WHO IS THE BEST ENDORSER ON
INSTAGRAM?
The differences between social media influencer, expert and typical consumer endorsers for the promotion of search and experience goods
Emily van Waardenburg 12507040
Master thesis - Graduate school of Communication Master’s programme: Persuasive Communication
Supervisor: Peter Neijens Wordcount: 7047
1 Abstract
Social media influencers seem to be promising brand endorsers on Instagram. Their likeability could have positive effects on the way consumers perceive advertisements and their effectiveness could be tested when compared to older endorser types such as a typical consumer or an expert. This study examined the effect of the type of endorser on Instagram users’ message and brand attitude, through some underlying processes such as source likeability, source credibility and users’ awareness of persuasive intent. Additionally, the effect of product type was explored for the relationship between the type of endorser and source credibility. Results of an online experiment (N = 180) with a 3 (type of endorser: social media influencer vs. typical consumer vs. expert) by 2 (product type: search good vs. experience good) between-subjects design showed that the typical consumer has a more positive effect on users’ message and brand attitude compared to a social media influencer and an expert endorser. This relationship was only mediated by the likeability of the source. Furthermore, product type did not seem to have a moderating effect on the relationship between the type of endorser and source credibility.
Keywords: social media influencer, endorser, Instagram, typical consumer, expert,
2 Introduction
Over the years, several studies have found evidence for the fact that the use of famous people in advertisements can be a helpful strategy in persuading consumers (McNamara, 2009). Many brands, therefore, are now aware of the fact that these celebrities are an opportunity to enhance the credibility and liking of an advertisement and eventually could impact consumers’ purchase intentions (Ford, 2018). In the past years, the online community of celebrities on platforms such as Instagram has expanded with the upswing of social media influencers that have established themselves as potential endorsers as well. Booth and Matic (2011) stated that social media influencers are able to have great influence in consumer persuasion and intentions for purchase. Another type of endorser that has been used in advertisements is a relatively unknown endorser displaying a typical consumer. These
endorsers might speak to people’s minds as a part of their reference group because they could be perceived in the same way as a family member or a friend, a group that has a great
influence on people’s purchase behavior (Ahmad & Ahmad, 2014). A third type of endorser commonly used is an expert on the promoted product. Expert endorsers are likely to be perceived as a credible source because of their expertise (Maddux & Rogers, 1980). These three types of endorsers might have different effects on consumers’ attitudes towards an advertisement and the promoted brand and therefore are interesting to compare with each other. Additionally, since these types of endorsers differ in terms of likeability and credibility, the effects on the consumers’ message and brand attitude could differ for the promotion of different kinds of products. Looking into the difference between products that must be assessed through experiencing it, also known as experience goods, or products that can be assessed in terms of clear characteristics, that are referred to as search goods,
Mudambi and Schuff (2010) found that consumers tend to perceive recommendations for experience goods as less believable because of its' subjective nature, compared to search
3 goods. Here, the credibility of a source could be an important factor in terms of the
effectiveness of an advertisement that promotes either an experience or a search good. In other words, this study will investigate to what extent the type of endorser of a brand (social media influencer vs. typical consumer vs. expert) influences the effects and users’ processing of brand-related posts on Instagram, with the type of product as a moderating factor.
This study aims to provide some insights in terms of social relevance. For advertising agencies, it is important to know what type of endorser works best for a certain kind of product. Some products could be harder to advertise compared to others, in a way that
influencers could be more successful in promoting experiences while an expert could be more successful in promoting products that are more difficult in their use. In terms of scientific relevance, this study could provide another angle in the different kinds of endorsements by comparing relatively new type of endorsers, also referred to as social media influencers, and older types of endorsers such as experts or typical consumers. Since the expectation of the effect of social media influencers is partially based on the way celebrities influence consumers, it is useful to see if these groups are comparable in their characteristics and if celebrity endorsement theories are still up to date, for the influencer endorser type as well. Additionally, Instagram as an advertising platform is relatively new to many companies nowadays and therefore, research on endorsement types on this particular platform could broaden the perspectives on the effectiveness of this form of online advertising.
Theory
Product promotion through human endorsement seems to be a promising way to advertise for companies. One platform that has recently grown big as an advertising platform with revenues greater than for example Google in the United States, is Instagram (Vizard,
4 2015). According to Djafarova and Rushworth (2017), pages on Instagram with the highest number of followers are the ones that belong to celebrities, which makes them an attractive group for the communication of marketing messages. This might be the case for social media influencers as well. Most social media influencers have become famous through their use of social media (De Veirman et al., 2016) and are therefore somewhat different types of endorsers compared to celebrities, mostly known from other events such as movies or tv shows. Even though they have become famous in a different way, the line between celebrities and social media influencers is blurry and theories on celebrity endorsement could provide insights on influencer endorsement as well. This study will focus on social media influencers as endorsers in particular, keeping their resemblance to celebrity endorsers in mind.
Social media influencers are nowadays seen and recognized by companies as opinion leaders (Uzunoglu & Kip, 2014). These companies therefore seek to use these people as an ambassador for their brand, to disseminate messages in a trustworthy way because of highly engaged relations (Lim, Radzol, Cheah, & Wong, 2017). This engagement starts with the relationship between the influencer and his/her follower and could eventually lead to a higher engaged relationship between the endorsed brand and the follower, with the influencer as the ‘bridge’ in between (Silva, Farias, Grigg, & Barbosa, 2019). The potential power of social media influencers as endorsers could be examined through the comparison with other types of endorsers, such as a typical consumer endorser or an expert on the product. Comparing influencers to a typical consumer promoting a product, an unknown endorser that could be a neighbor, a friend or a family member, might show another effect on people’s attitudes towards the product and the brand. Ahmad and Ahmad (2014) argued that friends and family are part of people’s reference groups and these reference groups influence their purchase behavior. An example of expert endorsement could be a dentist promoting a specific toothpaste brand. These experts distinguish themselves from other endorsers by their
5 expertise on the product which is seen as superior to ordinary people’s expertise (Friedman, Termini, & Washington, 1976) and therefore, this endorser type is likely to have different effects on consumers’ message and brand attitude. Conditions for this type of endorsement to be successful are real authority and being an expert on the product (Sagarin et al., 2002). Several underlying processes could affect the relationship between different kinds of endorsers and users’ message and brand attitude. This study will examine three processing variables: source likeability, source credibility and users’ awareness of the persuasive intent of the message.
