video “How to start a healthy lifestyle” pertained to healthy lifestyle and fitness and was sponsored by a workout App, 8 fit. After watching the video, participants were asked to answer questions that measured their skepticism regarding tendency to distrust in sponsorship claims, followed by attitude towards the brand recommended in the video, and individual level of persuasion knowledge.
Persuasion knowledge questions were presented after questions of skepticism and brand attitude. This was done to ensure that participant's skepticism and brand attitude would not be affected by the persuasion knowledge questions. After measuring persuasion knowledge,
15 manipulation check items were asked. The experiment ended with participants’ remarks or comments, and an elaborate debriefing, informing participants the real purpose of the study.
Manipulation
In all conditions participants were exposed to a YouTube video on a healthy lifestyle and fitness by YouTuber Valeria Lipovetsky. The video was adapted to reflect the three
conditions: explicit disclosure, implicit disclosure and no disclosure. The only difference was the explicitness of disclosure In the explicit disclosure condition the video lasted almost three minutes and included a twenty seconds disclosure. The YouTuber explicitly makes the following statement in the beginning of the video: “I want to say huge thank you to 8fit for sponsoring and collaborating with me on this video. 8fit is a mobile app that lets you customize your meals and workouts.” -- together with a textual disclosure “How to start a healthy lifestyle | ft. 8fit” presented in the title of the video directly under the video. “Ft.” indicated the sponsored nature of a YouTube video and this was a common way of sponsorship disclosure on YouTube platform. In the implicit disclosure condition, the YouTube video lasted two and half minutes without any oral disclosure statement from the YouTuber in the video. A textual sponsorship disclosure “How to start a healthy lifestyle | ft. 8fit” in video title was presented right under the video. The no disclosure condition, as a control condition, did not have either disclosure by the YouTuber in the video, or disclosure of “ft. 8fit” in the video title.
Measures
Skepticism. Nine items adopted from Obermiller and Spangenberg (1998 ) were used to measure skepticism. Participants were asked to indicate their agreement with the following statements: “We can depend on getting the truth in this YouTube video”, “The aim of this
16 YouTube video is to inform the consumer,” “ I believe this YouTube video is informative,” “ This YouTube video is generally trustworthy,” “ This YouTube video is a reliable source of information about the quality and performance of products,” “ This YouTube video is truth well told,” “ In general, this YouTube video presents a true picture of the product being advertised,” “ I feel I have been accurately informed after viewing this YouTube video,” and “ This YouTube video provides consumers with essential information.” The scale was
reliable (Cronbach’s alpha = .93). All scores in the nine items are recode reversely so that the higher score means more skepticism. The overall average score of the recoded nine items is the skepticism level (M = 3.81, SD = 1.24).
Brand attitudes. Nine items were selected from the scale developed by Spears and Singh (2004) after modifying and removing the irrelevant items, as follows: appealing / unappealing; pleasant / unpleasant; favorable / unfavorable; distinctive /not distinctive; useful/useless; exciting/ dull; superior / inferior; lacks important benefits/offers important benefits; advisable to choose /not advisable to choose. Similar to most prior studies, I used a seven-point semantic differential format in the scales (1 = negative attitudes, 7 = positive attitudes). The reliability test shows the scale is highly reliable (Cronbach’s alpha = .93). The mean of all items is the score of brand attitudes and the higher score means more positive brand attitudes (M = 4.43, SD = 1.08).
Persuasion knowledge The persuasion knowledge of participants is measured in the survey after the measurements of skepticism and brand attitude. Five subscales are used to test participant's trait persuasion knowledge following a scale developed by Boerman, Van Reijmersdal, Rozendaal, and Dima (2018). The subscales are from the conceptual dimension measuring a) the recognition of sponsored content and b) commercial source, c)
17 content. The recognition of sponsored content was measured by whether participants saw a brand logo in a poster (0 = no, 1 = yes) and how they describe the concrete sponsorship content. This item was coded into a 5-point scale firstly based on the preciseness of their description, following the measurement developed in study by Boerman et al. (2018). It was later recoded into 7-point Likert scale. The final score of persuasion knowledge scale for each participant was the average score of the five subscales. The reliability tests show the
individual scales were fairly reliable with all their measuring items (c: Cronbach’s alpha = .688; d: Cronbach’s alpha = .838; e: Cronbach’s alpha = .883). The scale of persuasion knowledge was more reliable if the recognition item (a) was deleted compared with the persuasion knowledge scale including all the items (Cronbach’s alpha = .881 vs. Cronbach’s alpha = .872). Therefore, the other four subscales are kept except the recognition subscale for the scale of persuasion knowledge and participants scoring higher have higher persuasion knowledge (M = 5.05, SD = 0.81).
