Celebrity Endorsement on Social Media and Its Effects on Brand
Attitude and Attitude Towards the Celebrity
Does the level of scepticism influence people’s attitudes when exposed to celebrity endorsement posts?
Yorrian Mathot 10002988
Master’s Thesis Persuasive Communication
Graduate School of Communication Master’s programme Communication Science
Abstract
The aim of this study was to investigate the effects of celebrity endorsement on social media. A distinction was made between implicit and explicit celebrity endorsement. The effect of type of celebrity endorsement on consumers’ brand attitude and attitude towards the celebrity was examined. Exposure to a persuasive message, like a social media post
containing celebrity endorsement, is expected to activate people’s persuasion knowledge. Especially, when explicit celebrity endorsement is used. Explicit endorsement is expected to lead to more activated persuasion knowledge than implicit endorsement. It is assumed that activation of persuasion knowledge is more likely to occur for people who are highly sceptical. High levels of scepticism combined with the activation of persuasion knowledge might lead to more negative attitudes. An online experiment (n=126) was conducted to test these assumptions. Participants were exposed to a social media post containing celebrity
endorsement and were asked to fill in a questionnaire containing questions about this celebrity endorsement, their attitudes towards the celebrity, the brand and advertising in general. The expected moderated mediation effect, was not confirmed by the results of this research. Instead the results showed significant effects of consumers’ scepticism on persuasion
knowledge and of persuasion knowledge on brand attitude and attitude towards the celebrity. No relationship between celebrity endorsement and brand attitude and attitude towards the celebrity was found. Type of celebrity endorsement does not seem to influence the activation of persuasion knowledge. Persuasion knowledge is more likely to be activated when people are highly sceptical. High levels of scepticism and activated persuasion knowledge seem to lead to more negative attitudes towards the celebrity and the brand.
Introduction
Nowadays, more than two billion people are actively using social media (Valentini, 2015). Over the last couple of years, social media has started to play a big role in people’s everyday life (Kaplan & Haenlein, 2010). This also holds true for celebrities. They often use social media to promote themselves and their work. Most celebrities are able to reach a large audience through their social media accounts. This wide reach makes them interesting for marketers. Marketers employ celebrities to promote their brands and products on these celebrities’ personal social media accounts. For advertisers and marketers, it is difficult to reach and convince consumers, because people are exposed to advertising very often and a lot of people are becoming sceptical about advertising (Helm, Moulard & Richins, 2015). In order to gain more access to and attract the attention of consumers, advertisers and marketers are forced to think of creative ways to reach consumers. One way to attract attention is by using celebrity endorsement. People follow celebrities on social media because they are interested in their lifestyles. As a result, these followers will pay more attention to the posts these celebrities will publish. By tapping into these celebrities’ posts marketers and
advertisers can reach and attract the attention of these followers.
For example, when popstar Nick Jonas posted a picture on social media in which he thanked the company Lyft for the ride to the Hamptons. As became known later, the ride was paid for by the company (Notopoulos, 2016). So, indirectly the company paid Nick Jonas to promote their brand. This new form of ‘brand placement’, i.e. integrating brands or products into editorial media content (Van Reijmersdal, Smit & Neijens, 2010), is an interesting opportunity for marketers to reach consumers online. In this example, the brand was placed on a celebrity’s personal social media account. This new way of reaching consumers is interesting for marketers, because the awareness of the persuasive intent of the brand or company is lower when celebrity endorsement is used (Bhatnagar, Aksoy & Malkoc, 2004;
Dekker & Van Reijmersdal, 2010; Silvera & Austad, 2004). Still, it is important to note that celebrity endorsement might not only positively influence consumers, but can also create negative effects, such as consumers’ resistance (Doss, 2011). It is, therefore, important for advertisers to know how and when to use celebrity endorsement to create the desired effect.
How celebrities endorse a product or brand (the characteristics of the endorsement) on social media can influence how the product or brand is received by consumers. Two types of celebrity endorsement can be distinguished: implicit celebrity endorsement and explicit celebrity endorsement. Implicit celebrity endorsement is very subtle and requires only the showing of the brand or product on the celebrity’s personal social media page. For example, posting a picture while wearing a sweater of the endorsed brand, without mentioning the brand. The celebrity is endorsing the brand by doing this but it is not that obvious because the brand is not mentioned. It is also possible to explicitly promote the brand or its product by very clearly mentioning the brand in the content of the post. Like publishing a post with products from the endorsed brand and saying: ‘Love this [endorsed brand] gear, always such good quality!’. Endorsing a brand like this is much less subtle and draws more attention. With this type of endorsement, it is more likely people will recognize the persuasive intent of the message.
This research focuses on the use of celebrity endorsement on the personal social media accounts of these celebrities. Celebrity endorsement can influence people’s attitude towards a celebrity and a promoted brand (Dean & Biswas, 2001; Erdogan, 1999; Lafferty, Goldsmith & Newell, 2002). When the endorsement becomes too obvious, which is more likely to occur with explicit celebrity endorsement, people will probably see the post as advertising. When people are aware of the persuasive intent of a message they will address this message more critically, especially when they were already sceptical towards advertising. Scepticism is generated in evaluating and coping with advertising messages (Bailey, 2007). When people
are already highly sceptical towards advertising, they will most likely use defensive strategies (Burgoon, Alvara, Grandpre & Voulodakis, 2002). People with a higher level of scepticism, will be more likely to activate their persuasion knowledge when exposed to a persuasion attempt. Being exposed to a persuasive message in combination with consumers’ scepticism, can lead to activation of people’s persuasion knowledge. Persuasion knowledge can, in effect, have an influence on people’s attitudes towards the celebrity in the persuasive message and towards the promoted brand. Consequently, this study will focus on the following research question: ‘In what way does implicit or explicit celebrity endorsement influence brand attitude and attitude towards the celebrity and does this relationship depend on consumers’ scepticism towards advertising and consumers’ persuasion knowledge?’
