Positive Biases: Effects of Para-social Relationship and Sponsorship Disclosure Prominence Level on Brand
Attitudes in the Context of Sina Weibo
Changyan Li (13012541) Master’s Thesis
Graduate School of Communication Master’s Program Communication Science
Supervisor: Eva van Reijmersdal Word count: 6126
Date: 28-06-2022
Abstract: With the rise of social media influencers (SMIs), marketers are making use of influencer marketing to help promote their brands. While the mechanism of
influencer marketing lies in the para-social relationships (PSRs) between SMIs and consumers, this variable has been largely ignored in previous research. Besides, the loose administration of sponsorship disclosure gives way to sponsorship disclosures of different prominence levels. Moreover, sponsorship disclosures on Sina Weibo, China’s most used social media, still lack investigation. This study therefore investigates how PSR and sponsorship disclosure prominence level affect brand attitudes via message credibility by means of an experiment with a 3 (disclosure prominence level: prominent vs vague vs no disclosure) × 2 (PSR: strong vs weak) design in the context of Sina Weibo. Results suggest that the prominence level of a sponsorship disclosure alone does not directly affect message credibility or indirectly affect brand attitudes via message credibility. However, participants with strong PSRs with the SMI report higher perceived message credibility when the sponsorship disclosure is prominent (vs vague). In contrast, no significant difference between the three disclosure conditions is found among participants with weak PSRs. Implications for marketers and SMIs are discussed.
Introduction
With the pervasion of social media in everyday life, social media influencers, a group of social media users who actively share self-generated content and gain many followers have stood out in the media landscape (Lin et al., 2018). Given their huge influence, marketers seek to cooperate with them, paying them to create content that speaks positively of the brands in order to foster positive brand attitudes and enhance purchase decision (Martínez-López et al., 2020). Defined as influencer marketing, this kind of marketing strategy raises ethical concerns as it entails a combination of
branded content and individual content creation, blurring the line between paid
advertisements and organic content (Boerman & van Reijmersdal, 2020). To empower consumers, third parties like the Federal Trade Commission (FTC) have enacted regulations on disclosing sponsorship on social media (FTC, 2019). One of the most popular social media, Instagram, released a built-in tool to help disclose sponsored content (Instagram Business Team, 2020).
Though there are regulations and tools of sponsorship disclosures, the execution is neither consistent nor compulsory. The loose administration gives way to disclosures of different forms. While previous research has investigated specific disclosure characteristics such as timing, position, and language (Binford et al., 2021; Evans et al., 2017; Pfeuffer & Huh, 2020; Wojdynski & Evans, 2016), the more overarching concept, the prominence level of a sponsorship disclosure, still lacks investigation.
Therefore, instead of comparing specific sponsorship disclosure characteristics or only the absence and presence of a sponsorship disclosure, this research firstly aims to
clarify how sponsorship disclosures of different levels of prominence affect brand attitudes.
Under the paradigm of the Persuasion Knowledge Model (PKM) (Friestad &
Wright, 1994), research has gone into ways of revealing the effects of sponsorship disclosures. The PKM suggests that people gain persuasion knowledge from previous experience and use this knowledge to cope with marketers’ persuasive attempts (Friestad & Wright, 1994). Based on this theory, research indicates that sponsorship disclosures activate people’s persuasion knowledge, resulting in a decrease in source credibility (Pfeuffer & Huh, 2020; Wojdynski & Evans, 2016), and brand responses such as brand attitudes, purchase intention, and intention to share electronic word-of- mouth (Boerman et al., 2017; Eisend et al., 2020; Lim et al., 2021). However, while some research fails to find significant effects (Jung & Heo, 2019; Pfeuffer et al., 2020), the others find that disclosures increase source credibility (Carr & Hayes, 2014; Dhanesh & Duthler, 2019), and that a sponsorship disclosure can mitigate the negative effect of recognition of advertising via increased sponsorship transparency (Campbell & Evans, 2018; Evans et al., 2019).
The ambiguity of the effects of sponsorship disclosures implies that boundary conditions may exist. One possible explanation could be that the mechanism of the effectiveness of influencer marketing lies in the psychological factor of para-social relationship (PSR) (Hu et al., 2020; Sokolova & Kefi, 2020). However, only a few previous studies about sponsorship disclosures (Boerman & van Reijmersdal, 2020;
Breves et al., 2021) take the relationship between the participants and the influencers
into account (for execptions see: De Veirman & Hudders, 2020; Evans et al., 2017;
Pfeuffer et al., 2020; Pfeuffer & Huh, 2020; Stubb & Colliander, 2019). In their experiment about YouTube sponsored videos among children, Boerman and van Reijmersdal (2020) find that PSR moderates the negative effect of disclosures on brand attitudes via activation of persuasion knowledge, indicating that a high PSR can eliminate the negative effects of sponsorship disclosures. Therefore, the second aim of this study is to introduce the variable of PSR to further explore the boundary effects of sponsorship disclosures among adults.
In addition, this study also aims to unravel the underlying mechanism of how sponsorship disclosures function, in which credibility is indispensable. Credibility has long been an important topic in persuasive communication and has been proved to be a mediator of the relationship between sponsorship disclosure and brand responses (Balaji et al., 2021; Eisend et al., 2020). However, it is usually seen as a source attribute only (Appelman & Sundar, 2016), or research does not make a clear differentiation between different types of credibility. According to Appelman and Sundar (2016), credibility in mediated communication contains three different perspectives, namely, medium, source, and message credibility. In the context of influencer marketing, where PSR plays a crucial role (Hu et al., 2020; Sokolova &
Kefi, 2020), the differentiation between source credibility and message credibility is necessary because a strong para-social relationship already entails high trust in the source (Lou, 2021.; Tukachinsky & Stever, 2019). Moreover, an investigation into message credibility is also necessary since sponsorship disclosure as a non-source
factor can affect message credibility (Appelman & Sundar, 2016). Through the activation of conceptual persuasion knowledge, sponsorship disclosures help people understand persuasive intents (Boerman et al., 2017), which serve as heuristics that affect people’s evaluation of message credibility (Metzger et al., 2010). Hence, the third aim of this research is to explore the concept of perceived message credibility as a mediator of the effect of sponsorship disclosures on brand attitudes.