Control Variable To exclude the influence of other characteristics of participants, some demographic information was measured as well in the beginning of the experiment. Participants were asked to provide their age in a text box and choose their gender (Female/ Male/ Other), education level (No schooling completed/ Elementary school/ Secondary school/ Bachelor’s degree/ Master’s degree/ Doctor degree and higher/ Other), YouTube watching frequency (Everyday/Several times a week/ Once a week/ Several times a month/ Several times a year/ Never), and whether like fitness videos by checking the preferences among the options of YouTube video types)
Manipulation Check At the end of the online experiment, participants were asked to choose if they noticed a sponsorship statement when they were watching the YouTube video. (Yes/ No/ Not sure) and if yes, they needed to explain where they saw it (In the video/ under the video/ Both). In line with previous research (Boerman et al., 2017), participants in the
18 control condition who falsely recognized a disclosure and participants in the experimental condition who didn’t recognize a disclosure were excluded.
Analyses plan
To check whether the groups are comparable with respect to certain demographic
characteristics and background variables, randomization check will be done. Chi-square tests will be used for the variables in nominal level (gender, education level, YouTube watching frequency and whether like fitness videos), and an one-way ANOVA will be used for age, the interval level variable. To test the whole moderated mediation model, PROCESS Model 7 will be used, with disclosure explicitness as independent variable, skepticism as mediator and brand attitude as dependent variable. As this research is highly interested in individual
differences of consumers’ ability to recognize sponsored content besides the whole persuasion knowledge level, two PROCESS model 7 analyses will be conducted with the recognition subscale as moderator in one analysis and persuasion knowledge in another.
Result
Randomization check
The results reflect that the three groups did not differ in age, F ( 2, 121) = -1. 32, p = .270, gender, χ2 (2) = 1. 27, p = .530, education level, χ2 (8) = 5.40, p = .714, YouTube watching frequency, χ2 (10) = 6.80, p = .748 and whether like fitness videos, χ2 (2) = 3.42, p = .181. Due to the non-significant differences, this research did not include any of these control variables into covariates the following analyses.
Moderated mediation effect of disclosure explicitness on brand attitude, moderated by persuasion knowledge
19 To test the moderated mediation effect proposed in H1 the PROCESS macro (Hayes, 2015) was used (Model 7). Disclosure explicitness was entered as the independent variable, persuasion knowledge as a moderator, skepticism as mediator, and brand attitude as the dependent variable. Both the direct and indirect effects in the moderated mediation model were tested. By applying the multi-categorical setting for the disclosure explicitness variable, both explicit disclosure condition and implicit disclosure condition were compared with the control condition (no disclosure condition) separately. The sample was bootstrapped with 5,000 replications to examine 95% confidence intervals for all the relationships.
Disclosure explicitness did not show a significant main effect on skepticism (explicit disclosure vs. no disclosure: b = 2.12, SE = 1.81, p = .244, 95% CI [ - 1.46, 5.70]; implicit disclosure vs. no disclosure: b = - 1.46, p = .384, 95% CI [ - 0.48, 1.85]. There was also no significant main effect of skepticism on skepticism found, which means skepticism level did not differ between people with high persuasion knowledge and people with low persuasion knowledge (b = 0.11, SE = 0.24, p = .650, CI[ - 0.37, 0.58]). The predicted interaction effect between disclosure explicitness and persuasion knowledge on skepticism was nonsignificant (explicit disclosure vs. no disclosure: b = - 0.45, SE = 0.35, p = .207, 95% CI [ - 1.15, 0.25]; implicit disclosure vs. no disclosure: b = - 0.30, SE = 0.33, p = .358, 95% CI [ - 0.35, 0.95]). This indicates that persuasion knowledge did not moderate the effect of disclosure
explicitness on skepticism at all.