An important element in individuals’ attitudes toward a promoting source is their perceived likeability (De Veirman, Cauberghe, & Hudders, 2017). Amos, Holmes, and Strutton (2008) found endorsers’ likeability to be an important factor in the effectiveness of an advertisement. This could be explained by the fact that individuals tend to be more influenced by likeable others (Cialdini, 2001). In other words, the likeability of a particular source could enhance positive feelings that could transfer to the message (Wang, 2015). De Veirman, Cauberghe and Hudders (2017) identified the number of followers of social media influencers as the most important predictor of their high likeability, which
differentiates these influencers from other endorser types. Reidenbach and Pitts (1986) noted that ‘the "typical consumer" approach is an attempt on the part of the marketer to demonstrate the similarity between the endorser and the potential user and, in so doing, to increase the endorser's credibility’ (p.31). This feeling of similarity, which is found to be a component of likeability (Burgoon, Dunbar & Segrin, 2002; Taylor, Peplau & Sears, 2003), might arise from shared demographic characteristics that the consumer has in common with the presented endorser, such as social status, age, gender or ethnicity (Comiati & Plăiaş, 2010). A girl-next-door promoting a product might feel more like a part of people’s reference groups than social media influencers, but social media influencers might be perceived as more likeable and
6 therefore more persuasive in their messages, since likeability seems to be one of the key and most influential attributes given to influencers (Moore, Yang, & Kim, 2018). Additionally, Stallen et al. (2010) stated that consumers are more often exposed to the face of a famous endorser, compared to an unknown endorser, and the earlier exposures to the famous endorser might create more positive feelings when exposed to such an endorser. Expert endorsers, however, are found to be less likeable compared to famous endorsers (Frieden, 1984) and since there is no personal relationship between the expert endorser and the consumer, they might be not as effective as social media influencers or typical consumer endorsers. This leads to the following hypotheses:
H1: Source likeability is a mediating factor in the relationship between the type of endorser and users’ message attitude and brand attitude.
H1a: Source likeability is positively related to users’ attitude toward the message and the brand.
H1b: Social media influencer endorsers are likely to be perceived as more likeable than typical consumer and expert endorsers.
H1c: Typical consumer endorsers are likely to be perceived as more likeable than expert endorsers.
Together with likeability, perceived credibility also seems to be an important factor in individuals’ attitudes toward a promoting source. Based on the source-credibility model of Hovland and Weiss (1951), source trustworthiness, which refers to validness and honesty, and source expertise, referring to a well-informed and experienced person (Yale & Gilly, 1995), are the key determinants for the perceived credibility of a source. Later, source
7 homophily was added to these determinants by Pan and Chiou (2011), which concerns the social relationships of individuals that influence the credibility of a particular source. Social homophily refers to individuals trying to look for the same demographic characteristics or perceived attributes in others that match their own (Ismagilova et al., 2019). Petty, Briñol and Tormala (2002) explained that a credible source is likely to create positive thoughts for individuals about advertisements which eventually could lead to favourable attitudes. Since there is also evidence for the positive relationship between source credibility and the adoption of information (Shuang, 2013), meaning that more information will be adopted when a source is credible, it is assumed that source credibility will positively enhance people’s attitude toward both the endorser and the information provided. Evans et al. (2017) argued that endorsement through influencers might be seen as credible because influencer-generated content reflects their genuine opinion. However, according to Tom et al. (1992), created endorsers, that could be explained as unknown people playing the role of a typical consumer, are better in creating a link with the product they are endorsing through their credibility, compared to famous endorsers, who often promote more than one product making the endorsement less genuine and credible. Nonetheless, numerous researches have studied the way product or brand endorsements enhance the trustworthiness and credibility of an advertisement, and most have come up with the argument that expert endorsing could enhance the credibility of an ad, for the main reason that the source credibility is high (Maddux & Rogers, 1980). Thus, in terms of credibility, expert endorsers might be the most successful, since source credibility for both social media influencer and typical consumer endorsers might not be based on actual expertise. This leads to H2:
H2: Source credibility is a mediating factor in the relationship between the type of endorser and users’ message attitude and brand attitude.
8 H2a: Source credibility is positively related to users’ attitude toward the message and the brand.
H2b: Social media influencer endorsers are likely to be perceived as less credible than typical consumer endorsers.
H2c: Expert endorsers are likely to be perceived as more credible, compared to social media influencer and typical consumer endorsers.
A moderating factor that might influence the relationship between the type of endorser and the consumers’ perceived source credibility is product type. Park and Lee (2009) wrote about the comparison that can be made between search goods and experience goods. They defined search goods as products from which complete information about the product itself can be obtained before purchase. In addition, Huang et al. (2009) argued that there is no interaction with the product needed to acquire product attributes, which are mostly objective. In this case, judgement is made before purchasing (Micu & Pentina, 2015). For experience goods, on the other hand, information is difficult to obtain and the attributes are subjective (Mudambi & Schuff, 2010). Therefore, Hsieh et al. (2005) stated that, besides evaluation through other people’s reviews, interaction with experience products is needed to personally evaluate the product. Mudambi and Schuff (2010) found that consumers tend to perceive recommendations for experience goods as less believable because of its' subjective nature, whereas Hsieh et al. (2005) noted that the attributes of search goods are seen as more stable and reliable, compared to experience goods. Surprisingly, Franke, Huhmann and
Mothersbaugh (2004) found that advertisements promoting search goods significantly provided more informational content compared to ads promoting experience goods. This finding contradicts the idea that information demand is higher for experience goods. Micu
9 and Pentina (2015) also argued that any brand information about experience goods will be perceived as interesting for the main reason that consumers see benefit in every information detail they can get due to the lack of their own experience of the product. In terms of this study, social media influencers or typical consumers might be better endorsers for experience goods because of their high likeability, compared to experts, that might be better endorsing search goods because of their expertise, which leads to the third hypothesis:
H3: The relationship between the type of endorser and source credibility is moderated by the type of product that is promoted by the endorser, in the sense that:
H3a: In case of experience goods, source credibility of social media influencer and typical consumer endorsers is higher than for expert endorsers.
H3b: In case of search goods, source credibility of an expert endorser is higher than for social media influencer and typical consumer endorsers.
