Thesis
Vlog 1: Consumers’ responses to sponsorship disclosure
modality in vlogs: the role of persuasion knowledge and
brand familiarity
By: Chantal van Bergen Student number :11414537
Supervisor: Raoul Kübler
MSc Business Administration – Digital Business Date: 22 June 2018
Statement of originality
This document is written by Student Chantal van Bergen who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents
Abstract
The current study proposes a model to test whether sponsorship disclosure has an effect on brand attitudes and purchase intentions and if brand familiarity moderates the effect and persuasion knowledge mediates the effect. Moreover, the study tests whether different forms of sponsorship disclosure modality have an effect on brand attitudes and purchase intentions. Three hundred twenty women participated in an online experiment with a 3 (dual-modality disclosure, visual disclosure and auditory disclosure) by 2 (familiar vs unfamiliar brand). The experiment demonstrated that sponsorship disclosure has no effect on brand attitudes and purchase intention, independent of its effect on persuasion knowledge. However, a negative effect of sponsorship disclosure on purchase intentions for familiar brands is found, independent from its effect on persuasion knowledge. The results also show that sponsorship disclosure has a significant negative impact on brand attitudes and purchase intentions, through its effect on persuasion knowledge. Moreover, different forms of sponsorship modality do not lead to different outcomes of consumers responses. Demonstrating that current guidelines of sponsorship disclosure on YouTube are sufficient. The study contributes to literature by explaining the inconsistent results on the effect of sponsorship disclosure and show that in the context of vlogs persuasion knowledge is an important mechanism to explain the effect of sponsorship disclosures. Moreover, marketers should question to use sponsored content in vlogs as a marketing strategy, because it can have negative effects.
Keywords: Sponsorship disclosure; Persuasion knowledge; Brand familiarity; Disclosure
Table of contents
1. Introduction ... 9 2. Conceptual framework ... 12 3. Theoretical background... 13 3.1. Influencer marketing ... 13 3.2. Sponsorship disclosure ... 14 3.3. Persuasion knowledge ... 153.4 The effect of sponsorship disclosure ... 19
3.5. Moderation effect of brand familiarity ... 25
3.6. The effect of sponsorship disclosure modality ... 28
3.7. Source credibility ... 33 4. Research design ... 35 4.1. Research approach ... 35 4.2. Stimuli material ... 36 4.2.1. Pre-test ... 36 4.2.2. Pilot-test ... 37 4.3. Data collection ... 37 4.3.1. Data cleaning ... 38 4.3.2. Manipulation check ... 38 4.3.3. Final dataset ... 39 4.4. Measurement scales ... 40 5. Results ... 43
5.1. Variables and measurements ... 43
5.1.2. Skewness and Kurtosis ... 43
5.2. Hypothesis testing ... 44
5.2.1. Effect sponsorship disclosure ... 45
5.2.3. Moderation effect of brand familiarity... 49
5.2.4. Conditional mediation effect of brand familiarity ... 51
5.2.5. Effect of sponsorship disclosure dual-modality ... 53
5.2.6. Mediation effect of persuasion knowledge: Dual-modality disclosure ... 58
5.2.7. Effect of single-modality sponsorship disclosure ... 61
5.2.8. Mediation effect of persuasion knowledge: Single-modality disclosure ... 63
5.2.9. Summary results ... 65
6. Discussion ... 66
7. Implications ... 73
8. Limitation and future research ... 74
References ... 77 Appendix A ... Appendix A1 ... Appendix A2 ... Appendix B ... Appendix B1 ... Appendix B2 ... Appendix B3 ...
List of figures and tables
Figure 1. Conceptual framework p.13
Table 4.1. Conditions experimental design p.35
Table 4.2. Sponsorship disclosure type recall p.39
Table 5.1. Skewness & Kurtosis p.44
Table 5.2. Sponsorship disclosure influence on Purchase intention p.46 Table 5.3. Sponsorship disclosure influence on Brand Attitude p.47 Table 5.4. Mediation effect of persuasion knowledge on brand attitude p.48 Table 5.5. Mediation effect of persuasion knowledge on purchase intention p.48 Table 5.6. Bootstrap confidence interval mediation effect of persuasion p.48
knowledge on brand attitude
Table 5.7. Bootstrap confidence interval mediation effect of persuasion p.48 knowledge on purchase intention
Table 5.8. Moderation effect of brand familiarity on brand attitude p.50 Table 5.9. Moderation effect of brand familiarity on purchase intention p.50 Table 5.10 Moderation effect of brand familiarity on brand attitude p.50
(conditional effect)
Table 5.11. Moderation effect of brand familiarity on purchase intention p.51 (conditional effect)
Table 5.13. Conditional mediation effect on purchase intention p.52 Table 5.14. Conditional mediation effect of brand familiarity on brand attitude p.53 Table 5.15. Conditional mediation effect of brand familiarity on purchase p.53
intentions
Table 5.16. Effect of a visual disclosure on brand attitude p.55 Table 5.17. Effect of an auditory disclosure on brand attitude p.55 Table 5.18. Effect of an auditory disclosure on purchase intention p.56 Table 5.19. Effect of a visual disclosure on purchase intention p.57 able 5.20. The effect of a visual disclosure on brand attitudes, mediated by p.59
persuasion knowledge
Table 5.21. The effect of an auditory disclosure on brand attitudes, p.59 mediated by persuasion knowledge
Table 5.22. Mediation effect PK on purchase intention: Visual disclosure p.59 Table 5.23. Mediation effect PK on purchase intention: Auditory disclosure p.59 Table 5.24. Bootstrap confidence interval mediation effect PK on p.60
brand attitude: Visual disclosure
Table 5.25. Bootstrap confidence interval mediation effect PK on p.60 brand attitude: Auditory disclosure