Upon inspection of the direct effects, as predicted participants less skeptical towards the sponsored brand had more positive brand attitude than participants more skeptical towards it, (b = - 0.60, SE = 0.06, p < .001, 95% CI [ - 0.72, - 0.49]), which was identified at the second stage of the moderated mediation model. Additionally, there was a significantly negative direct effect of disclosure explicitness on brand attitude comparing the no disclosure condition with explicit disclosure (b = - 0.41, SE = 0.19, p = .031, CI [ - 0.79, - 0.04]). The
20 difference between implicit disclosures and no disclosures was marginally significant (b = - 0.33, SE = 0.19, p = .089, CI [ - 0.70, 0.05]). This means that people who watched a
sponsored YouTube video with explicit disclosure had a less positive brand attitude than people watching it without any disclosure. And that there was a trend that more explicit disclosure leads to less positive brand attitude.
The indirect effect of disclosure explicitness via skepticism on brand attitude was not significant, no matter participants were with high or low persuasion knowledge (see Table 1). There was no evidence of moderated mediation model as predicted in the hypothesis (index of moderated mediation = 0.270, boot SE = 0.25, CI [ -0.33, 0.68]).
Therefore, the results showed that YouTube users’ skepticism level did not
significantly differ, when they watched a sponsored video with explicit disclosure or implicit disclosure, or even without a disclosure. The effect of disclosure explicitness on skepticism was not different at all between YouTube users with high persuasion knowledge and YouTube users low persuasion knowledge, which is not line with the prediction. However, people who were more skeptical to the sponsored brand had a less positive brand attitude, just as what was predicted. Overall, skepticism did not mediate the effect of disclosure
explicitness on brand attitude and persuasion knowledge did not moderate the effect of disclosure explicitness on skepticism. H1 was thus rejected.
Table 1 Results of the PROCESS-analysis with the sponsorship disclosure as an independent variable, persuasion knowledge scale as a moderator, skepticism as a mediator and brand attitude as the dependent variable.
Independent variable
b SE t p 10,000 BC Boat
LL 95 UL95 Skepticism (mediator)
21 Constant 3.31 1.22 2.70 .008 0.88 5.73 Disclosure explicitnessX1 -1.46 1.67 - 0.87 .384 - 0.48 1.85 Disclosure explicitnessX2 2.12 1.81 1.17 .244 - 1.46 5.70 Persuasion knowledge scale 0.11 0.24 0.46 .650 - 0.37 0.58 DE X1 x Persuasion knowledge scale 0.30 0.33 0.92 .358 - 0.35 0.95 DE X2 x Persuasion knowledge scale - 0. 45 0.35 - 1.27 .207 - 1.15 0.25
Brand attitude (dependent variable)
Constant 7.03 0.27 26.28 < .001 6.50 7.56 Disclosure explicitness X1 - 0.33 0.19 - 1.72 .089 - 0.70 0.05 Disclosure explicitness X2 - 0.41 0.19 - 2.18 .031 - 0.79 - 0.04 Skepticism - 0.60 0.06 - 10.61 < .001 - 0.72 - 0.49 Conditional effect of Disclosure explicitness X1
on brand attitude Persuasion
Knowledge scale
Indirect effect
Boat SE BootLLCI BootULCI
22
Median 5.24 0.19 - 0.44 0.31
High 5.75 0.25 - 0.65 0.32
Conditional effect of Disclosure explicitness X2 on brand attitude
Low 4.17 0.30 - 0.64 0.55
Median 5.24 0.20 - 0.24 0.54
High 5.75 0.27 - 0.25 0.78
Note. X1 is comparing implicit disclosure condition with no disclosure condition. X2 is comparing explicit disclosure condition with no disclosure condition. b is the regression coefficient unstandardized. N = 124
Moderated mediation effect on brand attitude, moderated by recognition of sponsored content x
To check whether it was not persuasion knowledge in general, but a person's ability to recognize sponsored content that moderates the effect of disclosure explicitness on skepticism, another PROCESS analysis was conducted with the recognition subscale as moderator (Model 7). Same as previous analysis, disclosure explicitness was independent variables, and brand attitude was dependent variable. The whole model was mediated by skepticism and the sample was bootstrapped with 5,000 replications (Boerman et al, 2017).
As identified in the first analysis, there is no significant main effect of disclosure explicitness on skepticism (explicit disclosure vs. no disclosure: b = 0.19, SE = 0.54, p = 0.726, 95% CI [ - 1.27, 0.89]; implicit disclosure vs. no disclosure: b = 0.02, SE = 0.54, p = .972, 95% CI [ - 1.06, 1.10]). Recognition of sponsored content did not affect skepticism significantly, neither (b = 0.01, SE = 0.10, p = 0.925, 95% CI [ - 0.19, 0.27]. The analysis did
23 find the predicted significant interaction effect of disclosure explicitness and recognition of sponsored content on skepticism (explicit disclosure vs. no disclosure: b = 0.01, SE = 0.13, p = .918, 95% CI [ - 0.24, 0.27]; implicit disclosure vs. no disclosure: b = 0.02, SE = 0.13, p = .903, 95% CI [ - 0.24, 0.27]).