Another process that might lie in between the effect of the endorser type on users’ attitude toward the message and the brand, is awareness of the persuasive intent of the
message. The main assumption in literature about persuasion knowledge is that the amount of knowledge individuals have on the sender’s intention to persuade determines how receptive they are to the message and how well they can withstand the persuasive intent, in a way that more knowledge will lead to less receptiveness and more resistance (Tutaj & Van
Reijmersdal, 2012). Abidin (2015) noted that, since the endorsement through influencers seems to be very personal, people are less likely to perceive the messages of influencers as biased or of clear persuasive intent. Here, the persuasive message has become a part of an influencers’ editorial content and therefore, the intention to persuade could stay hidden
10 (Dekker & van Reijmersdal, 2013). When persuasion knowledge is not triggered, attitudes are likely to stay more favorable (Tutaj & van Reijmersdal, 2012). For typical consumer endorsers, who might be perceived as part of people’s reference groups, the intention to persuade could be noticed less compared to expert endorsers. The absence of a personal relationship in expert endorsement might lead to consumers’ thinking how genuine an advertisement is and therefore, it is expected that the expert endorser can create a higher awareness of the persuasive intent in a message, leading to the following hypotheses:
H4: Users’ awareness of persuasive intent is a mediating factor in the relationship between the type of endorser and users’ message attitude and brand attitude.
H4a: Users’ awareness of persuasive intent is negatively related to users’ attitude toward the message and the brand.
H4b: Social media influencer endorsers are likely to create less awareness of persuasive intent than typical consumer endorsers.
H4c: Expert endorsers are likely to create more awareness of persuasive intent than social media influencer and typical consumer endorsers.
This study focusses on the effects of different kinds of endorsements on both message attitude and brand attitude. Foroudi (2019) explained that the attitude toward a brand is a large concept, consisting of a consumer’s assessment of associations, beliefs and reputation of the brand. Message attitude might therefore be a solid first measurement, since
MacKenzie, Lutz and Belch (1986) argued that an individuals’ attitude toward a message influences individuals’ brand attitude. Therefore, the last hypothesis states:
11 H5: A positive attitude toward the message will lead to a more positive attitude toward the promoted brand.
Figure 1. Overall conceptual model
Method
Design
Data was gathered through an online experiment using Qualtrics, with a 3 (type of endorser: social media influencer vs. typical consumer vs. expert) by 2 (product type: search good vs. experience good) between-subjects design.
Stimulus materials
In each of the six conditions, respondents were exposed to one (out of three)
endorser(s), promoting a particular product in a post, displayed in a familiar Instagram frame. The photograph contained a small caption about which product was being promoted. For this
12 study, Instagram posts were created with the use of a photoshop program. Non-existing brands were made up to make sure people were not biased about the promoted products and brand, which resulted in a teeth whitening pen from ‘New Smile’ and a watch from ‘Moon Watches’. The expert and typical consumer used in the posts were (relatively) unknown people. The social media influencer chosen for the experiment was Anna Nooshin. Last year, she was chosen as the most influential influencer of the Netherlands (Marketing Tribune, 2018). The six Instagram posts can be found in Appendix A. Besides the differences between the type of endorser and the promoted product, the posts also differed in the number of likes and the caption underneath the picture. The two posts with the social media influencer had the same number of likes, just as the two posts of the expert and the two posts of the typical consumer. The caption underneath the posts, per product, was the same for the social media influencer and the typical consumer. In the expert endorser post, a couple of words were added in the caption to clarify the expertise of the source.
Operationalization & measurements
(Parts of) existing scales (full scales are shown in Appendix B) were used to measure the processing variables and dependent variables in this study, those are:
Source likeability. The Reysen Likability Scale (Reysen, 2005) measured the
perceived likeability of a target individual. The original scale contained 11 items, from which 4 were used in this study (M = 4.44, SD = 1.19), such as ‘this person is friendly’ and ‘this
person is warm’, rated through a 7-point Likert-scale that ranged from very strongly disagree
(1) to very strongly agree (7). A principal axis factor analysis showed that the 4 items form one scale with one component with an eigenvalue above 1 (eigenvalue 2.93, KMO = .770). The reliability of the scale was good, Cronbach’s 𝛼 = .875.
Source credibility. To measure source credibility, the most used and well-established
13 important elements of source credibility; trustworthiness, expertise and goodwill (intentions). Each of these elements consists of 6 contradictions of two items, that have to be rated with a score from 1 to 7. From each element, 2 out of 6 items were extracted (M = 3.84, SD = 1.14). An example for trustworthiness is the contradiction between honesty (a score of 7) and
dishonesty (a score of 1), for expertise the contradiction between informed (7) and uninformed (1) and for goodwill, the contradiction between not centred (7) and self-centred (1). A principal axis factor analysis showed that the 6 items form 2 scales instead of
3, with two components with an eigenvalue above 1 (eigenvalue 3.40 & eigenvalue 1.17, KMO = .782). The scale’s reliability was good, Cronbach’s 𝛼 = .782.
Awareness of persuasive intent. Participants’ perception of the persuasive intent of the
message (M = 4.58, SD = 2.04) was measured by the agreement with the following
statement: ‘this post is created to persuade’. Participants had to answer this question with the use of a seven-point scale in which 1 stands for strongly agree and 7 for strongly disagree (van Noort, Atheunis, & van Reijmersdal, 2012).
Message attitude and brand attitude. 3 items about the distinction between ‘bad/good’, ‘unattractive/attractive’ and ‘negative/positive’ extracted from the study of Smink et al. (2019), were used to measure message attitude (M = 2.95, SD = .74) and brand attitude (M = 2.97, SD = .67), rated on a 5-point Likert-scale. A principal axis factor analysis showed that the 3 items from each measurement formed 1 scale, with one component with an eigenvalue above 1 for message attitude (eigenvalue 1.96, KMO = .655) and 1 for brand attitude (eigenvalue 2.23, KMO = .721). Reliability of the message attitude scale was reasonable, Cronbach’s 𝛼 = .655 and the reliability for the brand attitude scale was good, Cronbach’s 𝛼 = .822.
To make sure the presented materials were well understood, two questions were added. To test whether people recognized the specific type of endorser in their condition, they
14 were asked ‘how they perceived the endorser’ with answer options ‘as an expert on the
product’, ‘as a social media influencer’, ‘as a typical consumer’ or ‘otherwise, like (open entry)’. The following question was asked to test whether people perceived the chosen
products as an experience or a search good: ‘what type of product is a watch/a teeth whitening pen in your opinion?’ with two answer options ‘a product from which I have
knowledge, before purchase, how/if it works and how it functions’ or ‘a product from which I need to experience its functions and how it works myself (own evaluation)’.