Table 5.26. Bootstrap confidence interval mediation effect PK on p.60 purchase intention: Visual disclosure
Table 5.27. Bootstrap confidence interval mediation effect PK on p.60 purchase intention: Auditory disclosure
Table 5.28. Effect of single-modality disclosure on brand attitude p.61 Table 5.29. Effect of single-modality disclosure on purchase intention p.62 Table 5.30. Mediation effect of PK on relationship between single-modality p.63
disclosure and brand attitude
Table 5.31. Mediation effect of PK on relationship between single-modality p.64 sponsorship disclosure and purchase intention
Table 5.32. Bootstrap confidence interval mediation effect PK on brand p.64 attitude: Single-modality disclosure
Table 5.33. Bootstrap confidence interval mediation effect PK on purchase p.64 intention: Single-modality disclosure
1. Introduction
Advertisements of Kruidvat’s eyelash serum have only been in the traditional media for a few weeks now, but it has already been sold out in every Kruidvat store for months. The product was first used by social influencers in their vlogs and sales have boosted since then (Dorlo, 2018).
Kruidvat paid influencers to advertise the product. It was part of their marketing strategy. This is a typical example of where traditional media, such as television has become less important for companies. Social media and vlogs have become more essential. It is expected that the proportions will be 70/30: 70 percent of advertising will consist of social media and 30 percent of traditional media (Dorlo, 2018). However, social media will never cover 100 percent of the advertising market, because the products need to be seen in as many channels as possible, this also includes traditional media (Dorlo, 2018).
According to Dorlo (2018) vloggers are seen as friends, sisters or brothers. People trust vloggers and want to mirror their behaviour. This is different in traditional media were the advertising source is a company or brand. People don’t think that claims made are honest, because a company will be positive about their products. A vlogger is viewed as an independent person, and as such, they can carry a lot more influence. Consumers think that if they copy the influencer they will become as pretty as them. It is the new word-of-mouth (WOM) advertising. The influencer is more convincing than an advert (Dorlo, 2018).
The danger of this strategy is that vloggers do not disclose the sponsorship properly (Dorlo, 2018). People find it hard to notice the embedded advertising on social media because it is inferior to the entertainment content. Viewers may have less knowledge, judgement and attention to process the information, which can lead to positive “brand attitudes”. Therefore, sponsored content is generally invoked as ambiguous and dishonest
(Cain, 2011). To prevent consumers being misled, consumer organizations, such as the Commission of reclame code and policy makers, developed regulations to disclose the alliance between brand and vlogger (Stichting reclame code, 2016; Federal trade commision, 2013).
Legally, it is impossible to check all these vloggers according to the Reclame Code (Dorlo, 2018). However, vloggers do get caught and if sponsorship is not clearly disclosed to consumers, it is considered a violation of the regulation and the brand and vlogger may be fined. This happened to Marcha Foekistova, a famous vlogger in the Netherlands who has a beauty vlog channel on YouTube. She did not disclose her sponsored relationship with Swiss Sense (Stichting reclame code, 2016).
This frequently used marketing strategy, which blends sponsored content with social media content, is explained as the brand or products’ persuasive message, integrated within the entertainment content. It occurs on television, in magazines, games or social media (Boerman & Van Reijemersdal, 2016; Boerman, Van Reijmersdal & Neijens, 2012; Smit, Van Reijmersdal & Neijens 2009). In return for advertising the product or brand, people, such as vloggers or bloggers are compensated.
Companies who make consumers aware of their alliance with a vlogger, risk decreasing consumers’ loyalty and enthusiasm towards the brand (Boerman, Van Reijmersdal & Neijens, 2015), and may lower their purchase intentions if they understand the disclosure (Tessitore & Geuens, 2013). Boerman et al. (2015) found that disclosing sponsorship increased the recognition of the advertisement. Consequently, viewers were more critical towards the content, which led to diminished brand attitudes. Wei, Fischer and Main (2008) also found that disclosing sponsorship have diminishing effects on brand attitudes. The effect of sponsorship disclosure on brand attitude and purchase intention
only lower their brand attitudes when consumers are exposed to a sponsorship of an unfamiliar brand.
Results of studies covering the effect of sponsorship disclosure on consumer responses (i.e. brand attitude and purchase intention) remain inconsistent. According to Hwang and Jeong (2015), a honest disclosure can increase positive attitudes towards the advertisement. However, Matthes and Naderer (2016), found no evidence of sponsorship disclosure harming brand attitudes. Moreover, research in the context of vlogs remain unknown and generalizing results across different media is not possible, because vlogs have different characters. For example, ‘advergames’ differ from other media, because they require more cognitive resources to effectively play the game (Evans & Hoy, 2016). Furthermore, they show that presenting information like a disclosure in different forms (i.e. modality), influences the ability of consumers to process disclosures. Hence, the effect of a disclosure on brand attitude and purchase intention in the context of vlogs might be influenced by the modality of disclosures.