While the effects of sponsorship disclosures have been studied on social media like Instagram and YouTube (e.g. De Veirman & Hudders, 2020; Pfeuffer et al., 2020), Sina Weibo (below refer as Weibo), China’s largest microblogging platform, still lacks investigation. With 550 million active users per month (Sina Weibo Data Center, 2021), Weibo is an important place where marketers execute influencer marketing. However, while only paid native ads will be disclosed, Weibo has no specific disclosure regulation for sponsored content, resulting in a low disclosure rate.
Therefore, the final aim of this research is to reveal how sponsorship disclosures work in the context of Weibo, providing implications for different stakeholders.
The effects of sponsorship disclosure prominence level on message credibility According to Friestad and Wright (1994), people develop knowledge about persuasion tactics from previous experience with marketing content and they use this knowledge to cope with marketers’ persuasion attempts. In the context of covert advertising, where marketers try to hide the advertising nature of the messages, people need to firstly recognize the messages as advertising to activate their persuasion knowledge (Wojdynski & Evans, 2020). However, the fact that the format and the
content of a sponsored social media post (SSMP) resemble organic posts leads to covertness (De Veirman & Hudders, 2020). Moreover, posting by an individual social media user (influencer) instead of a brand makes its advertising nature even more difficult to recognize. As a result, it is likely that people fail to recognize SSMPs as paid advertisements, and their persuasion knowledge is hence not activated (Boerman et al., 2017; Wojdynski & Evans, 2020). In this scenario, a tool that helps people recognize the advertising nature of sponsored posts seems necessary. The FTC suggests that a sponsorship disclosure can serve as a cue for people to identify advertising in covert advertising. Previous research also demonstrates that
sponsorship disclosures increase advertising recognition in news websites (Wojdynski
& Evans, 2016) and social media (Boerman et al., 2017; Eisend et al., 2020).
Once people’s persuasion knowledge is activated, they will use this knowledge to cope with the persuasive attempt, either resist it or accept it. According to the
Psychological Reactance Theory (Brehm & Brehm, 2013), people try to retrieve their freedom and autonomy when it is threatened. Therefore, when people realize the persuasion attempts contained in the SSMPs, they are likely to cope with the attempts with skepticism and resistance. Previous empirical research about social media supports this effect, indicating that sponsorship disclosures can damage source credibility (e.g., De Veirman & Hudders, 2020; Deng et al., 2020; Pfeuffer & Huh, 2020) through the activation of persuasion knowledge. However, to the best of the author’s knowledge, no previous research has explicitly investigated message credibility in the context of sponsorship disclosure in social media.
Message credibility refers to individuals’ perception of trustworthiness and veracity of the content of communication (Appelman & Sundar, 2016). One related concept, attitudinal persuasion knowledge, which refers to feelings of distrust, skepticism and dislike toward advertising has been proved to be negatively affected by sponsorship disclosures through recognition of advertising (Boerman et al., 2017).
Besides, research about sponsorship disclosures on news websites also manifests that people find the news less credible when they recognize it as advertising (Wojdynski
& Evans, 2016). Based on the argumentation above, it is anticipated that a sponsorship disclosure in a social media post will activate people’s persuasion knowledge and will consequently decrease perceived message credibility.
Nonetheless, disclosures of different prominence levels may have different effects on message credibility. In accordance with An et al(2019), a prominent disclosure should meet three requirements. Firstly, it should be noticeable in terms of color and font. Secondly, it should be placed near to the ad itself. Finally, it should be understandable, avoiding technical or industry jargon. This research will focus on the first and the second point, namely, the visual prominence of disclosures. While Weibo has not offered related guidelines, the FTC (2019) provides clear guidelines for
sponsorship disclosures on picture-based platforms like Instagram and Snapchat.
Given that Weibo is based on the function of photo sharing (Qin, 2020), the FTC guidelines about picture-based platforms should be appliable to Weibo. The FTC states that the disclosures should be placed on the pictures and hard to miss. Besides, it also recommends that the color of the disclosure should stand out against the
background. On the contrary, it discourages mixing the disclosures in hashtags because people tend to ignore them.
In line with the FTC guidelines, this research will compare a disclosure presented on the picture (prominent) and a disclosure mixed with hashtags (vague). Because a prominent disclosure is hard to miss, it is more likely to assist people in recognizing the advertising nature of the post, resulting in lower perceived message credibility compared with no disclosure and vague disclosure. Supporting this assumption, an eye-tracking research by Boerman and Müller (2022) indicates that people pay more visual attention to brand tags appear on pictures than hashtags placed at the end of the caption. In terms of the comparison between the vague disclosure and no disclosure conditions, due to the covertness of the vague disclosure, it is likely that it has little effect on facilitating people to recognize the ad. Hence, it will not decrease people’s perceived message credibility. Therefore, it is expected that:
H1: (a) A prominent disclosure, compared to a vague and no disclosure, will lead to lower message credibility. (b) There is no significant difference in message
credibility between the vague and the no disclosure conditions.
The effects of message credibility
The concept of credibility is sometimes used interchangeably with trust. Trust is a critical element in interpersonal communication (Ridings et al., 2002), and its
importance also applies in the online environment (Corritore et al., 2003; Urban et al., 2009). However, while trust entails beliefs, predispositions, and behaviors connected to the acceptance of risks and vulnerability (Weitzl, 2016), credibility refers to the
quality of the message or the source perceived by the audience (Rieh & Danielson, 2007). Hence, as said by Corritore et al, “credibility provides a reason to trust but is not trust itself”. In addition, differentiation between message credibility and source credibility is also necessary. Source credibility refers to how much the audience believes in the sender of the message (Wu & Wang, 2011). Distinct from source credibility, message credibility measures the veracity of the content of the message (Appelman & Sundar, 2016).
In research about advertising, advertisement credibility, defined as the extent to which the audience believes the claims about the brand made in the message is truthful and believable (MacKenzie & Lutz, 1989), is a positive predictor of attitudes toward the advertisement (MacKenzie & Lutz, 1989), which in turn positively affects brand attitudes (Choi & Rifon, 2002; MacKenzie et al., 1986). Evidence from
research about Weibo native ad posts also indicates that message credibility plays the role of a mediator of the effect of number of likes on brand attitudes (Seo et al., 2019).