Same as the results in the first analysis, a significant result was found for the
predicted direct effect of skepticism on brand attitude (b = - 0.60, SE = 0. 06, p < .001, 95% CI [ - 0.72, - 0.49]). The additional finding of direct effect of disclosure explicitness on brand attitude was also found significantly between explicit disclosure and no disclosure (b = - 0.41, SE = 0.19, p = .031, CI [ - 0.79, - 0.04]) and marginally significant between implicit
disclosure and no disclosure (b = - 0.33, SE = 0.19, p = .089, CI [ - 0.70, 0.05]). The
moderated mediation model was not significant again (index of moderated mediation = -0.01, boot SE = 0.09, CI [ -0.18, 0.17]). The indirect effect of disclosure explicitness on brand attitude via skepticism was completely not significantly upon differently levels of recognition of sponsored content (See Table 2). Overall, same as what was found in prior analysis the results reflected the effect of disclosure explicitness on participants’ skepticism level did not differ by the degree of their ability to recognize a sponsored content. More skepticism was found leading to less positive brand attitude as predicted. Additionally, more explicit disclosure also resulted in less positive brand attitude. As we saw all the results in this analysis were the same as what we found in the previous model, H1 was still rejected.
Table 2 Results of the PROCESS-analysis with the sponsorship disclosure as an independent variable, recognition as a moderator, skepticism as a mediator and brand attitude as the dependent variable.
Independent variable
24 LL 95 UL95 Skepticism (mediator) Constant 3.82 0.44 8.64 < .001 2.94 4.69 Disclosure explicitnessX1 0.02 0.54 0.03 .972 - 1.06 1.10 Disclosure explicitnessX2 - 0.19 0.54 - 0.35 .726 - 1.27 0.89 Recognition of sponsored content 0.01 0.10 0.10 .925 - 0.19 0.27 DE X1 x Recognition of sponsored content 0.02 0.13 0.12 .903 - 0.24 0.27 DE X2 x Recognition of sponsored content 0.01 0.13 0.10 .918 - 0.24 0.27
Brand attitude (dependent variable)
Constant 7.03 0.27 26.28 < .001 6.50 7.56 Disclosure explicitnessX1 -0.33 0.19 - 1.72 .089 - 0.70 0.05 Disclosure explicitnessX2 - 0.41 0.19 - 2.18 .031 - 0.79 - 0.04 Skepticism - 0.60 0.06 - 10.61 < .001 - 0.72 - 0.49 Conditional effect of Disclosure explicitness
X1 on brand attitude
25 sponsored content effect
Low 1.40 0.28 - 0.56 0.54
Median 2.10 0.24 - 0.50 0.45
High 7.00 0.31 - 0.70 0.54
Conditional effect of Disclosure explicitness X2 on brand attitude
Low 1.40 0.28 - 0.44 0.66
Median 2.10 0.29 - 0.37 0.59
High 7.00 0.37 - 0.68 0.78
Note. X1 is comparing implicit disclosure condition with no disclosure condition. X2 is comparing explicit disclosure condition with no disclosure condition. b is the regression coefficient unstandardized. N = 124
Discussion
There are discrepancies among previous studies on how disclosure explicitness affects skepticism and brand attitude. Some studies show more explicit disclosure increase
skepticism and decrease positive brand attitude (Boerman et al, 2017; Sweetser et al, 2016; Tessitore and Geuens, 2013), while others find the opposite effects on both skepticism and brand attitude (Han et al., 2018; Carr and Hayes, 2014). With respect to the goal of this research, it aimed to find out the interaction between disclosure explicitness and persuasion knowledge on YouTube users’ skepticism and a mediation effect of skepticism upon the effect of disclosure explicitness on brand attitude. It was predicted that the more explicit sponsorship disclosure would result in less skepticism, resulting in more positive brand attitude, for YouTube users with high persuasion knowledge, but the effect would be the
26 other way around for YouTube users with low persuasion knowledge that the explicit
sponsorship will generate higher skepticism compared to an implicit disclosure or no disclosure, resulting in less positive brand attitude.