Theoretical Background
Celebrity Endorsement and Persuasion Knowledge
Celebrity endorsement can be an effective strategy for advertisers and marketers to try to attract consumers’ attention (Belch & Belch, 2015). Traditionally, celebrity endorsement is defined as the use of a celebrity in an advertising campaign, published by the brand.
Advertisements containing celebrity endorsement showed to have a better chance of breaking through the advertising clutter (Choi & Rifon, 2007). Celebrity endorsement is often used by marketers because the persuasive intent of the message attracts less attention. Consumers often do not directly recognize the persuasive intent of celebrity endorsement, because most attention goes to the celebrity (Bhatnagar, Aksoy & Malkoc, 2004; Dekker & Van
Reijmersdal, 2010; Silvera & Austad, 2004). This allows advertisers to conceal the fact that the message is actually advertising. In the present research, we focus on a different type of celebrity endorsement, it can be seen as a combination of brand placement and online celebrity endorsement. The brand is promoted by the celebrity, on the celebrity’s own social media account. It is expected that the perceived persuasive intent of the celebrity endorsement
on his or her social media account is lower than regular celebrity endorsement. Therefore, we decided to focus on celebrity endorsement on their personal social media accounts.
Previous research shows that consumers do not have favorable attitudes towards advertising on social media (Kelly, Kerr & Drennan, 2010) and towards advertising in general (Brehm, 1966; Calfee & Ringold, 1994). As a result, marketers need to find ways to reach people without these people becoming critical. We expect that using celebrities’ own social media accounts, will make consumers less critical towards the persuasive content. Consumers usually immediately perceive advertising posted by brands as persuasive. For advertisements where celebrities endorse a brand, it is often believed that these celebrities are genuine about their opinion about the product (Silvera & Austad, 2004). Consequently, it can be argued that consumers do not directly recognize the persuasive intent of the celebrity endorsement. When the endorsement occurs on the celebrity’s personal social media account, it can be argued that people will not directly perceive it as persuasive. Which can result in consumers being less critical about the content of the post. This type of advertising, the promotion of products or brands on the celebrity’s personal social media account, can be seen as a form of brand placement. The brand or its products are placed within a non-commercial setting (the social media account of the celebrity) to influence the receiver, without disclosing the persuasive intention (Tiwsakul, Hackley & Szmigin, 2005). Brand placement can be seen as a subtle way of advertising; it creates positive effects on brand attitude (Balasubramanian, Karrh &
Patwardhan, 2006). When celebrity endorsement is subtle, it is expected that it would create the same effect. Subtly endorsed brands and products on celebrities’ personal social media accounts, are expected to create positive effects on brand attitude as well. Therefore, in the current research we decided to study the effects of implicit compared to explicit celebrity endorsement.
When celebrity endorsement is explicit, the persuasive intent might become more obvious, causing people to see the celebrity endorsement as advertising. People might pay more attention to the endorsement in the post, which results in these people noticing the selling intent of the endorsement and becoming more critical (Cowley & Barron, 2008). The research of Boerman, Van Reijmersdal, and Neijens (2012) also showed that when people become more aware of the persuasive intent of a message, they are more inclined to process this message critically. Therefore, it is argued that people are more likely to process the explicit endorsement message more critically than the implicit endorsement message. This critical processing makes people more aware of the persuasive intent of a message containing explicit celebrity endorsement than implicit celebrity endorsement. A higher level of
awareness of the persuasive intent, will lead to the activation of persuasion knowledge, which, in turn, will make consumers more critical about the message. Persuasion knowledge refers to theories and beliefs consumers have about persuasion and marketers’ motives, strategies and tactics (Van Reijmersdal, Lammers, Rozendaal & Buijzen, 2014). Persuasion knowledge is expected to play an important role in the effects of celebrity endorsement and brand
placement.
Persuasion knowledge is people’s knowledge about the source and persuasive intent of a message and how to cope with the persuasive attempt. The Persuasion Knowledge Model (Friestad & Wright, 1994) emphasizes that people have to be aware of a persuasive attempt in order to activate their persuasion knowledge. Therefore, higher levels of attention towards the advertising will activate persuasion knowledge (Campbell, 1995). This was confirmed by the results of Tutaj and Van Reijmersdal (2012). They studied the effects of two types of online advertising formats; sponsored content and banner advertising. The results showed that people are more likely to recognize banners as advertising than sponsored content because the
and prominent formats of advertising than subtle online ad formats. The more people
recognize the format of the message, the more they understand the persuasive intent. When a format is subtle, the ad was found to be more informative, more amusing and less irritating (Tutaj & Van Reijmersdal, 2012). However, when the persuasive intent is more obvious, it might lead to the activation of people’s persuasion knowledge. Wei, Fischer and Main (2008) showed that explicit product placement made viewers more aware of the persuasive and commercial intent. This awareness led to activated persuasion knowledge. Consequently, it can be argued that the persuasive intent of explicit celebrity endorsement on social media is more easily recognized than the persuasive intent of implicit celebrity endorsement, because with explicit celebrity endorsement the persuasive intent is more obvious (just like the
banners ads) than with implicit celebrity endorsement. So, explicit celebrity endorsement will probably lead to more activated persuasion knowledge than implicit celebrity endorsement. Therefore, the following hypothesis was created:
H1: Exposure to explicit celebrity endorsement activates persuasion knowledge to a greater extent compared to exposure to implicit celebrity endorsement.