The aim of the current study is to investigate the effect of sponsorship (modality) in vlogs on brand attitude and purchase intention, and the role of persuasion knowledge and brand familiarity. The above illustrates the importance of obtaining more insights into the relationship between sponsorship disclosure and consumers’ responses in the context of vlogs, especially as people become more and more experienced with this new way of advertising. According to Friestad and Wright (1994), consumers learn to recognize persuasive attempts over the years. Therefore, they might question the honesty of vloggers. It could be that consumers will eventually see the vloggers as advertising, which takes away their credibility (Dorlo, 2018). A disclosure can increase this recognition. Thus, vlogs may not be more effective in advertising a product than advertising in traditional media with the company or brand as a source.
This leaves us with the following questions:
What is the effect of sponsorship disclosure on brand attitude and purchase intention and what is the role of persuasion knowledge and brand familiarity in this relationship, in the context of vlogs?
What is the effect of sponsorship modality on brand attitude and purchase intention and what is the role of persuasion knowledge in this relationship, in the context of vlogs?
2.Conceptual framework
The theoretical background in the next chapter will lead to the development of the following conceptual model, which is illustrated below. In this model the relationships between the variables are shown. This study considers how sponsorship disclosure modality interacts with persuasion knowledge, brand attitudes and purchase intentions. Moreover, current research will study whether sponsorship disclosure has the same effects on brand attitudes and purchase intentions for (un)familiar brands.
3. Theoretical background
In this section a detailed theoretical background for the empirical study will be provided. Important research in the area of disclosures is covered. This chapter is divided into an introduction of the concepts, influencer marketing, sponsorship disclosures and persuasion knowledge, and a detailed explanation of the effect of sponsorship disclosure (modality) on consumers response, the mediation effect of persuasion knowledge and the moderation effect of brand familiarity.
3.1. Influencer marketing
Video blogging, is known as user-generated videos and has recently become popular with millions of people, producing and sharing them (Gibson, Molyneaux, O’Donnell, Singer, 2008). Video blogs are known as ‘vlogs’, and are textual blogs enriched with nonverbal behaviour, in other words, blogs created in video (Gibson, et al., 2008; Biel & Perez, 2003). Vlogs are a form of online publishing, like blogs, are a form of online communication. Everyone with internet access can create vlogs. Vlogs are used for social commentary, alternative newscasts, creative outlets or personal online diaries (Gibson et al., 2008)
YouTube has over a billion users, almost one-third of all people on the internet use it and a billion hours of video is viewed every day by those users (YouTube, 2018). YouTube has become the main source of online videos and user community (Gibson et al., 2008). Vlogs are created on YouTube and some of these vloggers have reached 100 million followers (Lee & Watkins, 2015). Consequently, these vloggers have become celebrities and are able to reach a large community. The success of vlogs has led companies to use vlogs as a marketing tool to connect with consumers (Lee & Watkins, 2015).
Vlogs, including a celebrity’s confession about their love for L’Oréal mascara while putting the mascara on, or a friend posting a photo on Instagram with Nescafé coffee in her
hand, is known as influencer marketing. The difference between these two is, in first case, the celebrity is paid by an advertiser, and the second case, the friend is not (Woods, 2016). The fact that consumers can’t tell the difference is what makes influencer marketing so powerful (Woods, 2016). Influencer marketing is basically virtual WOM and the benefit is that consumers, tend to react more positively to messages from a friend or a persona, than from sponsored messages from companies (Woods, 2016). However, with the development of disclosure guidelines, vloggers must disclose their relationship with the company in a vlog. This may change the consumer’s response to the message and to the company.
3.2. Sponsorship disclosure
A disclosure is a declaration that communicates to consumers that the content they are reading, or viewing is intended to persuade them (Federal trade commission, 2013). The objective of a disclosure is to give consumers information about the sponsored content, this increases the knowledge of the advertising and persuasive intent (Boerman et al. 2012). To achieve its objective, a disclosure must be clearly communicated to consumers (Boerman, et al. 2015).
To ensure that a sponsorship is disclosed, each country and media channel have different sponsorship disclosure guidelines. The guidelines for television, radio and magazine have been established for years. According to the Federal Trade Commission in the USA, ''The disclosure should be presented simultaneously in both the audio and video portions of the television advertisement'' (The Office of the Federal Register, 1981, p. 285). In Europe, the Audiovisual Service Director states that:
''viewers shall be clearly informed of the existence of a sponsorship agreement.
service(s) or a distinctive sign thereof in an appropriate way for programmes at the beginning, during and/or at the end of the programmes'' (The Audiovisual commercial commutations, 2016, article 10).