Procedure
Respondents were shortly briefed before entering the Qualtrics survey and could not proceed without the agreement of informed consent. Starting with demographics, two
questions were asked that measured their age, and education level, with ‘none’, ‘high school’, ‘mbo’, ‘hbo’, ‘bachelor’ and ‘masters’ as options. Respondents also had to report their
Instagram use in terms of the amount of time they spend on the platform, in which
respondents could answer with ‘monthly use’, ‘weekly use’, ‘daily use’ or ‘usage multiple times a day’. After, participants were shown one of the 6 posts and were asked questions about this post in terms of source likeability, source credibility and if there was awareness of the persuasive intent of the message. Eventually, the attitude toward the message and the brand were measured. For the manipulation checks, two questions were asked concerning the way respondents perceived both the specific endorser type in their condition and the product the endorser was promoting.
Participants
For this quantitative study, 180 Dutch female Instagram users, in between the ages of 18 to 30 years old (M = 23.69, SD = 2.32), were recruited through online snowball sampling (non-probability). Most of these respondents had a ‘HBO’ degree (25.6%), followed by 24.4% with a master's degree, 20.6% with a bachelor's degree and 20.6% with a high school
15 diploma. Only 8.3% had a ‘MBO’ diploma and 0.6% had no degree at all. A large group of 66.7% of all respondents made use of Instagram multiple times a day, followed by 28.3% that uses Instagram on a daily base, 3.9% on a weekly base and 1.1% on a monthly base.
Results
Randomization
In terms of demographics, it is preferable that all groups are the same. Equally divided groups are assumed when participants are randomly assigned. In this study, a one-way
ANOVA showed that, in terms of age, the six conditions are not significantly different,
F(5,174) = 2.25, p = .051. The experimental conditions did differ with respect to education,
𝑋2(5) = 43.81, p = .011 and Instagram use, 𝑋2(5) = 26.23, p = .036.
Manipulations
In the two social media influencer conditions, 98.3% recognized the endorser as a social media influencer. Just as in the two typical consumer conditions, in which 85% of the participants recognized the endorser as a typical consumer. This points out that these two manipulations, divided over four conditions, were well understood in the experiment. In the two expert conditions, only 10% of all participants recognized the endorser as an expert on the product, meaning that this manipulation did not work.
The product manipulations were well understood: the teeth whitening pen was
recognized as an experience good by 84.4% of the participants and the watch was recognized as a search good by 83.3% of the participants.
Hypotheses
Before testing the whole model, the hypotheses were tested separately. Both source credibility, F(2,177) = .732, p = .482 and the awareness of persuasive intent, F(2,177) = .468,
16
p = .627, did not differ between the three types of endorsers. Source likeability on the other hand, was significantly different for the type of endorser, F(2,177) = 17.51, p = .000 (<.001), with the highest score on likeability for the typical consumer (M = 5.13, SD = 1.13),
compared to the social media influencer (M = 4.04, SD = 1.05) and the expert (M = 4.17, SD = 1.11). Post Hoc Bonferroni tests showed that source likeability was significantly different between the typical consumer and the social media influencer (Mdifference = 1.09, p <.001) and between the typical consumer and the expert (Mdifference = .95, p <.001). No significant
difference was found between the social media influencer and the expert in terms of likeability (Mdifference = .13, p > .05). Therefore, this study was not able to find support for H1b and H1c, since the typical consumer was the most likeable endorser compared to other endorsers. Additionally, no support was found for H2b and H2c for the main reason that source credibility did not differ significantly between the three endorser types. Just as no significant differences were found between the endorsers for users’ awareness of persuasive intent. Therefore, this study could not provide support for H4b and H4c.
A significant positive relationship was found between source likeability and both message and brand attitude; when source likeability gets higher, the attitude toward the message gets more positive, r =.29, p =.000, just as the attitude toward the brand, r = .16, p = .014, supporting H1a. Additionally, source credibility had a significant positive relationship with message and brand attitude; the more credible the source, the more positive the attitude toward the message, r =.21, p = .003 and the brand, r = .19, p = .005, and therefore supports H2a. The awareness of persuasive intent, on the other hand, did not have a significant relationship with message attitude, r = -.02, p = .412 and brand attitude, r = -.07, p = .191, meaning that H4a, although the relationship seems to be negative, could not be confirmed. Lastly, message attitude has a strong positive relationship with brand attitude, so the more positive the attitude toward the message, the more positive the attitude toward the brand, r =
17 .60, p = .000, which provides support for H5.
To test the moderation in this study, a one-way ANOVA was conducted through univariate analysis. In this test, the effect of the type of endorser and the product type on source likeability was examined. This analysis also takes the interaction effect, which has the ability to explain a moderation effect, into account. This interaction between the type of endorser and the type of product on source credibility was not significant, F(2, 174) = .139,
p = .139. In other words, no support was found for H3, H3a and H3b.
Even though some relationships between variables in the model were insignificant, the full model (see figure 2 and 3) as presented in the theoretical framework of this study will be tested by using the PROCESS v3.4 function by Hayes. This model was tested separately for the two outcome variables; message attitude and brand attitude, with the three types of endorsers as independent variable and source credibility, source likeability and awareness of persuasive intent as processing or mediating variables. In accordance with the findings described above, the only significant indirect effect occurred between the type of endorser and the attitude toward the message, through likeability, b = .0875, 95% CI
18
Figure 2. Full model with message attitude as outcome variable
The same significant indirect effect was found for the relationship between the type of endorser and the attitude toward the brand, through likeability, b = .0481, 95% CI
[.0028,.1067], supporting H1. This study could not provide support for H2 and H4.
19 Since the relationship between the type of endorser and source likeability was the only significant relationship, in which the previously discussed bivariate analysis showed that the typical consumer had a significant effect compared with the social media influencer and the expert, the final model will be limited to a smaller amount of variables (figure 4). Further on, the relationship between the typical consumer as an endorser and message attitude and brand attitude, with source likeability as a mediating factor, will be examined.
Figure 4. Final model
The use of a typical consumer in a post compared to another type of endorser is a significant predictor of the attitude toward a message, b =.3209, t = 2.66, p = .008.