Celebrity Endorsement and Persuasion Knowledge Affecting Attitudes
Messages with implicit celebrity endorsement disguise the persuasive intent, whereas messages containing explicit celebrity endorsement have a more visible persuasive intent. Consequently, the persuasive intent of explicit endorsement is more likely to be noticed. Awareness of the persuasive intent will lead to the activation of persuasion knowledge. This persuasion knowledge, in turn, can influence people’s attitudes towards the message, the product promoted and the spokesperson (Cowley & Barron, 2008). When messages are considered as advertisement, the effect on brand reputation will be different than when the
persuasive intent is not recognized. The more obvious the persuasive intent becomes, the more negative the brand attitude will be (Boerman, Van Reijmersdal & Neijens, 2015). This research will focus on people’s attitudes towards the celebrity (endorsing the brand) and consumers’ brand attitude.
When people’s persuasion knowledge is activated, the brand placement will interrupt their viewing experience, according to the Persuasion Knowledge Model by Friestad and Wright (1994). This interruption is expected to cause negative feelings, which can negatively impact people’s behavior and attitudes (Boerman, Van Reijmersdal & Neijens, 2012; Cowley & Barron, 2008; Van Reijmersdal, et al., 2014). It can be argued that when people are
scrolling through their social media feed, their viewing experience will be interrupted by the celebrity endorsement, and their persuasion knowledge will get activated. The researches of Boerman, Van Reijmersdal and Neijens (2012; 2014; 2015) about disclosures in television advertising also showed that the recognition of the persuasive intent of a message, activates persuasion knowledge, which can make people more critical. They showed a mediating effect of persuasion knowledge on brand attitude. Higher persuasion knowledge led to more
negative brand attitudes. This can be explained by the assumption that people are aware of the fact they are being influenced and they do not like to be persuaded and told what to do
(Brehm, 1966). It is argued that people’s brand attitude and attitude towards the celebrity will be more negative after persuasion knowledge is activated, which is most likely to occur when exposed to explicit celebrity endorsement. This leads to the following hypotheses:
H2a: There is a negative correlation between persuasion knowledge and brand attitude and attitude towards the celebrity.
H2b: The effect of celebrity endorsement on brand attitude and attitude towards the celebrity is mediated by the activation of persuasion knowledge.
Consumers’ Scepticism towards Celebrity Endorsement
People differ in the extent to which they are sceptical towards advertising. Scepticism is a psychological and personal character trait. Obermiller and Spangenberg (1998) described scepticism as the general tendency of disbelief about advertising claims and basic marketplace beliefs that vary across persons and are related to general persuasion. This sceptical tendency may vary within each person, due to the fact that scepticism is based on personal views about the marketplace. Scepticism can influence the level of defensive strategies people will use against the persuasive attempt. People are more likely to activate their persuasion knowledge when they are highly sceptical (Sher & Lee, 2009). Being sceptical makes people more critical towards messages, especially when these messages are intended to persuade them. People do not like to be told what to do and feel like they are being persuaded (Brehm, 1966). When a message tells people what to do, they will be more critical about this persuasive message (Knowles & Linn, 2004). People who are highly sceptical are more likely to perceive a message as persuasive, because they will process the message more critically (Helm,
Moulard and Richins, 2015). Highly sceptical consumers are more likely to recognize the persuasive intent of a message. When people are aware of the persuasive intent, they will be more prone to activate their persuasion knowledge. Sceptical people are more aware of the content of a message. The recognition of the persuasive intent of a message activates
persuasion knowledge (Van Reijmersdal, Neijens & Smit, 2005). Therefore, sceptical people are more prone to activate persuasion knowledge. When people are highly sceptical and have activated their persuasion knowledge, they are difficult to reach for marketers. The research of Van Reijmersdal, Tutaj and Boerman (2005) confirms these views. It is argued that when
people are more critical towards a message, they will be more likely to notice the persuasive intent, which will lead to the activation of persuasion knowledge. As mentioned earlier, explicit celebrity endorsement is more likely to be perceived as advertising. People who are aware of the persuasive intent of explicit celebrity endorsement, are more likely to activate their persuasion knowledge, this especially happens when people are highly sceptical.
H3a: The effect of celebrity endorsement on persuasion knowledge is stronger for people with a high level of scepticism than for people with a lower level of scepticism.
As mentioned above, it is argued that people who are highly sceptical will be more likely to activate their persuasion knowledge when they are exposed to explicit (versus implicit) celebrity endorsement. Activated persuasion knowledge, resulting from high scepticism, will negatively influence consumers’ attitudes (Boerman, Van Reijmersdal & Neijens, 2012; Cowley & Barron, 2008; Van Reijmersdal, et al., 2014). People who are exposed to explicit celebrity endorsement are expected to be more aware of the persuasive intent of the message than people who are exposed to implicit celebrity endorsement, because explicit celebrity endorsement is more prominent. This will especially be true for people who are highly sceptical. Highly sceptical consumers, who are exposed to explicit celebrity endorsement, will be most likely to activate their persuasion knowledge (Sher & Lee, 2009). The relationship between the different concepts mentioned in the theoretical framework will be shown below in Figure 1. As mentioned earlier, the activation of the persuasion knowledge is expected to lead to a more negative brand attitude and attitude towards the celebrity (Gupta, Balasubramanian & Klassen, 2000). This all leads to the last hypothesis:
H3b: The effect of exposure to celebrity endorsement on brand and celebrity attitude, which is mediated by persuasion knowledge, is moderated by consumers’ scepticism such that the effect is stronger for people with high versus low scepticism.