With the development of social media, new guidelines have been defined. Likewise, these guidelines also explain how to disclose alliance between the brand and the influencer. The execution can be done by using hashtags (#spon, #adv) in microblogs, such as Instagram and webblogs, or it can be a spoken text (e.g. I have been paid by brand X; I received product Y from brand X) in vlogs (Stichting reclame code, 2014). In the Netherlands, initiatives have resulted in the development of special guidelines for YouTube, ‘Social Code; YouTube.’ This community also developed guidelines especially for vloggers.
‘Social Code; YouTube,’ distinguishes between paid sponsorship, free product/discount sponsorship, and paid sponsorship for charity. Both paid sponsorship and paid sponsorship for charity have the same guidelines. Before the video begins, the vlogger must disclose sponsorship for at least 3 seconds, or disclose advertising in a spoken text within the vlog. Besides, the vlogger should include a disclosure in the description of the vlog. If the vlogger is sponsored by receiving a discount or free products, the vlogger should disclose this within the video description (Social Code: YouTube, 2017).
Sponsorship disclosures have been found to influence brand evaluation of viewers (Boerman, et al. 2015). According to Campbell, Mohr and Verlegh (2013), post-disclosure has a negative effect on brand attitude, because it results in unwanted inferences. However, findings of studies examining how sponsorship disclosure affects brand attitudes differ.
3.3. Persuasion knowledge
Negative effects on brand attitudes can occur by activation of persuasion knowledge, which is a result of sponsorship disclosure (Van Reijmersdal, Tutaj & Boerman, 2013). According to
Friesland and Wright (1994), persuasion knowledge is a resource of consumers that enables them to recognize the persuasive attempt of a message. Persuasion knowledge includes ideas about marketer’s goal (i.e. persuasion motives) and tactics that are used to achieve the goals (Campbell & Kirmani, 2000). The idea of the persuasion knowledge model (PKM) is that consumers can use their persuasion knowledge to recognize the persuasive attempt of the marketers and to cope with these influences (Campbell & Kirmani, 2000). As mentioned, over the years consumers learn to recognize persuasive tactics (Friestad & Wright, 1994)
Persuasion knowledge consists of conceptual persuasion knowledge and attitudinal persuasion knowledge (Boerman et al. 2012; Rozendaal, Lapierre, Van Reijmersdal & Buijzen, 2011) According to Rozendaal et al. (2011), conceptual persuasion knowledge consists of, recognition of the advertising and the source, perception of the intended audience, understanding advertising’s selling intent, persuasive intent and persuasive tactics, and advertising’s bias. Current study defines conceptual persuasion knowledge, as the viewer’s ability to recognize the advertising (Boerman, Willemsen & AA, 2017) and persuasive and selling intent.
On the other hand, the attitudinal persuasion knowledge includes attitudinal mechanism, which enables consumers to effectively cope with advertising tactics (Boerman et al. 2012). The attitudinal persuasion knowledge is established through, scepticism, disliking towards advertising and distrust, applied to a specific persuasive attempt (Rozendaal, et al. 2011). It involves critical attitudes, such as feelings of honesty, trustworthiness, and credibility (Boerman et al. 2012). According to Boerman et al. (2017), consumers must first activate their conceptual persuasion knowledge before they can use their attitudinal persuasion knowledge. Critical attitudes towards the message are normally developed on
recognition of the advertising content (Campbell & Kirmani, 2000). The current study defines attitudinal persuasion knowledge as critical feelings and scepticism towards the advertising.
In some situations, persuasion knowledge may not be activated automatically. Tutaj and Van Reijmersdal (2012) argue that the advertising format influences the recognition of the advertising, such that more pronounced advertising formats are easier to be recognized by consumers, as being of persuasive nature. They found that advertising banners are more recognizable as advertising than sponsored product placement. Moreover, they found that the higher the recognition and understanding of the advertising, the more people will understand the persuasive intent of advertising. However, new advertising setups and sponsored content embedded in entertainment content are situations in which the advertising is not directly inferred as advertising by consumers (Boerma et al., 2017). The above situations apply to sponsored content in vlogs.
Including a sponsorship disclosure will inform the viewers of the persuasive attempt and consequently activate conceptual persuasion knowledge (Boerman et al. 2017), because the main objective of a disclosure is to make the viewers able to distinguish between commercial and entertainment content (Boerman et al. 2012). Current study defines persuasion knowledge as a combination of two concepts, the conceptual and attitudinal persuasion knowledge. A disclosure will activate conceptual persuasion knowledge, which consequently activates attitudinal persuasion knowledge.
The first study which investigated the effect of sponsorship disclosure on persuasion knowledge is Boerman et al (2012). Results of the study support the reasoning above; recognition of advertising is the first step of persuasion knowledge, which leads to activation of the conceptual persuasion knowledge. Consumer’s conceptual persuasion knowledge was activated after exposure to a sponsorship disclosure in a television program. Consequently, viewers were more able to distinguish commercial from editorial content compared to viewers
who were not exposed to a disclosure. The activation of conceptual persuasion knowledge resulted in higher rates of attitudinal persuasion knowledge (Boerman et al. 2012).
Other studies, which focused on sponsorship disclosure and the use of persuasion knowledge in blogs (Campbell et al., 2013; Van Reijmersdal, et al. 2017), Facebook (Boerman et al. 2017), video clip (Matthes & Naderer, 2015), traditional media, such as radio and television shows (Boerman, Van Reijmersdal & Neijens, 2014; Boerman, et al., 2015; Campbell et al., 2013), and ‘advergames’ (Evans & Hoy, 2016), found similar results.