When a SSMP is accompanied with a prominent disclosure, people are likely to recognize the post as an advertisement (Boerman et al., 2017; Wojdynski & Evans, 2020). Therefore, under the condition of prominent sponsorship disclosure, the role of message credibility is anticipated to be equivalent to advertisement credibility, which has a positive effect on brand attitudes.
Regarding the SSMPs with vague disclosures or without disclosures, the concealment of the commercial relationship between the brand and the influencer leads to less possibility of people realizing the advertising nature of the posts. Hence,
the SSMP will be seen as a piece of electronic word-of-mouth (eWOM) from the influencer. Message credibility is a strong predictor of eWOM adoption and the change in attitudes (Weitzl, 2017). Empirical research also supports that the perceived credibility of eWOM positively affects brand attitudes via attitudes toward the eWOM itself (Hsu, 2021). Therefore, it is expected that the message credibility of SSMPs without disclosures or with vague disclosures also impacts brand attitudes positively.
Combining the argumentation above, it is assumed that message credibility has a positive effect on brand attitudes. In addition, this research predicts a mediating role of message credibility, explaining the effects of disclosure prominence on brand attitudes, which is supported by previous research (Balaji et al., 2021; Giuffredi-Kähr et al., 2022; Seo et al., 2019; Visentin et al., 2019).
H2: Message credibility mediates the effects of sponsorship disclosure
prominence level on brand attitudes, with higher prominence level leading to lower message credibility, which results in more negative brand attitudes.
The moderating role of para-social relationship
With transportation and identification, people often feel connected with media personalities even if this connection does not exist in real life. This relationship is coined as para-social relationship (PSR), which refers to the illusion of having an enduring and close relationship with media personalities (Horton & Richard Wohl, 1956). On social media where users closely follow the updates of the influencers whom they follow, PSR also plays an important role (Hu et al., 2020; Sokolova &
Kefi, 2020). However, unlike the one-sided and non-reciprocal characteristics of
traditional media (e.g., TV and radio), social media provides opportunities (e.g., comments and likes) for users to engage with influencers. Therefore, the PSRs between followers and influencers contain similar but different meanings. According to Lou (2021), the PSR on social media is redefined as a trans-parasocial relationship, which is developed via collectively reciprocal, synchronously interactive, and co- created interactions. A similar concept, para-social interaction (PSI) is often used interchangeably with PSR. Nonetheless, while PSI only refers to the perception of the media personality as an intimate person during a momentary interaction, PSR is a long-term relationship that takes a long time to build.
The development of PSR goes through a sequence of steps, including initiation, experimentation, intensification, and integration/bonding (Tukachinsky & Stever, 2019). According to Tukachinsky and Stever (2019), a strong PSR that goes to the stage of intensification is formed based on high trust in the media personality and intimacy. A qualitative research by Lou (2021) supports this theory, indicating that people assign high trust to their favorite influencers. Trust in the source has been proved to mitigate the negative effect of sponsorship disclosures by the research of Dekker & Reijmersdal (2013). Their results show that the acceptance of product claims of the people with high perceived source credibility is not affected by the disclosure, while people with low perceived source credibility report significantly lower acceptance.
In addition, a strong PSR entails a positive bias towards the influencers that can change people’s perception of sponsorship disclosures. Qualitative research reveals
that people understand the influencers’ business model and are supportive of their favorite influencers getting paid through endorsements (Lou, 2021.; van Dam & van Reijmersdal, 2019). In the context of high PSR, a sponsorship disclosure could be seen as improved advertising transparency and honesty rather than the cue of deceptive advertisements. Perceived transparency has been proved to alleviate the negative effects of sponsorship disclosures on purchase intention when people recognize a covert advertisement (Boerman & van Reijmersdal, 2020; Evans et al., 2019). In line with the argumentation above, it is assumed that people who have strong PSRs with the SMI will be less critical about the sponsored message even when there is a prominent sponsorship disclosure.
Conversely, people with strong PSRs may react in a different way if they notice a vague disclosure because it could be deemed as an inappropriate tactic that
deliberately makes the advertisement hard to recognize. Native advertising that is perceived to be deceptive due to vague disclosure is perceived to be less credible (Wojdynski, 2016). This effect could be explained by the Expectancy Violation Theory (Burgoon, 2015). During the formation of PSRs, people have validated consistency of the influencers and have formed expectations of them (Tukachinsky &
Stever, 2019). In general, people think the influencers they like are honest and genuine. A vague sponsorship disclosure that entails deceptiveness could be detrimental to the relationship as it violates people’s positive expectations, and consequently affects people’s evaluation of the message.
Nonetheless, the effects of prominent disclosures and vague disclosures stated
above should not occur for people with low PSRs with the influencers. For them, a sponsorship disclosure will play the role of a cue that activates their persuasion knowledge, which leads to reactance to the message. Based on the argumentation above and the argumentation for Hypothesis 1, it is therefore proposed that:
H3: PSR moderates the effects of sponsorship disclosure prominence level on message credibility: (a) participants with strong PSRs with the influencer will have higher perceived credibility of the message if the disclosure is prominent compared to vague and no disclosure, while (b) participants with weak PSRs will have less
perceived message credibility if the disclosure is prominent compared to vague and no disclosure.
Finally, the previous hypotheses can be merged, and an overall moderated- mediation model is proposed as following:
H4: PSR moderates the indirect effect of sponsorship disclosure prominence level on brand attitudes through message credibility.
Research Methods Pre-test
To select a proper vague disclosure for the stimuli and make sure that the
difference between the prominent disclosure and the vague disclosure is big enough, a between-subjects pre-test was conducted online. A questionnaire was developed on Qualtrics and shared via a link in the author’s social network (WeChat). In total, 67 people took part in the pre-test, among whom 4 were dropped because they were not Weibo users.