Figure 1: Conceptual model
Methods
Participants and Design
An online experimental study was conducted among 185 participants. 63 participants did not fully complete all the questions, they were excluded from the analyses. As a result, the research was conducted among 126 participants between 18-67 years old (M = 31.44, SD = 11.36), with 69.8% being female. 83.3% of the participants was highly educated (e.g. HBO or WO educated). The participants were randomly assigned to one of the two conditions, the explicit celebrity endorsement condition or the implicit celebrity endorsement condition, in a between-subjects design. The moderator in this research, the level of consumers’ scepticism, was measured as a continuous variable. Participants were recruited via the social media of the researcher and via email. Participating in the research was voluntary.
Procedure
In this research a between-subjects experiment was conducted to test whether the use of implicit or explicit celebrity endorsement influenced brand attitude and attitude towards the celebrity via the activation of persuasion knowledge. Through different types of social media
Celebrity Endorsement Consumers’ Scepticism Persuasion Knowledge
Brand Attitude & Attitude Towards
and email participants received a link they could use to participate in the research. After they were informed about the research and agreed with the terms and conditions, participants were randomly exposed to either the explicit celebrity endorsement or the implicit celebrity
endorsement. Next, participants were asked to answer a couple of questions about their brand attitude. Questions about two other brands were added as a cover. Then, participants’ attitude toward the celebrity was asked, followed by some questions about their persuasion knowledge. Finally, their scepticism was measured by answering questions about their view on
advertising, followed by the manipulation check and some question about their demographics (i.e. age, gender, education).
Celebrity Endorsement
For the manipulation of the independent variable (celebrity endorsement) an already existing social media post was edited. The celebrity used for the endorsement was the Dutch top model Doutzen Kroes. The brand she was endorsing, L’Oreal Paris, is a big international brand for beauty products. Doutzen Kroes is the face of many of their campaigns. An already existing social media post of Doutzen, promoting L’Oreal Paris, was used. With Photoshop the post was edited and two conditions were created. Both posts were showing the same image and were exactly the same on all characteristics. Only the type of endorsement differed; implicit or explicit celebrity endorsement. In one post, the picture had a caption, saying; ‘L’Oreal always does the trick, it makes my hair shine like a diamond #lorealparis’. This caption was added in the explicit celebrity endorsement condition. It clearly mentioned the brand that was endorsed in the social media post. In the implicit celebrity endorsement condition the photo was the same, but the caption did not mention the brand or products, it said; ‘Another day at the office #worklife #workhard’. No emphasis was put on the brand or product in the photo. For both conditions the pictures and profile were exactly the same except for the statements next to the photo.
To test whether participants paid attention to the manipulation a pre-test was
conducted. This pre-test was conducted among a convenience sample of 19 higher educated college students (M = 25.25, SD = 1.75) of which 64.7% was female. The pre-test was an online survey experiment. Participants were asked which statement they had seen in the social media post. They had to choose one of five statements. After that, they were asked which celebrity and brand they had seen and if they were familiar with the celebrity and brand before the current research. The results from the pre-test showed that participants actually had seen the manipulation. In the explicit celebrity endorsement condition 87.5% of the
respondents saw the statement. In the implicit celebrity endorsement 88.9% of the
respondents noticed the statement. Based on these results it was decided that the manipulation created for this study did not have to be edited.
Manipulation check
At the end of the questionnaire, participants were asked to indicate on a scale from one to seven, to what extent they agreed (1 = strongly disagree; 7 = strongly agree) with the item: ‘The social media post I just saw, was explicitly promoting a brand’. This was done to check whether the participants noticed the manipulation and evaluated it as being explicit or not (M = 3.34, SD = 2.31).
Brand Attitude and Attitude towards the Celebrity
The dependent variables brand attitude and attitude towards the celebrity were
measured with semantic differential seven-point scales created by Silvera and Austad (2004). The different items used scales with a value of one, associated with the more negative word and a value of seven, associated with the more positive word. The scales were adjusted for the purpose of this study, some items were removed and others were added to make the scales more relevant. The brand attitude was measured using the following five items: ‘desirable/not desirable’, ‘likeable/not likeable’, ‘favorable/unfavorable’, ‘good/bad’ and ‘high quality/poor
quality’. A factor analysis showed that the five items created a one-dimensional scale: there was only one component with an eigen value higher than one (EV = 4.12). All items are positively correlated with this first component. This component explains 82.37% of the variance, Cronbach’s Alpha = .94, which shows that the scale was very reliable (M = 5.04, SD = 1.13).
The attitude towards the celebrity was measured using the items ‘interesting/uninteresting’, ‘likeable/not likeable’, ‘trustworthy/untrustworthy,
‘sincere/insincere’ and ‘good/bad’. A factor analysis was conducted, showing that the five items created a one-dimensional scale: there was only one component with an eigen value higher than one (EV = 3.63). There was a positive correlation between all items with this first component. The component explains 73.63% of the variance, Cronbach’s Alpha = .90, which shows the scale was very reliable as well (M = 4.86, SD = 1.17).