Summarizing the results of these studies, consumers who were exposed to a sponsorship disclosure, reported higher levels of persuasion knowledge, compared to consumers who were not exposed to a disclosure. Thus, consumers recognized the advertising, selling and persuasion intent, had critical feelings and were sceptical towards the advertising.
In addition, Boerman et al., (2017) found that a Facebook post without a ‘sponsored label’ sent by a celebrity, lowered the activation of conceptual persuasion knowledge. The results demonstrate that people do not recognize advertisement and persuasion and selling intent of a post that is sent by a celebrity. People do not notice the advertising, because celebrities may have different reasons for posting something on Facebook. However, if the post was sent by a brand, a disclosure was not needed, persuasion knowledge was already activated and did not increase by exposure to a disclosure.
As mentioned above, the basic assumption of persuasion knowledge model is that consumers are able to identify the persuasive goal of the message. A result of the identification of the advertising is more scepticism and critical feelings of consumers towards the sponsored content and depreciation of the persuasive attempts. Therefore, this study
conceptual persuasion knowledge and consequently results in higher rates of critical attitudes towards the advertisement (i.e. attitudinal persuasion knowledge). In contrast, viewers who are not exposed to a sponsorship disclosure will not recognize the advertising and therefore they will show lower levels of persuasion knowledge, consequently they have less critical attitudes towards the advertisement.
3.4 The effect of sponsorship disclosure
Sponsorship disclosure may serve as a cue for consumers to change their brand attitudes and purchase intentions. The Reactance Theory postulates that consumers want freedom of choice, and a consequence of feeling threatened by no freedom of choice is reactance (Miron & Brehm, 2006). If people cannot get their freedom back, the alternative imposed seems subjectively less attractive than the denied alternative. According to Quick, Shen and Dillard to obtain certain emotions, attitudes and actions, is known as having freedom. In other words, freedom is to feel, as well as to have a specific evaluation or not to. People have freedom only when they know that they have it and know that they are able to accomplish their chosen behaviour (Miron & Brehm, 2006). Anything that is making it more difficult to obtain freedom is known as a threat. Reactance tends to follow when people are threatened by their freedom or freedom is eradicated (Quick, Shen & Dillard, 2013).
According to Wood and Quinn, who investigated the effect of forewarnings on attitudes, “Warnings are likely to generate a defensive orientation when they threaten some aspect of people's identity, their freedom, or their existing attitudes’’ (Wood & Quinn, 2003, p. 2). If a message is intending to change consumers views, their freedom is threatened and
reactance is likely to follow after a warning. Attitudes can be seen as a response to the threat set by a warning (Wood & Quinn, 2003).
Moreover, research into the effect of a threat to attitudinal freedom found that people negatively change their attitudes when people are told to hold a certain attitude about certain issues. They changed their attitudes in the opposite direction. The unfavourable attitudes arose because people felt threatened by no freedom to have a particular attitude towards a nuclear power plant near populated areas (Heller, Pallak & Picek, 1973).
Research on disclosure argues that a disclosure which uncovers the embedded sponsored content can lead to feelings of having been misled. These feelings arise because people in general do not want to be influenced (Boerman et al. 2012). They can get the feeling that their freedom to have particular attitude or feeling, is being threatened. Consequently, consumers develop more scepticism towards the advertising and these critical attitudes can lead consumers to oppose the persuasive message (Wei, Fischer & Main 2008).
According to, Wood and Quinn (2003) the impact of the warnings (i.e. disclosures) are based on the threat to existing attitudes, to one’s self-image of gullibility and to motivation to think about the issue. However, according to Campbell et al. (2013), consumers must react to a disclosure by activating their persuasion knowledge, identifying the hidden sponsored content as advertising, inferring that the advertising is undesirable and correcting for the influence, in order for a disclosure to result in adjustment of consumer’s responses. In other words, a disclosure must activate persuasion knowledge of viewers in order to conclude that the sponsored product placement is generating unwanted influence (Campbell et al., 2013). According to PKM, a consumer will only change their responses if they feel that the persuasive message has excessively influenced them (Friestad & Wright, 1998). Hence, only then people might change their brand attitudes and purchase intentions.
Consumers who access their persuasion knowledge to judge the advertisement on persuasive intent, can either resist the persuasive message or choose to be persuaded by the
message (Friestad & Wright, 19980. Therefore, people with activated persuasion knowledge are expected to act differently with persuasive attempts than people without activated persuasion knowledge (Boerman, et al. 2012), because people with activated persuasion knowledge tend to recognize persuasive attempts better.
As previously stated, attention to the undesirable advertising is likely to be triggered by a disclosure. By recognizing the undesirable advertising, consumers are motivated to change their responses and correct for the influence (Campbell et al., 2013). Consumers can change their attitudes towards the brand as a correction for the influence, which is the sponsored content that is trying to convince the consumers of the brand message (Boerman et al. 2012).