However, the findings rejected the hypothesis to a large extent, because there is no significant result for the whole moderated mediation model. No main effect of disclosure explicitness on skepticism was found, nor an interaction between disclosure explicitness and persuasion knowledge. As predicted people generated more skepticism was found having less positive brand attitude. Additionally, the result showed more explicit disclosure lead to less positive brand attitude.
Based on theory of reactance to persuasion (Brehm, 1966) and persuasion knowledge model (Friestad & Wright, 1994), it could be expected that a more or less explicit disclosure would generate different level of skepticism towards a brand, and this effect is dependent on people’s prior persuasion knowledge. Based on the prior knowledge on persuasion tactics, people with high or low persuasion knowledge were predicted to have different expectations of disclosures regarding whether there should be a disclosure and how explicit the disclosure was. And the persuasive attempt in sponsored content would thus be interpreted to positive or negative directions, resulting in more or less skepticism and positive brand attitude. However, in contrast to the hypothesis no significant interaction of disclosure explicitness and
persuasion knowledge on brand attitude was found.
For the first moderated mediation model with the persuasion knowledge scale as a moderator, this is likely caused by relatively high score of persuasion knowledge of all the participants. The result shows the scaled persuasion knowledge scores are relatively high and centralized (M=5.05, Median = 5.24, SD = .807) compared with other studies (Boerman et al., 2017; Boerman et al., 2018). Participants scoring at 4.17 were considered as people
27 with low persuasion knowledge in the analyses, which means a majority of participants stand on the high persuasion knowledge side of the scale. Hence, we got problem on correctly distinguishing between participants with high persuasion knowledge and participants with low persuasion knowledge. Some participants with high persuasion knowledge might be assigned to low persuasion knowledge, and some were reversely assigned. A mix of
participants in persuasion knowledge levels means a falsely participants assignment and thus resulted in a lack of findings in this experiment.
The predicted negative effect of skepticism on brand attitude was indeed found, but this effect was not affected by the explicitness of the disclosure, nor its interaction with PK. This finding confirmed the notion that people will tend to rate a brand relatively low when they are more skeptical to a brand, which was in line with most of empirical evidence (Kang, 2018; Yoo & Macinnis, 2005). This finding encouraged marketing practitioners to create the most credible advertising message to avoid more skepticism being generated. In general, consumers would be less likely to consider a brand positively when they question the reliability of its persuasive message, even when the message itself is likeable, which was aligned with many empirical evidence (Han et al., 2018; Carr & Hayes, 2014). In conclusion, the whole moderated mediation model was found not significant due to a nonsignificant effect of disclosure explicitness on skepticism and the nonsignificant moderation of persuasion knowledge.
The result showed an additional finding that more explicitly stated disclosure on YouTube would generate consumers less positive brand attitude, so it gave a clearer insight for marketeers conducting sponsored content campaign on YouTube. Besides, some previous researches studied the effect of disclosure formats related to explicitness but not specifically about disclosure explicitness on brand attitude (Sweetser et al., 2016; Boerman et al., 2015;
28 Tessitore & Geuens, 2013). This finding can thus serve as a conclusion and explanation of the previous results.
Limitations
This study has several limitations. A first limitation is that the measurement of persuasion knowledge did not fit this topic perfectly. In the first analysis with persuasion knowledge scale as moderator, the recognition of sponsored content subscale was deleted from the persuasion knowledge scale but recognition is a main focus for this study. We can also easily find the mean of scaled persuasion knowledge (M=5.05) is much higher than mean of
recognition subscale (M=3.29). The remaining subscales measuring persuasion knowledge were about the understanding of either persuasive intent, commercial source or persuasive tactics. The significant difference in means of scaled persuasion knowledge and recognition subscale are found (t persuasion knowledge (123) = 69. 68, p < .001, 95% CI [4.91, 5.19]; t recognition (123) = 16. 05, p < .001, 95% CI [2.88, 3.69]) and this indicated that people knew the marketing strategy well but they were lack of capability to recognize advertisement. Hence, the first analysis is not perfect to text the moderation of recognition.
For the second analysis with recognition of sponsored content as moderator, the recognition subscale is only measured by one item, which was very weak as a scale.