Persuasion Knowledge
To measure persuasion knowledge, items from the study of Boerman et al. (2012) were used. Participants’ critical feelings towards the post containing celebrity endorsement were measured with a seven-point attitudinal persuasion knowledge scale. Participants had to describe to what extent they agreed (1= strongly disagree, 7 = strongly agree) with statements like: ‘I think the social media post I just saw, is…’ followed by the items: ‘honest’,
‘trustworthy’, ‘convincing’, ‘biased’ and ‘not credible’. Another factor analysis was
conducted for the five items measuring persuasion knowledge. The five items created a one-dimensional scale: there was only one component with an eigen value higher than one (EV = 3.43). All items were positively correlated with this first component. 68.54% of the variance was explained by this component, Cronbach’s Alpha = .88, which shows that the scale was very reliable (M = 3.50, SD = 1.12).
Consumers’ scepticism was measured with nine items on a five-point Likert scale (1 = strongly disagree; 5 = strongly agree) created by Obermiller and Spangenberg (1998).
Respondents had to describe to what extend they agreed with statements like: ‘Advertising is generally truthful’, ‘Advertising is a reliable source of information’, ‘In general, advertising presents a true picture of the product being advertised’ and ‘I feel I’ve been accurately informed after viewing most advertisements’. These statements gave an overall view how participants think about advertising and to what extent they are sceptical about advertising. These scales have shown to be reliable for measuring consumers’ scepticism in the research of Obermiller and Spangenberg (1998). A factor analysis was conducted, the analysis showed that the nine items used to measure scepticism created a one-dimensional scale: there was only one component with an eigen value higher than one (EV = 5.86). All nine items were positively correlated with this first component. The component explains 65.12% of the
variance, Cronbach’s Alpha = .93, which shows that the scale was very reliable (M = 2.76, SD = 1.12).
Finally, participants had to answer a couple of demographic questions (i.e. age, gender, education). At the end of the questionnaire, participants were asked to tell their suspicion about the subject of the research. This was done to be able to exclude participants who had too much prior knowledge about the purpose of the study. None of the participants exactly knew what the research was about. Therefore, nobody had to be excluded.
Results
Manipulation Checks
An independent samples t-test was conducted to check whether the participants who saw the explicit celebrity endorsement, perceived this endorsement as being more explicit than participants who saw the implicit celebrity endorsement. A significant difference was found between the people who were in the implicit celebrity endorsement condition (M = 1.14,
SD = 0.44) and people who were in the explicit celebrity endorsement condition (M = 5.54, SD = 0.88); t (124) = -35.67, p < .001, 95% CI [-4.64, -4.15]. This suggests that people who have seen the explicit celebrity endorsement post actually perceived this post as being more explicit than people who had seen the implicit celebrity endorsement post. Further analyses showed that people in both conditions did not significantly differ on demographic
characteristics. No significant difference was found in age between the implicit celebrity endorsement condition (M = 31.87, SD = 11.89) and the explicit celebrity endorsement condition (M = 31.00, SD = 10.89); t (124) = .43, p = .668, 95% CI [-3.15, 4.89]. For
education no significant difference between the implicit condition (M = 5.37, SD = 1.20) and the explicit condition (M = 5.40, SD = 1.19) was found either; t (124) = -.15, p = .881, 95% CI [-0.45, 0.39]. This was also the case for gender 2
(1) = .60, p = .44. So, the randomization was successful and both groups can be seen as equal.
Testing Hypotheses
The PROCESS macro of Hayes (2013, Model 7) was used to test the five hypotheses in the moderated mediation model. The possibility to test conditional indirect effects was offered by this macro. It also provides confidence intervals based on bootstrapping for the mediated effect at the two levels of the moderator.
The first analysis showed that there was no significant direct effect of celebrity endorsement on persuasion knowledge, t (126) = -.43, p = .671, SE = .18 with coefficient a1i = -.0775.
Consequently, it can be concluded that celebrity endorsement does not have a significant effect on persuasion knowledge, which means Hypothesis H1 will be rejected. The second analysis did give a significant effect, from consumers’ scepticism on persuasion knowledge, t (126) = 5.64, p < .001, SE = .08 with coefficient a2i = .4669. This means that consumers’
scepticism has a direct effect on persuasion knowledge. No significant interaction effect was found between celebrity endorsement and consumers’ scepticism on persuasion knowledge; t
(126) = .87, p = .385, SE = .17 with coefficient a3i = .1448. Which results in the conclusion
that there is no significant interaction effect between celebrity endorsement and consumers’ scepticism on persuasion knowledge. Therefore, Hypothesis H3a will also be rejected. Then the effect of persuasion knowledge on attitude towards the celebrity was analyzed. The analysis showed that there was a significant relationship between persuasion knowledge and attitude towards the celebrity, t (126) = 3.85, p < .001, SE = .12 with coefficient bi = .4085.
When analyzing the effect of persuasion knowledge on brand attitude, the relationship was also significant, t (126) = 2.89, p =.005, SE = .13 with coefficient bi = .3664. This leads to the
conclusion that there is a significant relationship between persuasion knowledge and brand attitude and attitude towards the celebrity. Therefore, Hypothesis H2a can be confirmed. After that, the effect of celebrity endorsement on the attitude towards the celebrity was examined, it showed that this direct effect was not significant, t (126) = -1.16, p = .249, SE = .20 with coefficient ci = -.2271. Celebrity endorsement does not have a significant direct
effect on attitude towards the celebrity. For brand attitude, the analysis was also examined and showed there was no significant effect either, t (126) = -.62, p = .534, SE = .19 with
coefficient bi = -.1200. So, no direct effect from celebrity endorsement on brand attitude was
found. This leads to the rejection of Hypothesis H2b. There were no significant direct effects from celebrity endorsement on brand attitude or attitude towards the celebrity.