Participants were randomly assigned to one of the three conditions (prominent vs vague 1 vs vague 2 disclosure, see Figure 1). They were firstly asked to fill in some demographic information including gender and age. After that, a screenshot of a sponsored Weibo post with a sponsorship disclosure was shown. A question stating,
“did you see the disclosure ‘sponsored by @brand’ in the screenshot?” was then asked. Finally, participants were required to indicate to what extent they agreed that the post was an advertisement on a scale range from 1 (strongly disagree) to 5 (strongly agree). The results show that 70% of the participants in the prominent disclosure condition saw the disclosure, 50% of the participants in the vague 2 disclosure condition saw the disclosure while only 30% of the participants in the vague 1 disclosure reported that they saw it. The prominent disclosure and the vague 2 disclosure were then used for the main experiment. However, the post used in the pre-test was not adopted in the main experiment because most of the participants recognized it as an ad even without a disclosure.
Fig. 1. Pre-test stimuli of prominent disclosure (left), vague 1 disclosure (middle), and vague 2 disclosure (right)
Design and Sample
An online between-subjects experiment with a manipulation factor with three levels (disclosure prominence: prominent vs vague vs no disclosure) and a measured factor with two levels (PSR: strong vs weak) was conducted among 224 participants.
However, 28 participants were dropped from the sample. Among them, 13
participants were dropped because they were not Weibo users, and 15 participants did not provide enough responses. In the final sample of 196 participants (143 males, 51 females, 2 non-binary or do not want to say), the mean age was 24.59 (SD = 3.78) years old with 19 years old the youngest and 53 years old the oldest.
Procedure and stimuli
An online questionnaire was developed using Qualtrics and was distributed via
the author’s WeChat account. All participants took part in the research voluntarily without receiving any incentive. They were randomly assigned to one of the three conditions, and a cover story saying that the research was about people’s usage habits and patterns of Weibo was told to avoid reminding them of the existence of the disclosures.
The first block of the questionnaire contained demographic questions like gender, age, and how much time they spend on Weibo per day in minutes. A screenshot of three Weibo post mock-ups, including a sponsored post and two posts from normal people that were added to make the stimuli more natural, was shown in the second block (see Figure 2). The screenshots were identical except for the sponsorship disclosures across the three conditions. The sponsored post was sent by Kakakaoo-, a Weibo influencer who has more than 10 million followers. There was a picture showing Kakakaoo- and two PopMart toy models in the post, and the caption said (translated), ‘Nice weather to show off my new toy models! PopMart Molly is so cute and the quality of it is super good! I love it so much! #Molly’s day’.
In the prominent disclosure condition, a disclosure (sponsored by @PopMart) was presented in the top-left of the picture, with white fonts and a yellow background.
This disclosure was designed in accord with the FTC guidelines that in a picture- based social media, the disclosure should overlay the picture. Besides, the fonts and background color were designed to be noticeable. The vague disclosure condition was developed in line with a kind of disclosure that the FTC does not recommend, that is, mixing the disclosures with hashtags. The vague disclosure was presented at the end
of the caption as a hashtag (#sponsored by @PopMart) that comes after another hashtag about the Molly toy model. The manipulation was checked by asking participants to indicate whether they saw the disclosure at the end of the questionnaire.
Participants were instructed to read through the screenshot carefully before clicking the next button. The next button in this block was set to delay for 15 seconds to make sure that people pay enough attention to the screenshots. In the final block, participants were asked to fill out some relevant scales (see Appendix). After
finishing the questionnaire, a debrief about the real aim of the research was displayed.
Fig. 2. Stimuli of control condition (left), prominent (middle), and vague disclosure (right)
Measures
Message credibility was measured by the scale created by Appelman and Sundar
(2016) with responses ranging from 1 (strongly disagree) to 7 (strongly agree). The scale consisted of three items that asked the participants to indicate to what extent they agree that the Weibo post sent by Kakakaoo- was authentic, accurate, and believable. The mean score of these three items was computed to measure message credibility (M = 4.04, SD = 1.18, Cronbach’s alpha = .90).
Six seven-point semantic differential scales were used to measure brand attitudes (bad/good, unfavorable/favorable, dislike/like, poor quality/high quality
unpleasant/pleasant, and negative/positive) with higher points meaning more positive attitudes (Boerman et al., 2012). Factor analysis indicated that all six items load on one factor (eigenvalue = 4.36, explained variance = 72.62%). The mean value of these six items was adopted to measure brand attitudes (M = 4.54, SD = 1.13, Cronbach’s alpha = .92).
The scale with nine items by Chung and Cho (2017) was employed to measure PSR. Example items were “Kakakaoo- makes me feel comfortable as if I am with a
friend” and “I would like to have a friendly chat with Kakakaoo-”. The response scale ranged from 1 (strongly disagree) to 7 (strongly agree). Factor analysis showed that the items load on 3 different factors that were subconstructs of PSR, namely,
understanding (eigenvalue = 1.09, explained variance = 12.14%), friendship (eigenvalue = 4.21, explained variance = 46.81%), and identification (eigenvalue = 1.20, explained variance = 13.34%) (Chung & Cho, 2017). In accord with Chung and
Cho (2017), these three subconstructs were combined into one variable by computing the mean value of the nine items. After that, a binary variable was created using median-split (Median = 3.89) to categorize strong and weak PSR.
As for control variables, brand familiarity, influencer familiarity, and Weibo usage were measured. Brand familiarity was measured by asking participants to
indicate to what extent they agree with the statement “I am familiar with PopMart” on a scale ranging from 1 (strongly disagree) to 7 (strongly agree) (M = 3.93, SD =1.74).
Influencer familiarity was measured in a similar way. Participants were instructed to indicate to what extent they agree with the sentence “I am familiar with Kakakaoo-”
on a seven-point scale (M = 2.67, SD = 1.51). Besides, Weibo usage was measured by the time people spend on Weibo per day in minutes (M = 53.23, SD = 69.17).
Results
Preliminary analyses
A Chi-Squared test showed that there was no significant difference in terms of gender between the three conditions, x2(6) = 5.95, p = .428. A series of one-way ANOVAs also revealed that the three conditions did not differ in respect of age, F (2, 195) = 3.00, p = .052, Eta2 = .03, media usage, F (2, 195) = 1.35, p = .261, Eta2 = .01, SMI familiarity, F (2, 195) = 0.73, p = .484, Eta2 = .01, and brand familiarity, F (2, 195) = 0.96, p = .385, Eta2 = .01. Hence, no variable was used as covariate in the further analysis.