Moreover, it was measured a brand recognition check of brand logo placed in an ice skating poster ( see appendix) and participants’ description of sponsorship they saw. Then the item was recoded into a 7-point Likert scale. The result of this recognition in a poster might be different with the recognition of sponsored content in video, as there was many other texts in big size distracting participants and taking their attention from the disclosure of brand logo placement. However, this distraction would not happen in a YouTube video because there would not be more than one message conveying to audience at the same time. In short, the
29 measurement of persuasion knowledge did not completely fit the research topic of YouTube sponsored content. A better measurement of recognition subscale could be a recognition of disclosure placed in a video.
Second, the result showed that more explicit disclosure generated a bit less skepticism (M = 3.69, SD = 1.24) compared with implicit disclosure (M = 3.91, SD = 1.20) and no disclosure (M = 3.85, SD = 1.32). Although it is not significant, it is in line with the direction of what was predicted for people with high persuasion knowledge. As discussed before that majority of participants are actually with high persuasion knowledge, it might be a sign that the hypothesis for people with high persuasion knowledge is correct. The small sample size might be a reason for the little difference in skepticism scores, with only 124 participants valid after exclusion. With a larger sample size, the result might show more explicit
disclosure would decrease skepticism given the truth that most of participants are with high persuasion knowledge. In addition, the little difference might be caused by lack of interest on the sponsored content. When people were watching the sponsored fitness video in the
experiment, they did not watch it out of their own willingness. Only 26% participants chose that they would like to watch a fitness video on YouTube. According to the reactance theory, greater reactance is aroused more by greater personal relevance (Brehm, 1966). Due to the lack of personal relevance to the fitness topic and the sponsored product, participants’ reactance were likely not activated to a large extent and thus they would not be greatly skeptical to the brand (8fit) as much as they usually were in their personal lives to the things interesting to them. The disclosure explicitness therefore could not affect their skepticism level to a large degree.
With respect to the recommendations for future studies, disclosure explicitness regarding how explicit the language formed in a disclosure could be also interesting to
30 investigate. This study used the textual disclosure merely placed the title as implicit
disclosure and both textual disclosure in title and oral disclosure in the video as explicit disclosure, because these types of disclosures are mostly used by YouTubers in their
YouTube sponsored videos so that they are most interesting types to this research. There are also prior studies used different explicit levels of disclosure language about how clearly the sponsor is identified, as the measurement of disclosure explicitness (Amazeen & Muddiman, 2017; Carr & Hayes, 2014). Following the study by Amazeen and Muddiman (2017), the disclosure language explicitness can have three variations, in (a) low (i.e.: ‘partner content’), (b) medium (e.g.: ‘sponsored content’), (c) high (i.e.: ‘paid advertisement by [sponsor]). Furthermore, while no mediation effect of disclosures explicitness through skepticism was found on brand attitude, there was a significant direct effect of disclosure explicitness on brand attitude found. Therefore, there might be another variable that mediates the effect of disclosure explicitness on brand attitude. As described that reactance to persuasion is highly associated with persuasion knowledge and also dependent on the existence of disclosure (i.e.: forewarning), it could be a possible mediator for the effect of disclosure explicitness on brand attitude.
Conclusion
In summary, this research did not found that more or less explicit disclosure results in significantly more or less skepticism under the moderation effect of YouTube users’ persuasion knowledge. However, positive brand attitude was found decreasing for a higher skepticism level. Additionally, it was also found that more explicit disclosure decreased positive brand attitude. Due to the nonsignificant effect of disclosure explicitness on
skepticism moderated by persuasion knowledge, the proposed moderated mediation model does not exist.
31 Marketing practitioners can learn from the found evidence that YouTube users with high persuasion knowledge would not be more or less skeptical because the disclosure is more or less explicitly stated, or no disclosure is provided. And the nonsignificant effect works the same for YouTube users with low persuasion knowledge. Hence, an explicit disclosure is recommended because it won’t affect all the consumers’ skepticism and it may help the YouTuber to gain more credibility (Carr & Hayes, 2014). However, implicit
disclosure or no disclosure works better to increase YouTube users’ positive brand attitude. This is an explanation of why a large amount of disclosures on YouTube are not clearly stated or hidden in the description box with little notice from viewers. Therefore, this research recommend regulators to make more detailed standardization for the disclosure explicitness on YouTube.
Reference
37 Mind Blowing YouTube Facts, Figures and Statistics – 2019. (2019, January 5).
MerchDope. Retrieved from https://merchdope.com/youtube-stats/
Adyoulike (2015) Native advertising set to double by 2018. PR Newswire, 18 December.