Additionally, the likeability of a source is a significant predictor of the attitude toward the message as well, b = .1292, t = 2.71, p = .007. The explained variance in this Hayes model was also significant, 𝑅2 = .1204, F(2,177) = 12.12, p = .000, meaning that 12 per cent of the variance in message attitude can be explained by a typical consumer as endorser and source likeability. The indirect effect from the typical consumer on the message attitude through source likeability was significant which means the mediation effect is confirmed, b = .1319,
SE = .061, 95% CI [.0274, .2624]. In other words, both direct and indirect relationships were
found.
The typical consumer endorser, compared to other endorsers, is no significant
20 .149, just as source likeability does not seem to be a significant predictor for the attitude toward a brand, b = .0656, t = 1.45, p = .150. The explained variance in this regression was significant 𝑅2 = .0382, F(2,177) = 3.52, p = .032, which means that only 4% of the variance in brand attitude can be explained by the typical consumer as endorser and source likeability. Both final models can be found in Appendix C (figure 5 and 6).
Discussion
The results of this study show that the use of a typical consumer as a product endorser has a stronger positive effect on users’ attitude toward the message and brand compared to other endorsers. This finding contradicts the expectation of this research that social media influencers are more likely to have a positive effect on users’ message and brand attitude, compared to an expert and a typical consumer endorser. The positive relationship between the typical consumer endorser and message and brand attitude seems to be formed through
several processes. This study provides proof for one of those processes that stands in between this relationship, the likeability of a source. As expected on forehand, source likeability appeared to be positively related to message attitude and brand attitude.
This general finding can be explained through Ahmad and Ahmad’s (2014) argument that people are sensitive for others in their reference groups, which can influence their purchase behavior. A typical consumer could feel like a representative for the target
audience, mainly through perceived similarity, which is found to be a component of source likeability (Burgoon, Dunbar & Segrin, 2002; Taylor, Peplau & Sears, 2003). This feeling of similarity might originate from shared demographic characteristics with an endorser such as age, social status or ethnicity, leading to more credibility (Comiati & Plăiaş, 2010).
21 through social status might be the most important factor here. Besides, Tom et al. (1992) found in their study that created endorsers, such as unknown actors, are more effective in the creation of a link to the endorsed product, compared to celebrity endorsers. This might differ between product types, since Friedman & Friedman (1979) noted that the use of typical consumers could be useful as endorsers in the case of products that do not trigger any social or financial risk. The products used in this study, a teeth whitening pen and a watch, could be categorized under low social and financial risk in terms of consumers’ purchase
consideration. For companies, the use of a typical consumer endorser could also be the least risky in terms of the content of the advertisement. Since social media influencers write their posts on their own and manage their own account, companies are not sure of the content that will be displayed through an influencers’ channel. If they would create their own
advertisement using a typical consumer as an endorser, they can control the content and the way the advertisement spreads.
This study could not provide evidence for the expected positive influence of the social media influencer as an endorser on both message and brand attitude, which could be
explained by several reasons. As Comiati and Plăiaş (2010) stated in their research, ‘there are only a few famous persons who enjoy the appreciation from everybody’ (p.40). In this study, the choice of the social media influencer was mostly based on the large number of followers and likes on posts. However, choosing an influencer to endorse a product might depend on more than just the number of followers or likes. Besides the fact that not everyone can like or appreciate the same influencer, their image can also be damaged by bad publicity. A social media influencer might be as vulnerable for bad publicity as a celebrity, so their likeability could fluctuate in different periods. Doyle, Pentecost and Funk (2014) provided support in their studies for the fact that negative publicity for famous endorsers could potentially influence consumers’ brand attitude. Most social media influencers often collaborate with
22 more than one brand and are therefore associated with several advertising campaigns. This excessive exposure could also lead to unfavorable feelings toward the endorser and
eventually the message and the brand and might also trigger people in thinking that the endorsement is not genuine.
In short, according to this study, advertisement agencies should make use of a typical consumer as an endorser to positively influence Instagram users’ attitude toward the message and the brand because of their high likeability.
Limitations
This study has several limitations from which the most important ones will be pointed out. Starting with the main point of criticism; the operationalization of the manipulations. The executed experiment contained only one type of social media influencer endorser, one type of expert endorser and one type of typical consumer endorser. The likeability of the influencer used in this research (Anna Nooshin), was, against expectations, somewhat above average. This finding shows that the actual person, regardless of their ‘role’ as a particular type of endorser, makes a great difference. After all, the endorsers’ likeability is an important factor in the effectiveness of the advertisement (Amos, Holmes, & Strutton, 2008), and likeability is a key attribute given to influencers (Moore, Yang, & Kim, 2018). Taken into account that the chosen influencer might not be the best influencer for product endorsement, because of her moderate likeability, and another person could have created way more positive results, it is important to be aware of the difference in effectiveness of the use of these influencers for the simple reason that not all influencers are perceived in the same way. The same argument counts for the expert endorser, who was not perceived as an expert by many of the respondents. This might implicate that it is hard to let consumers believe that an expert endorser is a real expert on the product and not just an actor that is paid to pretend that he or she has intel on the product. Future research that includes social media influencers and expert
23 endorsers should thus focus on whether these types of endorsers are perceived in the right way and are, in the case of the influencer, a fit to the studies’ target group and not recently associated with bad publicity or something similar. These assumptions and overall
evaluations of endorsers could be tested with the use of a pre-test.
The second manipulation was the type of product; one experience good and one search good were chosen in this experiment. Although the manipulations of these products were well understood, the product type did not seem to affect the relationship between the endorser type and the credibility of the source. The distinction between these two types of products remains somewhat unclear, and other specific products might have provided different insights in this study. For future research, the distinction between experience and search goods should be made bigger, with extremes as examples of each product type. Experience goods could, therefore, better be operationalized through actual experiences, so less focus on a product that needs to be experienced and more focus on actual services that offer an experience.
Another limitation is the group of respondents of this study. 180 women participated in this study which means that each condition had only 30 participants. A larger number of participants per condition could help better interpret the results in terms of generalization for the overall population. Furthermore, regarding generalization, this sample appeared to be very well-educated, which is not a representative display of the population. Aside from trouble generalizing the results, DeLorme, Huh and Reid (2009) also noted that there is a positive relationship between education level and people’s level of scepticism toward advertisements. The higher the level of education, the more scepticism. Obermiller and Spangenberg (1998) defined scepticism as ‘the tendency toward disbelief of advertising claims’ (p. 160). Several researches provided evidence for the fact that higher consumer scepticism affects the perceptions of the endorsement as well as their attitude toward the
24 brand (Obermiller, Spangenberg, & MacLachlan, 2005; Bailey, 2007; Chen & Leu, 2011). More varied groups could have led to less overall scepticism which could have impacted the way participants perceived the endorsers, the message and eventually the brand. An
interesting viewpoint for future research might therefore be to find out the differences between groups, in terms of age but mostly education level, on Instagram and the effects of endorsement through Instagram on those different groups. Besides, since this study showed that Instagram already has a large number of well-educated people that are targeted as well, it might be interesting to find out how to tackle scepticism among this type of consumer and which way of advertising with what type of endorser or maybe no endorser would be the most effective.