The manipulation check indicated that 72.3% and 69.9% of the participants in the control condition and vague disclosure condition did not see a disclosure in the
screenshots. In the prominent disclosure condition, 41.4% of the participants confirmed that they saw the disclosure. However, a Chi-Squared test indicated that the relationship between disclosure condition and manipulation check results was non-significant, x2(4) = 3.11, p = .540.
Effects of disclosure prominence on message credibility
To test the first hypothesis, a one-way ANOVA was run with message credibility as the dependent variable and disclosure condition as the factor. The result showed that there was no significant difference between the three disclosure conditions in message credibility, F (2, 195) = 0.39, p = .681, Eta2 < .001. Bonferroni post hoc tests manifested that the difference between the vague condition and the control condition was non-significant, p = 1.000. Therefore, while H1(a) was not supported, H1(b) was supported.
Mediation analysis
The expected mediation in H2 was tested using Hayes’ (2017) PROCESS Marco (version 4.1) in SPSS. Model number 4 and 5000 bootstrap samples were used. In the model, disclosure condition was indicator coded and was included as the independent variable with the control condition as the reference group (x1 = vague disclosure, x2 = prominent disclosure). Besides, message credibility was included as the mediator and brand attitudes as the dependent variable. The results showed that the effect of disclosure condition on message credibility was non-significant, F = 0.38, p = .681, while the effect of message credibility on brand attitudes was significant, F = 26.11, p< .001. However, the indirect effect of prominent disclosure (vs no disclosure) on
brand attitudes was non-significant (indirect effect = -.031, SE = 0.12, 95% CI [-.27, .20]). Furthermore, the indirect effect of vague disclosure (vs no disclosure) on brand attitudes was non-significant as well (indirect effect = -.09, SE = 0.09, 95% CI [-.27, .10]). Hence, H2 was not supported.
Moderating effect of PSR
Hypothesis 3 was tested by a two-way ANOVA with disclosure condition and PSR (weak vs strong) as two factors and message credibility as the dependent
variable. The results showed a statistically significant interaction effect between PSR and disclosure condition, F (2, 196) = 4.57, p = .012, Eta2 = .04. Figure 3 shows that when the PSR was strong, participants who were exposed to the prominent disclosure perceived significantly higher message credibility of the post than those who were in the vague disclosure condition, Mdiff = 0.69, p = .043, 95% CI [0.02, 1.37]. The perceived message credibility of the control condition was also lower than the prominent condition, but the difference was non-significant, Mdiff = 0.32, p = .734.
Therefore, H3(a) was partly supported. In terms of the weak PSR group, there was no significant difference between the three conditions, resulting in a rejection of H3(b).
Besides, it is worth mentioning that the perceived message credibility of the strong PSR group was significantly higher than that of the weak PSR group in the prominent disclosure condition, Mdiff = 1.61, p < .001. The same situation applied to the control condition, Mdiff = 0.98, p < .001.
Fig. 3. Graphical depiction of interaction between disclosure prominence level and PSR Moderated mediation
To test the moderated mediation proposed in hypothesis 4, model number 7 of PROCESS Macro (version 4.1) was run in SPSS with 5000 bootstraps samples.
Disclosure condition was included as an indicator coded independent variable with the control condition as the reference group (x1 = vague disclosure, x2 = prominent disclosure) and brand attitudes was the dependent variable. Besides, message credibility was added as the mediator with PSR (weak vs strong) serving as the moderator. The results indicated that the moderated mediation was non-significant because the 95% CIs of the two indexes of moderated mediation both contained 0 (indexx1 = - .26, 95% CI [- .64, .09], indexx2 = .32, 95% CI [- .06, .73]). As a result, H4 was not supported.
Discussion
This study addressed the effects of the prominence levels of sponsorship
disclosures and PSR on brand attitudes. It firstly aimed to clarify how the prominence
3.00 3.50 4.00 4.50 5.00
control vague prominent
Message credibility
Disclosure prominence level
weak strong
level of a sponsorship disclosure affects the perceived message credibility of a SSMP.
Secondly, by introducing the variable of PSR, the potential boundary effect of sponsorship disclosure prominence level on message credibility was tested. Finally, this research posited and tested the potential mediator (message credibility) between the sponsorship disclosure prominence level and brand attitudes.
Results manifested that the sponsorship disclosure prominence level alone did not influence message credibility. There was no significant difference in perceived
message credibility between the prominent disclosure, vague disclosure, and no disclosure conditions. This result contradicted previous research which indicated that the existence of sponsorship disclosures can negatively affect attitudinal persuasion knowledge, that is, people distrust the content of the social media post more when it is accompanied by a sponsorship disclosure (Boerman et al., 2017; De Veirman &
Hudders, 2020; Lim et al., 2021). However, the results were in line with the research by De Cicco et al (2021), which revealed that the prominence level of a sponsorship disclosure did not have a significant main effect on advertising recognition, which will not elicit reactance that can affect attitudes toward message according to the Covert Advertising Recognition and Effects model (Wojdynski & Evans, 2020). One possible reason could be that this research was conducted in the context of Weibo where sponsorship disclosures are rare. It was probable that participants were not used to paying attention to sponsorship disclosures, or they did not understand the
implications of the disclosures well. Hence, a sponsorship disclosure alone did not influence participants’ evaluation of the message credibility of the SSMP.
Although the prominence levels of sponsorship disclosures did not directly affect message credibility, the moderation between PSR and the sponsorship disclosure prominence level was found to be significant. Specifically, the prominent sponsorship disclosure led to significantly higher message credibility than the vague disclosure for participants with strong PSRs. This interesting finding supported the notion of halo effects, which refers to the phenomenon that consumers have positive biases toward the SMIs to whom they feel connected (Breves et al., 2019). This halo can transform prominent sponsorship disclosures into improved transparency and honesty rather than deceptiveness that is activated by consumers’ persuasion knowledge. The results also align with the findings of Boerman and van Reijmersdal (2020) and Hwang and Zhang (2018) that strong PSRs can alleviate the negative effects of sponsorship disclosures. Furthermore, the relatively lower message credibility in the vague disclosure condition supported the Expectancy Violation Theory (Burgoon, 2015), implying that hiding the advertising nature of the sponsored content will backfire for people with strong PSRs.