Available at: http://www.prnewswire.com/news-releases/native-advertising-set-to-doubleby-2018-562919861.html
Amazeen, M., & Wojdynski, B. (2018). The effects of disclosure format on native advertising recognition and audience perceptions of legacy and online news publishers.
Journalism. https://doi.org/10.1177/1464884918754829
Benoit, W. L. (1998). Forewarning and persuasion. Persuasion: Advances through
meta-analysis, 139-154.
Brehm, J. (1966). A theory of psychological reactance. New York, N,Y., [etc: Academic Press.
32
Boerman, S., Van Reijmersdal, E., & Neijens, P. (2015). How audience and disclosure characteristics influence memory of sponsorship disclosures. International Journal of
Advertising, 34(4), 1–17. https://doi.org/10.1080/02650487.2015.1009347
Boerman, S., Willemsen, L., & Van Der Aa, E. (2017). “This Post Is Sponsored.” Journal of
Interactive Marketing, 38, 82–92. https://doi.org/10.1016/j.intmar.2016.12.002
Boerman, S., Reijmersdal, E., & Neijens, P. (2012). Sponsorship Disclosure: Effects of
Duration on Persuasion Knowledge and Brand Responses. Journal of Communication,
62(6), 1047–1064. https://doi.org/10.1111/j.1460-2466.2012.01677.x
Boerman, S., Van Reijmersdal, E., Rozendaal, E., & Dima, A. (2018). Development of the Persuasion Knowledge Scales of Sponsored Content (PKS-SC). International Journal
of Advertising, 37(5), 671–697. https://doi.org/10.1080/02650487.2018.1470485
Boush, D., Friestad, M., & Rose, G. (1994). Adolescent Skepticism toward TV Advertising and Knowledge of Advertiser Tactics. Journal of Consumer Research, 21(1), 165– 175. https://doi.org/10.1086/209390
Campbell, C., & Marks, L. (2015). Good native advertising isn’t a secret. Business Horizons, 58(6), 599–606. https://doi.org/10.1016/j.bushor.2015.06.003
Carr, C., & Hayes, R. (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, 00–00.
https://doi.org/10.1080/15252019.2014.909296
Dekker, K., & Van Reijmersdal, E. (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.
33
Matthes, J., & Naderer, B. (2015). Product placement disclosures: Exploring the moderating effect of placement frequency on brand responses via persuasion knowledge.
International Journal of Advertising, 35(2), 1–15.
https://doi.org/10.1080/02650487.2015.1071947
Fransen, M., Smit, E., & Verlegh, P. (2015). Strategies and motives for resistance to persuasion: an integrative framework. Frontiers in Psychology, 6, 1201. https://doi.org/10.3389/fpsyg.2015.01201
Friestad, M., & Wright, P. (1994). THE PERSUASION KNOWLEDGE MODEL - HOW PEOPLE COPE WITH PERSUASION ATTEMPTS. Journal Of Consumer Research, 21(1), 1–31. https://doi.org/10.1086/209380
Han, J., Drumwright, M., & Goo, W. (2018). Native Advertising: Is Deception an Asset or a Liability? Journal of Media Ethics, 33(3), 102–119.
https://doi.org/10.1080/23736992.2018.1477048
Hass, R., & Grady, K. (1975). Temporal delay, type of forewarning, and resistance to influence. Journal of Experimental Social Psychology, 11(5), 459–469. https://doi.org/10.1016/0022-1031(75)90048-7
Hayes, A. (2015). An Index and Test of Linear Moderated Mediation. Multivariate
Behavioral Research, 50(1), 1–22. https://doi.org/10.1080/00273171.2014.962683
Hullett, C. (2005). The impact of mood on persuasion - A meta-analysis. Communication
Research, 32(4), 423–442. https://doi.org/10.1177/0093650205277317
Isaac, M., & Grayson, K. (2017). Beyond Skepticism: Can Accessing Persuasion Knowledge Bolster Credibility? Journal of Consumer Research, 43(6), 895–912.
34
Janssen, L., Fransen, M. L., Wulff, R., & Reijmersdal, E. A. (2016). Brand placement disclosure effects on persuasion: The moderating role of consumer self-control.
Journal of Consumer Behaviour, 15(6), 503–515. https://doi.org/10.1002/cb.1587 John, D. R. (1999). Consumer socialization of children: A retrospective look at twenty-five
years of research. Journal of consumer research, 26(3), 183-213.