Attitudes towards the brand can be defined, as association between a branded product and the memory of an individual of the evaluation of the branded product (Farquhar, 1989). Research found that indeed disclosing a sponsorship leads to negative attitudes developing towards the brand, but only via activation of persuasion knowledge (Boerman, et al. 2012; Wei et al., 2008; Boerman, et al. 2014). Boerman et al. (2014), showed that recognition of the advertising, by a disclosure in a MTV program, resulted in more critical attitude towards the content (attitudinal persuasion knowledge), consequently people lowered their attitudes towards the brand.
However, contradicting results were found by Matthess and Naderer (2016), who found no effect of activated persuasion knowledge in both a music video and television show on brand attitudes. Furthermore, results of the study of Dekker and Van Reijmersdal (2013), who studied the effect of celebrity endorsement, found no effect of activated persuasion knowledge on brand attitudes. Nevertheless, they found that the persuasiveness of the message (i.e. acceptance of the ad claims) increases with higher levels of celebrity credibility.
Another study found positive effects of a sponsorship disclosure. Explicit honest sponsorship disclosure about a blogpost lead to positive attitudes towards the message and the source, compared to normal sponsorship disclosure, which led to negative attitude towards the message and the source (Hwang & Jeong, 2016).
On the whole, the majority of research focusing on sponsorship disclosure all found negative effects of sponsorship disclosure on consumers’ responses, but only as a result of activated persuasion knowledge. Research found negative effects, of engaging in EWOM (Boerman et al., 2017), on brand attitude (Boerman et al. 2012, 2014, 2015; Campbell et al., 2013; Van Reijmersdal, et al. 2016; Wei et al., 2008) and on purchase intentions (Tessitore & Geuens, 2015; Van Reijmersdal, et al. 2016). These finding contradicts with the Reactance Theory that postulates that the effect can also be negative, independent of activation of persuasion knowledge, because a sponsorship disclosure may be inferred as a threat to existing attitude and behaviour (Miron & Brehm, 2006; Quick, Shen & Dillard, 2013).
However, the negative effects were found to be moderated by sponsorship type (Boerman et al., 2015; Evans & Hoy 2016), brand familiarity, high versus low (Wei et al., 2008), timing of the disclosure: prior, concurrent or after the product placement (Boerman et al., 2014; Cambpell, et al., 2013), duration of the disclosure (Boerman, et al. 2012), product type, search goods versus experience goods (Lu, Chang and Chang, 2014), cognitive loading of consumers (Evans & Hoy, 2016), gamer’s mood (Van Reijmersdal, Lammers, Rozendaal & Buijzen, 2015) and credibility of the source, brand versus celebrity (Boerman et al. 2017).
As already explained, according to Cambpell et al. (2013), only a sponsorship disclosure after the product placement in a television show leads to diminished brand attitudes, whereas a prior disclosure did not lead to diminished brand attitudes. Boerman et al.
a 3 seconds disclosure. Moreover, Lu et al. (2014), who research the effects of two types of disclosure, namely the revealing of paid sponsorship or discount sponsorship, did not affect attitudes towards the blog. These variances in results demonstrate that the effect of disclosure is still an unresolved issue. In some situations persuasion knowledge is lower, and consequently less powerful in worsening brand attitudes, while in other situations persuasion knowledge does not lead to any diminished brand attitudes and purchase intentions.
Furthermore, attitudes are a determinant of product purchases (Farquhar, 1989). The Theory of Reasoned Action (TRA) (Fishbein & Ajzen, 2011) and Theory of Planned Behavior (TPB) (cited in, Wei, et al. 2008, as Ajzen, 1985), argues that attitudes are a determinant of behaviour. According to Fishbein & Ajzen (2011), “Once attitudes,... have been formed they are directly accessible and available to guide intentions and behaviour’’ (Fishbein & Ajzen, 2011, p. 21). In addition, Laroche, Kim and Zhou (1996), found that brand attitude is a determinant for purchase intention.
In the context of blogs, Lu et al. (2014) found that when consumers have high positive attitudes towards the sponsored recommendation post, it improves the purchase intentions. Hence, purchase intentions are likely to chance in the same direction as brand attitudes, when a disclosure is exposed to consumers. In support of this, Tessitore and Geuens, (2013) and Van Reijmersdal et al. (2016) found that when consumers understood the disclosure, they showed lower levels of purchase intention, through its effect on persuasion knowledge, in both a television show and a blog. These findings are in line with the effect of a disclosure on brand attitudes.
In sum, research into vlogs remains unknown. The current study will replicate testing the relationship between persuasion knowledge, brand attitude and purchase intentions, which is studied in the context of other media, but not in the context of vlogs. In general, activated
persuasion knowledge leads to more critical attitudes towards the sponsored content. On the basis of the PKM and prior research, which overall demonstrates that sponsorship disclosure negatively influences the attitudes towards the brand and purchase intention, the current study expects sponsorship disclosure to activate persuasion knowledge, and consequently negatively influence brand attitude and purchase intentions. Persuasion knowledge seems to be an underlying mechanism to explain the relation between sponsorship disclosure and consumers’ responses. However, in the context of vlogs the effect of sponsorship disclosure has never been investigated. Hence, this study is also interested in the direct effect of sponsorship disclosure on brand attitude and purchase intention. Using the Reactance Theory and research on warnings, this study expects to find a negative effect of sponsorship disclosure on brand attitude and purchase intention.