Eventually, the moderated mediation effect was analyzed. This analysis showed that there was no significant moderated mediation effect of celebrity endorsement on brand attitude [indirect effect = 0.05, boot SE = 0.07, 90% BCCI (-.07; .20)] and attitude towards the celebrity
[indirect effect = 0.06, boot SE = 0.07, 90% BCCI (-.08; .20)]. Based on these results Hypothesis H3b can also be rejected.
In sum, the results showed that there was no significant moderated mediation effect of celebrity endorsement on brand attitude and attitude towards the celebrity. However, there
were significant relationships between consumers’ scepticism and persuasion knowledge and between persuasion knowledge and brand attitude and attitude towards the celebrity1.
Conclusion
The aim of this study was to investigate whether the effects of celebrity endorsement on brand attitude and attitude towards the celebrity via the activation of persuasion knowledge were moderated by consumers’ scepticism. Tests were conducted to examine the effects of celebrity endorsement on social media. One of the conclusions that can be drawn from the results found in this study, is that the type of celebrity endorsement does not influence
people’s persuasion knowledge. Showing people a social media post containing either implicit or explicit celebrity endorsement, did not lead to different results in the activation of
persuasion knowledge. No effect between celebrity endorsement and brand attitude or attitude towards the celebrity was found either. Whether people saw a social media post containing implicit or explicit celebrity endorsement did not make a difference in the effect on their attitudes towards the brand or the celebrity. Even though, the results of Dean and Biswas (2001) showed that celebrity endorsement does have a positive effect on brand attitude and the study of Balasubramanian, Karrh and Patwardhan (2006) showed that product placement creates a positive effect on brand attitude because of its subtle character. The results of the current study did not confirm these findings. It was expected that implicit celebrity
endorsement, which is more subtle, would have a positive effect on attitude just like brand placement. Nevertheless, no difference between implicit and explicit celebrity endorsement on social media was found. A possible explanation might be participants’ awareness of their participation in the research. This awareness might lead to more critical evaluation of the social media post in either condition. As a result, there might be no difference between people’s attitudes after seeing either type of celebrity endorsement.
However, the results did show a relationship between persuasion knowledge and people’s brand attitude and attitude towards the celebrity. This is in line with the research of Cowley and Barron (2008). Cowley and Barron’s research showed that when persuasion knowledge was activated, people elaborated more on the message and the sender’s intent. The message was interfering with their viewing experience. As a result, people’s attitudes towards the celebrity and the endorsed brand became more negative. The researches of Boerman, Van Reijmersdal and Neijens (2012; 2014; 2015) about disclosures in television advertising
confirm these views. They showed that recognition of the persuasive intent of a message leads to activation of persuasion knowledge. Activated persuasion knowledge will make people more critical. A mediating effect of persuasion knowledge on brand attitude was found. The results of our study are adding value to these views. The relationship between persuasion knowledge and brand attitude and attitude towards the celebrity was confirmed. The activation of persuasion knowledge seems to influence people’s attitude towards the brand and the celebrity.
Another finding, was the effect of consumers’ scepticism on persuasion knowledge. A higher level of scepticism leads to more activation of persuasion knowledge. This confirms the findings from the study of Van Reijmersdal, Neijens and Smit (2005), who stated that sceptical consumers are more likely to activate their persuasion knowledge. Highly sceptical consumers are more likely to recognize the persuasive intent of a message. When people are aware of the persuasive intent, they will activate their persuasion knowledge. Consumers’ scepticism is likely to activate persuasion knowledge, this was confirmed by the results of our study as well. When the persuasive intent of a message gets more obvious, people are more prone to activate their persuasion knowledge. In turn, the activation of persuasion knowledge leads to a higher likelihood of consumers’ attitudes being more negative. This result is also in line with the results from previous research (Boerman, Van Reijmersdal & Neijens, 2012;
Cowley & Barron, 2008; Van Reijmersdal, et al., 2014). People who are more aware of the fact they are being exposed to a persuasive message will be more likely to activate their persuasion knowledge. The activation of persuasion knowledge is shown to lead to more negative attitudes. Our study confirms these findings. We found that people who activate their persuasion knowledge are more likely to have negative attitudes towards the celebrity and the brand. In this research, only implicit or explicit celebrity endorsement on social media did not seem to influence persuasion knowledge and people’s attitudes towards the brand and
celebrity. No relationship was found between the type of celebrity endorsement and the activation of persuasion knowledge or people’s attitudes towards the brand or the celebrity.
To summarize; it can be concluded that people who are more sceptical, have a higher chance of activating their persuasion knowledge and will be more likely to have a negative attitude towards the brand and the celebrity they have seen. The use of implicit celebrity endorsement, in comparison with explicit celebrity endorsement, does not seem to make a difference.
Discussion
This study had some limitations that could have influenced the results. To start with the fact that people were exposed to a single social media post within an artificial setting (participating in a research), which can lead to different reactions towards the celebrity
endorsement because people are paying more attention towards the social media post anyway. When people are scrolling through their social media feed in real-life, they will probably not pay that much attention towards one single post. The higher level of attention might lead to more awareness of the celebrity endorsement, even though this endorsement was implicit. The high level of attention for the social media post can create more awareness towards the
persuasive intent, in either condition. More awareness might lead to higher activation of persuasion knowledge and different results in brand attitude and attitude towards the celebrity
than when people were less aware of the persuasive intent. Even though participants
perceived the social media post in the explicit endorsement condition as being more explicit. The high level of awareness towards the social media post might lead to a more critical evaluation of both social media posts, which will lead to a higher chance of perceiving the implicit celebrity endorsement as persuasive as well. Therefore, for future research, it can be advised to show people social media feed containing a couple of posts and not just one single social media post. This way they will be exposed to more different posts, which will be more realistic and disguise the post containing the celebrity endorsement. It is expected that by doing this, the persuasive intent of the post with the implicit celebrity endorsement is less likely to be noticed than the post containing the explicit celebrity endorsement. At least, by conducting the research like that, the focus will be not only on the social media post
containing the manipulation.