A third limitation is the number of processing variables. Several researches describe a lot of other processes that lie in between the relationship between an endorser type and consumers’ attitude toward a message or a brand, such as source attractiveness, which is more based on actual looks than only likeability (Kahle & Romer, 1985; Till & Busler, 1998) or product fit (Kamins, 1990; Till & Busler, 1998). Since two out of three processing
variables, source credibility and Instagram users’ awareness of persuasive intent, did not seem to be significant mediators in this study and therefore are not in between the
relationship between the type of endorser and Instagram users’ message and brand attitude, it might be interesting to find out what other processes lie in between this relationship. Other relatively new and relatable variables could give more insights and other angles to go on. Besides the improvement of these limitations, another implication for future research can be made. The social media influencer endorser in this research can be described as a macro-influencer, since she has followers in the six- and seven-digit rage (Alassani & Göretz, 2019). It could be useful and interesting to test the differences between micro and macro- influencers, since a lot of famous influencers with large amounts of followers and likes could
25 almost be perceived as celebrity endorsers, which makes it a hard to find out if a social media influencer is perceived differently. It is likely that micro-influencers might have other effects than macro-influencers, since they have a smaller amount of followers, ranging from four- to five-digit number followers (Alassani & Göretz, 2019), that are more engaged with the influencer (De Veirman et al. 2017).
Conclusion
The use of social media influencers as endorsers in advertisements on Instagram is relatively new and therefore an interesting topic for research to look into. A lot of studies have provided proof for the fact that famous endorsers have positive effects on consumers’ brand attitude and purchase intention, but since the remaining blurry line between celebrities and social media influencers, it might be interesting to specify the concept of ‘a social media influencer’ and divide this endorser type into smaller categories, such as micro and macro-influencers. In order to understand the differences between types of endorsers and their effects on potential consumers, multiple endorsers of a particular kind should be compared to multiple endorsers of another kind, for several products or services. Nevertheless, in this study, the typical consumer endorser seems to be a promising tool to promote any kind of product. Through this endorsers’ high likeability, a more positive attitude toward the message and the brand was found. Given the fact that, in most other studies, the likeability of social media
influencers is high, this effect could have been found when another person was chosen in the social media influencer condition in this research. As expected, expert endorsers were not likeable in this study. Henceforward, research on endorsement should thus focus on what makes the difference between the typical consumer and the social media influencer, who could be split up into micro and macro-influencers to specify the characteristics of such endorsers. In terms of generalization, the well-educated sample of this study might not
26 provide generalizable insights on how effective endorsers are for different kinds of products. Another angle for future research could therefore be the differences between low and highly educated people, since the target-audience on Instagram is getting bigger and could get more diverse as well.
27 References
Abidin, C. (2015). Communicative intimacies: influencers and perceived interconnectedness. Retrieved from: https://adanewmedia.org/2015/11/issue8-abidin/
Ahmad, Z., & Ahmad, J. (2014). Consumer Purchase Behavior in Cellular Service Sector of Pakistan. IOSR Journal of Business and Management, 16(5), 62- 75. doi:
10.9790/487X-16546275
Alassani, R., & Göretz, J. (2019). Product placements by micro and macro influencers on Instagram. International conference on human-computer interaction, 251-267. doi: 10.1007/978-3-030-21905-5_20
Amos, C., Holmes, G., & Strutton, D. (2008). Transfer effects: exploring the relationship between celebrity and brand. International Journal of Advertising, 27(2), 209-234. 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. doi: 10.1080/13527260601058248
Basu, S. (2018). Information search in the internet markets: Experience versus search goods. Electronic Commerce Research and Applications, 30, 25-37. doi:
10.1016/j.elerap.2018.05.004
Blight, M., Ruppel, E., & Schoenbauer, K. (2017). Sense of community on Twitter and Instagram: exploring the roles of motives and parasocial relationships.
Cyberpsychology, Behavior, and Social Networking, 20(5), 314-319. doi:
10.1089/cyber.2016.0505
Booth, N., & Matic, J. (2011). Mapping and leveraging influencers in social media to shape corporate brand perceptions. Corporate Communications: An International
Journal, 16, 184–191. doi: 10.1108/13563281111156853
28
handbook: Developments in theory and practice, 445-473. doi:
10.4135/9781412976046.n23
Chen, F. P., & Leu, J. D. (2011). Product involvement in the link between skepticism toward advertising and its effects. Social Behavior and Personality: an international
journal, 39(2), 153-159. doi: 10.2224/sbp.2011.39.2.153
Cialdini, R. (2001). Harnessing the Science of Persuasion. Harvard Business Review, 72-79. Comiati, R., & Plăiaş, I. (2010). The impact of endorser’ characteristics on consumers’ reaction to the ads. Management & Marketing Journal, 8(1).
Dekker, K., & Van Reijmersdal, E. A. (2013). Disclosing celebrity endorsement in a television program to mitigate persuasion: How disclosure type and celebrity credibility interact. Journal of Promotion Management, 19(2), 224-240. doi: 10.1080/10496491.2013.769473
DeLorme, D. E., Huh, J., & Reid, L. N. (2009). Direct-to-consumer advertising skepticism and the use and perceived usefulness of prescription drug information sources. Health
Marketing Quarterly, 26(4), 293-314. Doi: 10.1080/07359680903304278
De Veirman, M., Cauberghe, V., & Hudders, L. (2017). Marketing through Instagram influencers: impact of number of followers and product divergence on brand attitude. International Journal of Advertising, 36(5), 798-828. doi:
10.1080/02650487.2017.1348035
Djafarova, E., & Rushworth, C. (2017). Exploring the credibility of online celebrities' Instagram profiles in influencing the purchase decisions of young female users.