As the indirect effect of the prominence level of sponsorship disclosures on message credibility was non-significant, the proposed mediation in H2 was non- significant as well. However, results showed that the direct effect of message
credibility on brand attitudes was significantly positive, which confirmed that similar to advertisements and eWOM, message credibility also played the role of a positive predictor of brand attitudes in the context of SSMPs (Choi & Rifon, 2002; Hsu, 2021;
MacKenzie et al., 1986; MacKenzie & Lutz, 1989).
In addition, though the moderation between PSR and the prominence level of sponsorship disclosures was significant, the moderated mediation model proposed in H4 was non-significant. One possible explanation was that though the moderation was significant, the effect size was rather small (Eta2 = .04).
Theoretical and practical implications
As stated in the introduction section, research about the effects of sponsorship disclosures has not reached a consistent standing point. This inconsistency implies the existence of potential moderators that previous research has ignored. Contrary to the real-life situation in which consumers often consume content created by the SMIs whom they have followed for a period, previous research often used made-up SMIs or did not take the relationships between the SMIs and the consumers into consideration (e.g. Pfeuffer et al., 2020; Pfeuffer & Huh, 2020). To fill this gap, this research addressed the importance of the PSRs between SMIs and consumers. The results also confirmed that PSR plays a moderating role in the mechanism of the effects of sponsorship disclosures. This research provides a deeper understanding of the
functions of sponsorship disclosures, indicating that it is worthwhile to investigate the effects of sponsorship disclosures beyond the paradigm of PKM.
In terms of practical implications, this research provides insights both for marketers and SMIs. As indicated by the results, sponsorship disclosures alone will not damage message credibility and brand attitudes. In addition, because followers usually possess stronger PSRs with the SMIs they follow (Breves et al., 2021), prominent disclosures are therefore recommended to generate higher perceived
message credibility. Conversely, the attempt to make the disclosure vague will lead to negative consequences. Therefore, SMIs should provide clear and prominent
sponsorship disclosures when posting sponsored content. Besides, since there is no clear regulation of sponsorship disclosures on Weibo, policymakers should enact related guidelines and regulations to empower consumers.
Limitations and future research
This research has several limitations. Firstly, a convenience sample was used due to the limited funding and time. Specifically, there were more females than males in the sample, leading to a decrease in the external validity of the results. Secondly, although the pre-test showed that more participants noticed the prominent disclosure than the vague disclosure, the difference between the two turned out to be smaller in the manipulation check of the main experiment, and many participants did not even notice the disclosures. However, this situation is common in research about
sponsorship disclosures (Boerman et al., 2017; van Reijmersdal et al., 2017;
Wojdynski & Evans, 2016), reflecting that people pay little attention to sponsorship disclosures. Thirdly, the prominence level of a sponsorship disclosure is a
combination of fonts, colors, and placement of the disclosure. To figure out which element contributes to the prominence of disclosure, future research can conduct separate experiments to gain a better understanding. Finally, this research was
conducted in the context of Weibo, where sponsorship disclosures are not compulsory and scarce. As a result, participants were likely to be less sensitive to sponsorship disclosures than the participants of the research on platforms like Instagram and
Facebook. Therefore, the generalizability of the findings of this research is limited. In addition,the moderation between PSR and the sponsorship disclosure prominence level seems to support the viewpoint of Jung and Heo (2019) that consumers might be savvy enough to recognize the advertising nature of native ads even without
disclosures. In this research, it was possible that sponsorship disclosures did not trigger advertising recognition but were an additional reference for the evaluation of message credibility. However, to dive deeper into this underlying mechanism, future research should measure advertising recognition as well and include it in the model.
To conclude, this research indicates that the presence of sponsorship disclosures will not harm the message credibility of SSMPs or brand attitudes. Nevertheless, for those who have strong PSRs with SMIs, prominent sponsorship disclosures will increase perceived message credibility. On the contrary, vague sponsorship disclosures can cause detrimental effects.
References
An, S., Kang, H., & Koo, S. (2019). Sponsorship disclosures of native advertising:
Clarity and prominence. The Journal of Consumer Affairs, 53(3), 998–1024.
https://doi.org/10.1111/joca.12212
Appelman, A., & Sundar, S. S. (2016). Measuring message credibility: Construction and validation of an exclusive scale. Journalism & Mass Communication Quarterly, 93(1), 59–79. https://doi.org/10.1177/1077699015606057
Balaji, M. S., Jiang, Y., & Jha, S. (2021). Nanoinfluencer marketing: How message features affect credibility and behavioral intentions. Journal of Business
Research, 136, 293–304. https://doi.org/10.1016/j.jbusres.2021.07.049
Binford, M. T., Wojdynski, B. W., Lee, Y.-I., Sun, S., & Briscoe, A. (2021). Invisible transparency: Visual attention to disclosures and source recognition in Facebook political advertising. Journal of Information Technology & Politics, 18(1), 70–
83. https://doi.org/10.1080/19331681.2020.1805388
Boerman, S. C., Reijmersdal, E. A. van, & Neijens, P. C. (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. C., & van Reijmersdal, E. A. (2020). Disclosing influencer marketing on YouTube to children: The moderating role of para-social relationship. Frontiers in Psychology, 10, 3042–3042. https://doi.org/10.3389/fpsyg.2019.03042
Boerman, S. C., Willemsen, L. M., & Van Der Aa, E. P. (2017). “This post is sponsored”: Effects of sponsorship disclosure on persuasion knowledge and electronic word of mouth in the context of facebook. Journal of Interactive Marketing, 38, 82–92. https://doi.org/10.1016/j.intmar.2016.12.002
Brehm, S. S., & Brehm, J. W. (2013). Psychological reactance: A theory of freedom and control. Academic Press.
Breves, P., Amrehn, J., Heidenreich, A., Liebers, N., & Schramm, H. (2021). Blind trust? The importance and interplay of parasocial relationships and advertising disclosures in explaining influencers’ persuasive effects on their followers.