H1a: Viewers who are exposed to a sponsorship disclosure show higher levels of brand
attitudes compared to viewers who are not exposed to a sponsorship disclosure.
H1b: Viewers who are exposed to a sponsorship disclosure show higher levels of purchase
intentions compared to viewers who are not exposed to a sponsorship disclosure.
H2a: Viewers who are exposed to a sponsorship disclosure show higher levels of persuasion
knowledge and consequently lower levels of brand attitudes, compared to viewers who are not exposed to a sponsorship disclosure.
H2b: Viewers who are exposed to a sponsorship disclosure show higher levels of persuasion
knowledge and consequently lower levels of purchase intentions, compared to viewers who are not exposed to a sponsorship disclosure.
3.5. Moderation effect of brand familiarity
The effect of sponsorship disclosure tend to differ by different levels of variables, known as moderators, for example brand familiarity. Brands differ in the amount of brand knowledge consumers have stored in their memory (Campbell & Keller, 2013). Brand familiarity is defined as a consumer’s product- related experience (Alba & Hutchinson, 1987). Consumers’ exposure to advertising, product purchases, communication with sellers, choices and decision making, information search and product usage in numerous situations, all illustrate product-related experiences by consumers. Consumers’ experiences with a product result in consumers’ expertise about the product (Alba & Hutchinson, 1987).
Awareness is a precondition for brand associations (hierarchy of effects model) (Koll & Wallpach, 2014). According to Keller, brand awareness, includes brand recognition, the ability to recall brand exposures and brand recall, the ability to name a brand by given product hints (Keller, 1993). Over time consumers develop associations with a brand, resulting from prior experiences with the brand (Alba & Hutchinson, 1987). Consumers do not have any experience with unfamiliar brands and therefore lack associations with them (Campbell & Keller, 2013)
In addition, associations result in evaluations of behavioural responses, for example trust and purchase intentions (Koll & Wallpach, 2014). Consumers tend to have more confidence towards a brand that they have experience with (Laroche, et al. 1996). Therefore, consumers are likely to respond to advertising from a familiar brand, such as by making purchases (Koll & Wallpach, 2014).
Not only association, but consumers lack also knowledge about unfamiliar brands, to ground their attitudes on. Consequently, they rely on attitudes towards the advertisement to form brand evaluations (Campbell & Keller, 2003). In contrast, consumers have knowledge about familiar brands, and therefore they use their prior brand evaluations to form attitudes
about the advertisement. In other word, consumers are more attitude certain about familiar brands than unfamiliar brands (Matthes & Nadarer, 2016). Thus, the influence of advertising attitude on brand attitude is greater for unfamiliar brands (Campbell & Keller, 2003). According to Keller (1993), higher brand awareness can increase the effectiveness of marketing communications. The message persuasion for familiar brands would be greater, than for unfamiliar brands.
Moreover, the processing of familiar brands is different than for unfamiliar brands. Consumers focus on learning about unfamiliar brands, while in contrast, they try to update their knowledge about familiar brands (Campbell & Keller, 2003). According to Campbell and Keller (1993), adverts for unfamiliar brands require more extensive processing, because they seem more interesting and appealing. A result of extensive processing is that consumers are more likely to be more judgemental about the (un)desirability of advertising of unfamiliar brands. As a result, consumers are likely to show higher rates of activated persuasion knowledge, and consequently the advertising message is less persuasive for unfamiliar brands, than for familiar brands.
Hence, brand familiarity can moderate the effect of sponsorship disclosure and consequently attitudes towards the brand and purchase intention of the product, recommended in the advertisement. Results of a study by, Wei et al. (2008) show that persuasion knowledge was less affective in diminishing brand attitudes for a familiar brand versus unfamiliar brand in a radio show, because people were more attitude certain about the familiar brand. Thus, disclosure had no effect for familiar brand in lowering brand attitudes in contrast to unfamiliar brands. In the context of recommended blog articles, Lu et al. (2014) found that consumers with high brand awareness (i.e. brand recognition and recall), believed the claims in a blog more, which resulted in a positive attitude towards the advertising.
Overall, one study investigated brand familiarity as moderating variable in the relation between activated persuasion knowledge and brand attitude (Wei et al. 2008). The current research expects that persuasion knowledge is less capable in worsening brand attitudes for familiar brands, because consumers are more attitude certain and rely more on existing knowledge to evaluate familiar brands. Hence, brand attitudes and purchase intentions of consumers are expected to not change by exposure to a sponsorship disclosure of a familiar brand. In contrast, consumers process information more extensively to form attitudes towards the unfamiliar brand and therefore it is expected that higher levels of persuasion knowledge will result in lower brand attitudes and purchase intentions. Thus, brand attitudes and purchase intentions of consumers, are expected to be lower by exposure to a sponsorship disclosure of a familiar brand against an unfamiliar brand. The following is therefore hypothesized:
H3a: Viewers who are exposed to a sponsorship disclosure of a familiar brand show higher
levels of brand attitudes, compared to viewers who are exposed to a sponsorship disclosure of an unfamiliar brand.