Computers in Human Behavior, 68, 1-7. doi: 10.1016/j.chb.2016.11.009
Doyle, J. P., Pentecost, R. D., & Funk, D. C. (2014). The effect of familiarity on associated sponsor and event brand attitudes following negative celebrity endorser
29 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), 138–49. doi:
10.1080/15252019.2017.1366885
Franke, G. R., Huhmann, B. A., & Mothersbaugh, D. L. (2004). Information content and consumer readership of print ads: a comparison of search and experience products. Journal of the Academy of Marketing Science, 32(1), 20.
Ford, J. B. (2018). What do we know about celebrity endorsement in advertising?. Journal of
Advertising, 58(1), 1-2. doi:10.2501/ JAR-2018-006
Foroudi, P. (2019). Influence of brand signature, brand awareness, brand attitude, brand reputation on hotel industry’s brand performance. International journal of hospitality
management, 76, 271-285. doi: 10.1016/j.ijhm.2018.05.016
Frieden, J. B. (1984). Advertising spokesperson effects: an examination of endorser type and gender on two audiences. Journal of Advertisement Research, 24(5), 33-41.
Friedman, H. H., & Friedman, L. (1979). Endorser effectiveness by product type. Journal of
Advertising Research, 19(5), 63-71.
Friedman, H. H., Termini, S., & Washington, R. (1976). The effectiveness of advertisements utilizing four types of endorsers. Journal Of Advertising, 5(3), 22–24.
doi:10.1080/00913367.1976.10672647
Hovland, C., & Weiss, W. (1951). The Influence of Source Credibility on Communication Effectiveness. Public Opinion Quarterly, 15(4), 635–650. doi: 10.1086/266350 Hsieh, Y. C., Chiu, H. C., & Chiang, M. Y. (2005). Maintaining a committed online
customer: a study across search- experience-credence products. Journal of Retailing, 81(1), 75-82. doi: 10.1016/j.jretai.2005.01.006
30 empirical examination of consumer behavior for search and experience goods.
Journal of Marketing, 73(2), 55–69. doi: 10.1509/jmkg.73.2.55
Ismagilova, E., Slade, E., Rana, N. P., & Dwivedi, Y. K. (2019). The effect of characteristics of source credibility on consumer behaviour: A meta-analysis. Journal of Retailing
and Consumer Services. doi:10.1016/j.jretconser.2019.01.005
Kahle, L. R., & Homer, P. M. (1985). Physical attractiveness of the celebrity endorser: A social adaptation perspective. Journal of Consumer Research, 11(4), 954-961. doi: 10.1086/209029
Kamins, M.(1990). An Investigation Into the ‘Match-Up’ Hypothesis in Celebrity
Advertising: When Beauty May Be Only Skin Deep. Journal of Advertising, 19 (1), 4–13. doi: 10.1080/00913367.1990.10673175
Lim, X. J., Radzol, A. F., Cheah, J., & Wong, M. W. (2017). The impact of social media influencers on purchase intention and the mediation effect of customer attitude. Asian
Journal of Business Research, 7(2), 19–36. doi:10.14707/ajbr.170035
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), 130-143.
Maddux, J. E., & Rogers, R. W. (1980). Effects of source expertness, physical attractiveness, and supporting arguments on persuasion: A case of brains over beauty. Journal of
Personality and Social Psychology, 39(2), 235-244. doi:10.1037/0022-3514.39.2.235
Marketing Tribune. (2018). Anna Nooshin is meest invloedrijke influencer. Retrieved from https://www.marketingtribune.nl/content/nieuws/2018/12/anna-nooshin-is-meest- invloedrijke-influencer/index.xml
McCroskey, J. C., & Teven, J. J. (1999). Goodwill: A reexamination of the construct and its measurement. Communications Monographs, 66(1), 90-103. doi:
31 10.1080/03637759909376464
McNamara, K. (2009). Publicizing private lives: celebrities, image control and the reconfiguration of public space. Social and Cultural Geography, 10(1), 1-23.
Meyers, C. B. (2017). Social media influencers: A lesson plan for teaching digital advertising media literacy. Advertising and Society Quarterly, 18(2), 1–31.
doi:10.1353/asr.2017.0018
Micu, A. C., & Pentina, I. (2015). Examining search as opposed to experience goods when investigating synergies of internet news articles and banner ads. Internet
Research, 25(3), 378-398. doi: 10.1108/IntR-11-2012-0242
Mudambi, S. M., & Schuff, D. (2010). What makes a helpful online review? A study of customer reviews on Amazon.com. MIS Quarterly, 34(1), 185–200. doi:
10.2307/20721420
Moore, A., Yang, K., & Kim, H. M. (2018). Influencer Marketing: Influentials’ Authenticity, Likeability and Authority in Social Media.
Obermiller, C., & Spangenberg, E. R. (1998). Development of a scale to measure consumer skepticism toward advertising. Journal of Consumer Psychology, 7(2), 159-186. doi: 10.1207/s15327663jcp0702_03
Obermiller, C., Spangenberg, E., & MacLachlan, D. L. (2005). Ad skepticism: The consequences of disbelief. Journal of Advertising, 34(3), 7-17. doi:
10.1080/00913367.2005.10639199
Pan, L. Y., & Chiou, J. S. (2011). How much can you trust online information? Cues for perceived trustworthiness of consumer-generated online information. Journal of
Interactive Marketing, 25(2), 67-74. doi: 10.1016/j.intmar.2011.01.002
Park, C., & Lee, T. M. (2009). Information direction, website reputation and eWOM effect: A moderating role of product type. Journal of Business Research, 62(1), 61-67. doi:
32 10.1016/j.jbusres.2007.11.017
Petty, R. E., Briñol, P., & Tormala, Z. L. (2002). Thought confidence as a determinant of persuasion: The self-validation hypothesis. Journal of Personality and Social
Psychology, 82, 722-741. doi:10.1037/0022-3514.82.5.722
Reidenbach, R. E., & Pitts, R. E. (1986). Not all CEOs are created equal as advertising spokespersons: Evaluating the effective CEO spokesperson. Journal of
Advertising, 15(1), 30-46. doi: 10.1080/00913367.1986.10672986
Reysen, S. (2005). Construction of a new scale: The Reysen Likability Scale. Social Behavior
and Personality, 33(2), 201-208. doi: 10.2224/sbp.2005.33.2.201
Sagarin, B. J., Cialdini, R. B., Rice, W. E., & Serna, S. B. (2002). Dispelling the illusion of invulnerability: The motivations and mechanisms of resistance to persuasion. Journal of Personality and Social Psychology, 83, 526–541. doi: 10.1037/0022-3514.83.3.526
Shuang, Y. (2013). Effects of information quality and source credibility on EWOM adoption in context of virtual community. International Conference on Management Science
and Engineering 20th Annual Conference Proceedings, 194-200. doi:
10.1109/ICMSE.2013.6586282
Silva, M. J. D. B., Farias, S. A. D., Grigg, M. K., & Barbosa, M. D. L. D. A. (2019). Online Engagement and the Role of Digital Influencers in Product Endorsement on
Instagram. Journal of Relationship Marketing, 1-31. doi: 10.1080/15332667.2019.1664872
Smink, A. R., Frowijn, S., van Reijmersdal, E. A., van Noort, G., & Neijens, P. C. (2019). Try online before you buy: How does shopping with augmented reality affect brand responses and personal data disclosure. Electronic Commerce Research and
33 Spears, N., & Singh, S.N. (2004). Measuring attitude toward the brand and purchase
intentions. Journal of Current Issues & Research in Advertising, 26(2), 53-66. doi: 10.1080/10641734.2004.10505164
Stallen, M., Smidts, A., Rijpkema, M., Smit, G., Klucharev, V., & Fernández, G. (2010). Celebrities and shoes on the female brain: The neural correlates of product evaluation in the context of fame. Journal of Economic Psychology, 31(5), 802-811. doi:
10.1016/j.joep.2010.03.006
Taylor, S. E., Peplau, L. A., & Sears, D. O. (2003). Social Psychology (11th edition). Upper Saddle River NJ: Pearson Prentice Hall.