International Journal of Advertising, 40(7), 1209–1229.
https://doi.org/10.1080/02650487.2021.1881237
Breves, P. L., Liebers, N., Abt, M., & Kunze, A. (2019). The perceived fit between Instagram influencers and the endorsed brand: How influencer–brand fit affects source credibility and persuasive effectiveness. Journal of Advertising Research, 59(4), 440–454. https://doi.org/10.2501/JAR-2019-030
Burgoon, J. K. (2015). Expectancy violations theory. In The International
Encyclopedia of Interpersonal Communication (pp. 1–9). John Wiley & Sons,
Ltd. https://doi.org/10.1002/9781118540190.wbeic102
Campbell, C., & Evans, N. J. (2018). The role of a companion banner and sponsorship transparency in recognizing and evaluating article-style native advertising.
Journal of Interactive Marketing, 43, 17–32.
https://doi.org/10.1016/j.intmar.2018.02.002
Carr, C. T., & Hayes, R. A. (2014). The effect of disclosure of third-party influence on an opinion leader’s credibility and electronic word of mouth in two-step flow.
Journal of Interactive Advertising, 14(1), 38–50.
https://doi.org/10.1080/15252019.2014.909296
Choi, S. M., & Rifon, N. J. (2002). Antecedents and consequences of web advertising credibility. Journal of Interactive Advertising, 3(1), 12–24.
https://doi.org/10.1080/15252019.2002.10722064
Chung, S., & Cho, H. (2017). Fostering parasocial relationships with celebrities on social media: Implications for celebrity endorsement. Psychology & Marketing, 34(4), 481–495. https://doi.org/10.1002/mar.21001
Corritore, C. L., Kracher, B., & Wiedenbeck, S. (2003). On-line trust: Concepts, evolving themes, a model. International Journal of Human-Computer Studies, 58(6), 737–758. https://doi.org/10.1016/S1071-5819(03)00041-7
De Cicco, R., Iacobucci, S., & Pagliaro, S. (2021). The effect of influencer–product fit on advertising recognition and the role of an enhanced disclosure in increasing sponsorship transparency. International Journal of Advertising, 40(5), 733–759.
https://doi.org/10.1080/02650487.2020.1801198
De Veirman, M., & Hudders, L. (2020). Disclosing sponsored Instagram posts: The role of material connection with the brand and message-sidedness when disclosing covert advertising. International Journal of Advertising, 39(1), 94–
130. https://doi.org/10.1080/02650487.2019.1575108
Dekker, K., & Reijmersdal, E. A. van. (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.
https://doi.org/10.1080/10496491.2013.769473
Deng, X., Li, M., & Suh, A. (2020). Recommendation or advertisement? The influence of advertising-disclosure language with pictorial types on influencer credibility and consumers’ brand attitudes. In HCI International 2020 – Late Breaking Papers: Interaction, Knowledge and Social Media (pp. 234–248).
Springer International Publishing. https://doi.org/10.1007/978-3-030-60152- 2_19
Dhanesh, G. S., & Duthler, G. (2019). Relationship management through social media
influencers: Effects of followers’ awareness of paid endorsement. Public Relations Review, 45(3), 101765. https://doi.org/10.1016/j.pubrev.2019.03.002
Eisend, M., van Reijmersdal, E. A., Boerman, S. C., & Tarrahi, F. (2020). A meta- analysis of the effects of disclosing sponsored content. Journal of Advertising, 49(3), 344–366. https://doi.org/10.1080/00913367.2020.1765909
Evans, N. J., Phua, J., Lim, J., & Jun, H. (2017). Disclosing Instagram influencer advertising: The effects of disclosure language on advertising recognition, attitudes, and behavioral intent. Journal of Interactive Advertising, 17(2), 138–
149. https://doi.org/10.1080/15252019.2017.1366885
Evans, N. J., Wojdynski, B. W., & Grubbs Hoy, M. (2019). How sponsorship
transparency mitigates negative effects of advertising recognition. International Journal of Advertising, 38(3), 364–382.
https://doi.org/10.1080/02650487.2018.1474998
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
Giuffredi-Kähr, A., Petrova, A., & Malär, L. (2022). Sponsorship disclosure of influencers—A curse or a blessing?
https://doi.org/10.1177/10949968221075686, 10.5167/uzh-217836
Horton, D., & Richard Wohl, R. (1956). Mass Communication and Para-Social Interaction. Psychiatry, 19(3), 215–229.
https://doi.org/10.1080/00332747.1956.11023049
Hsu, L.-C. (2021). Effect of eWOM review on beauty enterprise: A new interpretation of the attitude contagion theory and information adoption model. Journal of Enterprise Information Management, 35(2), 376–413.
https://doi.org/10.1108/JEIM-07-2020-0261
Hu, L., Min, Q., Han, S., & Liu, Z. (2020). Understanding followers’ stickiness to digital influencers: The effect of psychological responses. International Journal of Information Management, 54, 102169.
https://doi.org/10.1016/j.ijinfomgt.2020.102169
Jung, A.-R., & Heo, J. (2019). Ad disclosure vs. ad recognition: How persuasion knowledge influences native advertising evaluation. Journal of Interactive Advertising, 19(1), 1–14. https://doi.org/10.1080/15252019.2018.1520661
Lin, H.-C., Bruning, P. F., & Swarna, H. (2018). Using online opinion leaders to promote the hedonic and utilitarian value of products and services. Business Horizons, 61(3), 431–442. https://doi.org/10.1016/j.bushor.2018.01.010
Lou, C. (2021). Social media influencers and followers: theorization of a trans- parasocial relation and explication of its implications for influencer advertising.
Journal of Advertising, ahead-of-print(ahead-of-print), 1–18.
https://doi.org/10.1080/00913367.2021.1880345
MacKenzie, S. B., & Lutz, R. J. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing, 53(2), 48–65. https://doi.org/10.2307/1251413
MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the
ad as a mediator of advertising effectiveness: A test of competing explanations.