Adding to this, the fact that people are participating in an experiment will lead to them paying more attention towards the message they are exposed to, which will make them more critical towards the content. This higher awareness and critical attitude will influence the results from this study as well. People might score higher on persuasion knowledge and scepticism and score more negatively on brand attitude and attitude towards the celebrity than when they were unaware of the fact they are participating in a research. Even in the implicit celebrity endorsement condition, participants will pay more attention towards the post. These higher levels of attention might make them more critical, leading to higher levels of
scepticism and persuasion knowledge. As a result, their attitudes will become more negative in the implicit endorsement condition, as well as the explicit endorsement condition. Even though, the implicit endorsement condition was expected to have less negative effects on consumers’ attitudes. In future research, it can be interesting to study the effects of social media posts people saw in real-life. It will be interesting to measure their attitude towards
social media posts of celebrities they are already following and check whether they noticed any type of endorsement that might be present and what influence the type of endorsement has on their attitudes.
A highly perceived persuasive intent might also be of influence in this research, because of the manipulation used in the social media post. The picture that was used, was very obviously showing a L’Oreal hairspray. This prominent role in the picture, might make people more aware of the product and brand in the picture. The explicit or implicit
endorsement statement will have less impact, if the brand is already very prominently present. Even though people perceived the explicit celebrity endorsement as being more explicit, it is possible that the obvious presence of the product and brand in the social media post,
influenced people’s attitudes in the implicit celebrity endorsement condition anyway. For future research, it might be better to use a picture with the product or brand being less prominent. There is still some more research needed to draw conclusion about the use of implicit and explicit celebrity endorsement.
Practical Implications
Based on the results of this and earlier studies, it seems important for marketers to pay attention to persuasion knowledge. It is advisable to create an ad that does not activate
people’s persuasion knowledge. This research shows that when consumers’ persuasion knowledge is activated, their attitudes will be more likely to become negative. Persuasion knowledge will be more likely to get activated when people are highly sceptical. Celebrity endorsement did not seem to make a difference. To see whether this was caused by the celebrity endorsement used in this study or celebrity endorsement in general, more research should be done to see whether the different elements of the celebrity endorsement on social media, i.e. the celebrity, the type of product and brand and the social media post itself, influence people’s attitudes differently.
In contrary, for policymakers who are against hidden advertising, the focus should be on making people more sceptical towards advertising. By making people more sceptical, consumers will be more likely to activate their persuasion knowledge. Like the results from this study showed, people with activated persuasion knowledge will be harder to persuade. The activation of the persuasion knowledge will lead to more negative attitudes.
This research is adding to the general thought that consumers’ scepticism will lead to activation of persuasion knowledge. The combination of high levels of scepticism and
activated persuasion knowledge will lead to more negative attitudes towards the brand and the celebrity. The role implicit and explicit celebrity endorsement play in this relationship is not clear. To be able to draw more general conclusions about the use of explicit and implicit celebrity endorsement on social media, more research should be done.
References
Bailey, A. A. (2007). Public information and consumer skepticism effects on celebrity endorsements: Studies among young consumers. Journal of Marketing
Communications, 13(2), 85-107.
Balasubramanian, S. K., Karrh, J. A., & Patwardhan, H. (2006). Audience response to product placements: An integrative framework and future research agenda. Journal of
advertising, 35(3), 115-141.
Bhatnagar, N., Aksoy, L., & Malkoc, S. A. (2004). Embedding brands within media content: The impact of message, media, and consumer characteristics on placement
efficacy. The psychology of entertainment media, 99.
Belch, G. E., & Belch, M. A. (2015). Advertising and promotion: An integrated marketing
communications perspective. The McGraw− Hill.
Boerman, S. C., Reijmersdal, E. A., & Neijens, P. C. (2012). Sponsorship disclosure: Effects of duration on persuasion knowledge and brand responses. Journal of
Communication, 62(6), 1047-1064.
Boerman, S. C., van Reijmersdal, E. A., & Neijens, P. C. (2014). Effects of sponsorship disclosure timing on the processing of sponsored content: A study on the effectiveness of European disclosure regulations. Psychology & Marketing, 31(3), 214-224.
Boerman, S. C., van Reijmersdal, E. A., & Neijens, P. C. (2015). Using eye tracking to understand the effects of brand placement disclosure types in television programs. Journal of Advertising 44(3), 196-207.
Brehm, J. W. (1966). A theory of psychological reactance.
Burgoon, M., Alvara, E., Grandpre, J., & Voulodakis, M. (2002). Revisiting the theory of psychological reactance: Communicating threats to attitudinal freedom. In J.P. Dillard
& M. Pfau (Eds.), The Persuasion Handbook: Developments in Theory and Practice. California, Sage Publications.
Calfee, J. E., & Ringold, D. J. (1994). The 70% majority: Enduring consumer beliefs about advertising. Journal of public policy & marketing, 228-238.