Till, B. D., & Busler, M. (1998). Matching products with endorsers: attractiveness versus expertise. Journal of Consumer Marketing, 15(6), 576-586. doi:
10.1108/07363769810241445
Tom, G., Clark, R., Elmer, L., Grech, E., Masetti J., & Sandhar, H. (1992). The Use of Created versus Celebrity Spokesperson in Advertisement. The Journal of Consumer
Research, 20(4), 535-547. doi: 10.1108/07363769210037088
Tutaj, K., & E.A. Van Reijmmersdal. (2012). Effects of online advertising format and persuasion knowledge on audience reactions. International Journal of Advertising,
18(1), 5–18. doi:10.1080/ 13527266.2011.620765
Uzunoglu, E., & Kip, S. M. (2014). Brand communication through digital influencers: Leveraging blogger engagement. International Journal of Information Management,
34, 592–602. doi: 10.1016/j.ijinfomgt.2014.04.007
Van Noort, G., Antheunis, M. L., & Van Reijmersdal, E. A. (2012). Social connections and the persuasiveness of viral campaigns in social network sites: Persuasive intent as the underlying mechanism. Journal of Marketing Communications, 18(1), 39-53.
34
2-7.
Wang, C. (2015). Do People Purchase What They Viewed from Youtube? The Influence of
Attitude and Perceived Credibility of User-Generated Content on Purchase Intention.
School of Communication. Tallahassee, Florida, Florida State University.
Yale, L. J., & Gilly, M. C. (1995). Dyadic perceptions in personal source information search. Journal of Business Research, 32(3), 225-237. doi: 10.1016/0148- 2963(94)00048-J
Appendix A
35 Search good: watch
1. Expert endorser 2. Social media influencer
37 Experience good: teeth whitening pen
1. Expert endorser 2. Social media influencer
38 3. Typical consumer
39 Appendix B.
Measurement scales
1. Source likeability (Reysen, 2005) 1. This person is friendly. (used) 2. This person is likeable. (used) 3. This person is warm. (used)
4. This person is approachable. (used) 5. I would ask this person for advice. 6. I would like this person as a coworker. 7. I would like this person as a roommate. 8. I would like to be friends with this person. 9. This person is physically attractive. 10. This person is similar to me. 11. This person is knowledgeable.
7 point likert-scale: Very strongly disagree – strongly disagree – disagree – neutral – agree – strongly agree – very strongly agree
2. Source credibility (McCroskey & Teven, 1999) 1. Competence
Intelligent 1 2 3 4 5 6 7 Unintelligent Untrained 1 2 3 4 5 6 7 Trained Inexpert 1 2 3 4 5 6 7 Expert (used) Informed 1 2 3 4 5 6 7 Uninformed Incompetent 1 2 3 4 5 6 7 Competent (used) Bright 1 2 3 4 5 6 7 Stupid
2. Goodwill
Cares about me 1 2 3 4 5 6 7 Doesn’t care about me
Has my interests at heart 1 2 3 4 5 6 7 Doesn’t have my interests at heart (used) Self-centered 1 2 3 4 5 6 7 Not self-centered (used)
40 Insensitive 1 2 3 4 5 6 7 Sensitive
Not understanding 1 2 3 4 5 6 7 Understanding
3. Trustworthiness
Honest 1 2 3 4 5 6 7 Dishonest (used) Untrustworthy 1 2 3 4 5 6 7 Trustworthy (used) Honorable 1 2 3 4 5 6 7 Dishonorable Moral 1 2 3 4 5 6 7 Immoral Unethical 1 2 3 4 5 6 7 Ethical Phoney 1 2 3 4 5 6 7 Genuine
3. Awareness of persuasive attempt (van Noort, Atheunis, & van Reijmersdal, 2012)
’this post is created to persuade’’
7 point Likert-scale: Strongly disagree – disagree – slightly disagree- neutral – slightly agree – agree – strongly agree
4. Message attitude (Smink et al., 2019) ‘I think the message is..’
Bad 1 2 3 4 5 Good Unattractive 1 2 3 4 5 Attractive Negative 1 2 3 4 5 Positive
5. Brand attitude (Smink et al., 2019) ‘I think the brand .. is..’
Bad 1 2 3 4 5 Good Unattractive 1 2 3 4 5 Attractive Negative 1 2 3 4 5 Positive
41 6. Manipulation check
- How do you perceive this endorser? As an expert on the product As a social media influencer As a typical consumer
Otherwise, like …. (open entry)
- What type of product is a …. in your opinion?
A product from which I have knowledge, before purchase, how/if it works and how it functions.
A product from which I need to experience its functions and how it works myself (own evaluation).
42 Appendix C
Models
Figure 5. Final model with message attitude as outcome variable
43 Appendix D. Qualtrics survey.