Journal of Marketing Research, 23(2), 130-. https://doi.org/10.2307/3151660
Martínez-López, F. J., Anaya-Sánchez, R., Fernández Giordano, M., & Lopez-Lopez, D. (2020). Behind influencer marketing: Key marketing decisions and their effects on followers’ responses. Journal of Marketing Management, 36(7–8), 579–607. https://doi.org/10.1080/0267257X.2020.1738525
Metzger, M. J., Flanagin, A. J., & Medders, R. B. (2010). Social and heuristic approaches to credibility evaluation online. Journal of Communication, 60(3), 413–439. https://doi.org/10.1111/j.1460-2466.2010.01488.x
Pfeuffer, A., & Huh, J. (2020). Effects of different sponsorship disclosure message types on consumers’ trust and attitudes. International Journal of Advertising, 0(0), 1–32. https://doi.org/10.1080/02650487.2020.1807229
Pfeuffer, A., Lu, X., Zhang, Y., & Huh, J. (2020). The effect of sponsorship
disclosure in YouTube product reviews. Journal of Current Issues & Research in Advertising, 0(0), 1–23. https://doi.org/10.1080/10641734.2020.1859023
Ridings, C. M., Gefen, D., & Arinze, B. (2002). Some antecedents and effects of trust in virtual communities. The Journal of Strategic Information Systems, 11(3), 271–295. https://doi.org/10.1016/S0963-8687(02)00021-5
Rieh, S. Y., & Danielson, D. R. (2007). Credibility: A multidisciplinary framework.
Annual Review of Information Science and Technology, 41(1), 307–364.
https://doi.org/10.1002/aris.2007.1440410114
Seo, Y., Kim, J., Choi, Y. K., & Li, X. (2019). In “likes” we trust: Likes, disclosures
and firm-serving motives on social media. European Journal of Marketing, 53(10), 2173–2192. https://doi.org/10.1108/EJM-11-2017-0883
Sokolova, K., & Kefi, H. (2020). Instagram and YouTube bloggers promote it, why should I buy? How credibility and parasocial interaction influence purchase intentions. Journal of Retailing and Consumer Services, 53, 101742-.
https://doi.org/10.1016/j.jretconser.2019.01.011
Stubb, C., & Colliander, J. (2019). “This is not sponsored content” – The effects of impartiality disclosure and e-commerce landing pages on consumer responses to social media influencer posts. Computers in Human Behavior, 98, 210–222.
https://doi.org/10.1016/j.chb.2019.04.024
Tukachinsky, R., & Stever, G. (2019). Theorizing development of parasocial engagement. Communication Theory, 29(3), 297–318.
https://doi.org/10.1093/ct/qty032
Urban, G. L., Amyx, C., & Lorenzon, A. (2009). Online trust: State of the art, new frontiers, and research potential. Journal of Interactive Marketing, 23(2), 179–
190. https://doi.org/10.1016/j.intmar.2009.03.001
van Dam, S., & van Reijmersdal, E. A. (2019). Insights in adolescents’ advertising literacy, perceptions and responses regarding sponsored influencer videos and disclosures. Cyberpsychology, 13(2). https://doi.org/10.5817/CP2019-2-2
van Reijmersdal, E. A., Boerman, S. C., Buijzen, M., & Rozendaal, E. (2017). This is advertising! Effects of disclosing television brand placement on adolescents.
Journal of Youth and Adolescence, 46(2), 328–342.
https://doi.org/10.1007/s10964-016-0493-3
Visentin, M., Pizzi, G., & Pichierri, M. (2019). Fake news, real problems for brands:
The impact of content truthfulness and source credibility on consumers’
behavioral intentions toward the advertised brands. Journal of Interactive Marketing, 45, 99–112. https://doi.org/10.1016/j.intmar.2018.09.001
Weitzl, W. (2016). Conceptualization of the construct, research questions and
hypotheses. In Measuring Electronic Word-of-Mouth Effectiveness (pp. 95–162).
Springer Fachmedien Wiesbaden. https://doi.org/10.1007/978-3-658-15889-7_3 Wojdynski, B. W. (2016). The deceptiveness of sponsored news articles: How readers
recognize and perceive native advertising. The American Behavioral Scientist (Beverly Hills), 60(12), 1475–1491. https://doi.org/10.1177/0002764216660140
Wojdynski, B. W., & Evans, N. J. (2016). Going native: Effects of disclosure position and language on the recognition and evaluation of online native advertising.
Journal of Advertising, 45(2), 157–168.
https://doi.org/10.1080/00913367.2015.1115380
Wojdynski, B. W., & Evans, N. J. (2020). The covert advertising recognition and effects (CARE) model: Processes of persuasion in native advertising and other masked formats. International Journal of Advertising, 39(1), 4–31.
https://doi.org/10.1080/02650487.2019.1658438
Wu, P. C. S., & Wang, Y. (2011). The influences of electronic word‐of‐mouth message appeal and message source credibility on brand attitude. Asia Pacific Journal of Marketing and Logistics, 23(4), 448–472.
https://doi.org/10.1108/13555851111165020
Appendix Scales Message Credibility
你同意下列形容词能准确形容 Kakakaoo-（阿沁）发的微博吗？请点击下拉列表进选择。
完全不同
意 不同意 有些不同
意 中立 有些同意 同意 完全同意 可信的
真实的 准确的
Brand Attitudes
你认为 Kakakaoo-微博中的品牌（泡泡玛特）是：
好的 坏的
令人不 快的
令人愉 悦的
喜欢 不喜欢
正面 负面
有利的 不利的
品质好 的
品质差 的
Brand Familiarity
你在多大程度上同意以下说法？
完全不同
意 不同意 有些不同
意 中立 有些同意 同意 完全同意 我熟悉泡
泡玛特这 个品牌
PSR
你在多大程度上同意下列关于 Kakakaoo-的说法？
完全不同
意 不同意 有些不同
意 中立 有些同意 同意 完全同意 1. 她让
我感觉很 舒适，就 好像一个 朋友
o o o o o o o
2. 我愿 意和她友
好地交谈
o o o o o o o
3. 我想 我挺理解
她的
o o o o o o o
4. 如果 她是个普 通人，我 们或许会 成为朋友
o o o o o o o
5. 当她 做出某些 行为时，
我想我能 理解背后 的原因
o o o o o o o
6. 某些 情况下，
我能感受 到她的情
绪
o o o o o o o
7. 她似 乎了解我 想知道的
事情
o o o o o o o
Influencer Familiarity
你在多大程度上同意以下说法？
完全不
同意 不同意 有些不
同意 中立 有些同
意 同意 完全同
意 我熟悉
Kakakaoo- 8. 她让 我想起自
己
o o o o o o o
9. 我能 和她产生
共鸣