Kang, J. (2018). The Effect of Ad Skepticism and Celebrity Preference on Brand Attitude Change in Celebrity-Endorsed Advertising: Role of ad skepticism in celebrity-endorsed advertising. Japanese Psychological Research.
https://doi.org/10.1111/jpr.12242
Kuykendall, D., & Keating, J. (1990). Mood and persuasion: Evidence for the differential influence of positive and negative states. Psychology and Marketing, 7(1), 1–9. https://doi.org/10.1002/mar.4220070102
Lewinski, P., Fransen, M. L., & Tan, E. S. (2016). Embodied resistance to persuasion in advertising. Frontiers in psychology, 7, 1202.
Mathur, A., Narayanan, A., & Chetty, M. (2018). Endorsements on Social Media: An Empirical Study of Affiliate Marketing Disclosures on YouTube and Pinterest. https://doi.org/10.1145/327438
Matthes, J., Schemer, C., & Wirth, W. (2007). More than meets the eye: Investigating the hidden impact of brand placements in television magazines. International Journal of
Advertising, 26(4), 477–503. https://doi.org/10.1080/02650487.2007.11073029
Obermiller, & Spangenberg. (1998). Development of a Scale to Measure Consumer Skepticism Toward Advertising. Journal of Consumer Psychology, 7(2), 159-186. Obermiller, C., Spangenberg, E., & MacLachlan, D. L. (2005). Ad skepticism: The
35
Petty, R., & Cacioppo, J. (1979). Effects of Forewarning of Persuasive Intent and Involvement on Cognitive Responses and Persuasion. Personality and Social
Psychology Bulletin, 5(2), 173–176. https://doi.org/10.1177/014616727900500209
Peter Lewinski, Marieke L. Fransen, & Ed S Tan. (2016). Embodied Resistance to Persuasion in Advertising. Frontiers in Psychology, 7, 1202.
https://doi.org/10.3389/fpsyg.2016.01202
Spears, N., & Singh, S. (2004). Measuring Attitude toward the Brand and Purchase Intentions.
Journal of Current Issues & Research in Advertising, 26(2), 53-66.
Sweetser, K., Ahn, S., Golan, G., Hochman, A., Wojdynski, B., & Golan, G. (2016). Native Advertising as a New Public Relations Tactic. American Behavioral Scientist, 60(12), 1442–1457. https://doi.org/10.1177/0002764216660138
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.
https://doi.org/10.2501/IJA-32-3-419-442
Tutaj, K., & Van Reijmersdal, E. (2012). Effects of online advertising format and persuasion knowledge on audience reactions. Journal of Marketing Communications, 18(1), 5–18. https://doi.org/10.1080/13527266.2011.620765
van Reijmersdal, E., Fransen, M., van Noort, G., Opree, S., Vandeberg, L., Reusch, S., … Golan, G. (2016). Effects of Disclosing Sponsored Content in Blogs: How the Use of Resistance Strategies Mediates Effects on Persuasion. American Behavioral Scientist, 60(12), 1458–1474. https://doi.org/10.1177/0002764216660141
van Reijmersdal, E., Tutaj, K., Boerman, S., & Persuasive Communication. (2013). The effects of brand placement disclosures on scepticism and brand memory.
36 Communications : The European Journal of Communication Research, 38(2), 127–146.
https://doi.org/10.1515/commun-2013-0008
Wojdynski, B., & Evans, N. (2015). Going Native: Effects of Disclosure Position and Language on the Recognition and Evaluation of Online Native Advertising. Journal
of Advertising, 45(2), 1–12. https://doi.org/10.1080/00913367.2015.1115380
Wood, W., & Quinn, J. M. (2003). Forewarned and forearmed? Two meta-analysis syntheses of forewarnings of influence appeals. Psychological Bulletin, 129(1), 119.
Wu, K. (2016). YouTube marketing: Legality of sponsorship and endorsements in advertising.
JL Bus. & Ethics, 22, 59.
Yoo, C., & Macinnis, D. (2005). The brand attitude formation process of emotional and informational ads. Journal of Business Research, 58(10), 1397–1406.
https://doi.org/10.1016/j.jbusres.2005.03.011
Zuwerink Jacks, J., & Devine, P. G. (2000). Attitude Importance, Forewarning of Message Content, and Resistance to Persuasion. Basic & Applied Social Psychology, 22(1), 19–29. https://doi-org.proxy.uba.uva.nl:2443/10.1207/15324830051036243
Appendix