Campbell, M. C. (1995). When attention-getting advertising tactics elicit consumer inferences of manipulative intent: The importance of balancing benefits and investments. Journal
of Consumer Psychology, 4(3), 225-254.
Campbell, M.C., & Kirmani, A. (2000). Consumers' use of persuasion knowledge: The effects of accessibility and cognitive capacity on perceptions of an influence agent. Journal of Consumer Research, 27(1), 69-83.
Choi, S., & Rifon, N. J. (2007). Who is the celebrity in advertising? Understanding dimensions of celebrity images. The Journal of Popular Culture, 40(2), 304-324. doi:10.1111/j.1540-5931.2007.00380.x
Cowley, E., & Barron, C. (2008). When product placement goes wrong: The effects of program liking and placement prominence. Journal of Advertising, 37(1), 89-98.
Dean, D. H., & Biswas, A. (2001). Third-party organization endorsement of products: An advertising cue affecting consumer prepurchase evaluation of goods and services. Journal of Advertising, 30(4), 41-57. doi:10.1080/00913367.2001.10673650
Dekker, K., & van Reijmersdal, E. (2010). Waarschuwingen, beroemdheden en brand placement: De effecten van type waarschuwing en geloofwaardigheid op kijkerreacties. Tijdschrift voor Communicatiewetenschap, 38(4), 320. Doss, S. (2011). The transference of brand attitude: The effect on the celebrity
endorser. Journal of Management and Marketing Research, 7(1).
Erdogan, B. Z. (1999). Celebrity endorsement: A literature review. Journal of marketing
Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with persuasion attempts. Journal of consumer research, 21(1), 1-31.
Gupta, P. B., Balasubramanian, S. K. & Klassen, M. L. (2000). Viewers' evaluations of product placements in movies: Public policy issues and managerial implications. Journal of Current Issues & Research in Advertising, 22(2), 41-52.
Hayes, A.F. (2013). Introduction to mediation, moderation, and conditional process analysis:
A regression-based approach. Guilford Press.
Helm, A. E., Moulard, J. G., & Richins, M. (2015). Consumer cynicism: developing a scale to measure underlying attitudes influencing marketplace shaping and withdrawal
behaviours. International Journal of Consumer Studies, 39(5), 515-524. Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and
opportunities of Social Media. Business horizons, 53(1), 59-68.
Kelly, L., Kerr, G., & Drennan, J. (2010). Avoidance of advertising in social networking sites: The teenage perspective. Journal of interactive advertising, 10(2), 16-27.
Knowles, E. S., & Linn, J. A. (2004). The importance of resistance to persuasion. In E. S. Knowles & J. A. Linn (Eds.), Resistance and Persuasion, Mahwah, New Jersey: Lawrence Erlbaum Associates.
Lafferty, B. A., Goldsmith, R. E., & Newell, S. J. (2002). The dual credibility model: The influence of corporate and endorser credibility on attitudes and purchase intentions.
Journal of Marketing Theory and Practice, 1-12.
Notopoulos, K. (2016). Is This An Ad? Nick Jonas & Lyft Edition. BuzzFeed. Retrieved from
https://www.buzzfeed.com/katienotopoulos/is-this-an-ad-nick-jonas-lyft-edition?utm_term=.kjRBoPrw0#.ufgLljB7g
Obermiller, C., & Spangenberg, E. R. (1998). Development of a scale to measure consumer skepticism toward advertising. Journal of consumer psychology, 7(2), 159-186.
Rossiter, J.R. (2011). Marketing measurement revolution: the C-OAR-SE method and why it must replace psychometrics. European Journal of Marketing, 45(11/12), 1561-1588.
Sher, P. J., & Lee, S. H. (2009). Consumer skepticism and online reviews: An elaboration likelihood model perspective. Social Behavior and Personality: an international
journal, 37(1), 137-143.
Silvera, D. H., & Austad, B. (2004). Factors predicting the effectiveness of celebrity
endorsement advertisements. European Journal of marketing, 38(11/12), 1509-1526.
Tiwsakul, A.R., Hackley, C., & Szmigin, I. (2005). Explicit, non-integrated product placement in British television programmes. International Journal of Advertising, 24(1), 95-111.
Tutaj, K., & van Reijmersdal, E. A. (2012). Effects of online advertising format and persuasion knowledge on audience reactions. Journal of Marketing
Communications, 18(1), 5-18.
Valentini, C. (2015). Is using social media ‘good’ for the public relations profession? A critical reflection. Public Relations Review, 41(2), 170-177.
Van Reijmersdal, E. A., Lammers, N., Rozendaal, E., & Buijzen, M. (2014). Disclosing the persuasive nature of advergames: Moderation effects of mood on brand responses via persuasion knowledge. International Journal of Advertising, 34(1), 70-84.
Van Reijmersdal, E. A., Neijens, P.C., & Smit, E. (2005). Readers’ reactions to mixtures of advertising and editorial content in magazines. Journal of Current Issues & Research in Advertising, 27(2), 39-53.
Van Reijmersdal, E. A., Smit, E., & Neijens, P. (2010). How media factors affect audience responses to brand placement. International Journal of Advertising, 29(2), 279-301.
Van Reijmersdal, E. A., Tutaj, K., & Boerman, S. C. (2013). The effects of brand placement disclosures on skepticism and brand memory. Communications-The European Journal
of Communication Research, 38(2), 127-146.
Wei, M., Fischer, E., & Main, K. J. (2008). An examination of the effects of activating persuasion knowledge on consumer response to brands engaging in covert marketing. Journal of Public Policy & Marketing, 27(1), 34-44.