H3b: Viewers who are exposed to a sponsorship disclosure of a familiar brand have higher
levels of purchase intentions, compared to viewers who are exposed to a sponsorship disclosure of an unfamiliar brand.
H4a: Viewers who are exposed to a sponsorship disclosure of a familiar brand show lower
levels of persuasion knowledge and consequently higher levels of brand attitudes, compared to viewers who are exposed to a sponsorship disclosure of an unfamiliar brand.
H4b: Viewers who are exposed to a sponsorship disclosure of a familiar brand show lower
levels of persuasion knowledge and consequently higher levels of purchase intentions, compared to viewers who are exposed to a sponsorship disclosure of an unfamiliar brand.
3.6. The effect of sponsorship disclosure modality
Not only brand familiarity can change the effect of sponsorship disclosure, which can potentially lead to deception of consumers (Evans & Hoy, 2016), but also how the disclosure it presented to viewers can influence the effect of sponsorship disclosure. According to Evans and Hoy (2016) a viewer must fully observe the disclosure, in order for the disclosure to fully inform the viewers that the information presented in the content is of persuasive attempt. In other words, the viewer must be able to recall the disclosure and understand that a disclosure serves as a cue of persuasive attempts (Boerman et al. 2012). In support of this, research found that only using a symbol as a disclosure didn’t lead to more resistance of the persuasive message, because participants could hardly recall the symbol and didn’t understand the meaning of the symbol (Tessitore & Geuens, 2013).
Sponsorship disclosures can be presented in different formats and modalities. Modality is known as the way in which information is presented to people and how it is encoded (Pavio, 1991). Information can be encoded either visually (printed words vs. visual objects), auditory (spoken word vs. environmental sounds) or haptic (tactual and motor feedback from writing vs. manipulation of objects) (Pavio, 1991). Information can either be presented in a dual-mode (i.e. dual-modality), or as a single-mode (i.e. single-modality).
According to Penny (1989), short-term memory is improved by presenting information in dual-mode, rather than a single-mode. People use different mechanisms to process information, therefore recall is better when two presentation modalities are used. People can make better connections between visual and auditory information, which is stored in working memory. This also means that when people do two concurrent tasks, performance is better with information presented in two modalities.
The Dual-Code Theory of Pavio (1991), has the same assumption of Penny (1989), this theory highlights that processing of verbal and nonverbal systems is different, and if information is presented in two different modalities people can recall the information more easily. However, this study emphasises the importance of verbal and nonverbal symbolic contrast. An example, of a non-verbal symbolic is a logo.
Findings of various research supports this view. Frick (1984) who studied the short-term memory of digits, found that a sequentially presentation of auditory and visual digits, resulted in more memory than showing the digits in one mode. Research on risk disclosures found that knowledge and awareness were higher for dual-modality presentation than for single-modality presentation (Morris, Mazis & Brinberg ,1989). The study investigated risk disclosures formats in a drug commercial on television. Single-modality presentations resulted in less profound processing than dual-modality disclosures. The dominance of dual-modality presentation was also found by Murray, Manrai and Manrai (1998), who studied viewers ability to process information in a video super (i.e. the visual presentation of verbal presentations in a commercial). They found that dual-modality, a video including audio voice over, is most effective to guarantee awareness.
One recent study researching the effect of sponsorship disclosure and brand placement in a television program on visual attention, found that a combination of logo (nonverbal) and text (verbal) disclosure, was most effective in attracting visual attention and persuasion knowledge (Boerman, et al. 2015). They used eye tracking techniques to study the effect. They found no direct effect of different modalities of disclosures, but only an indirect effect via viewers’ attention to the advertisement and persuasion knowledge. Hence, the influence of a disclosure on persuasion knowledge depends on the viewers’ attention to a disclosure and activation of persuasion knowledge.
Summarizing the above reasoning, dual-modality disclosures are processed differently from single-modality disclosure, consequently resulting in higher levels of promoting, awareness (Paivio 1991; Morris et al., 1989), understanding (Morris, et al. 1989; Murray, et al. 1998; Tessitore & Geuens, 2013), and recall (Tessitore & Geuens, 2013).
However, another recent study found contradicting results. Evans and Hoy (2016), who studied the effect of sponsorship modality on persuasion knowledge in ‘advergames’, found that single-modality disclosure was more effective in activating persuasion knowledge, than dual-modality. They argued that ‘advergames’ differ in characteristics to other type of media, such that participants were unable to strike the advertising, because of distraction by gameplay and sound effects.
In sum, no study focused on the effects of sponsorship disclosure modality in the context of vlogs. The results of recent studies are inconsistent and generalizing the findings to the context of vlogs is not possible, because vlogs are a different form of media with different characteristics. Nevertheless, based on the Dual-Code Theory and Penny’s (1989) assumptions, which is demonstrated by various research to be true (e.g. Frick, 1984), the current research expects superiority of dual-modality disclosures, over single-modality disclosure in effective activating persuasion knowledge. Consequently dual-modality disclosure is more effective in decreasing brand attitudes and purchase intention, than single-modality disclosure. The following is therefore hypothesized:
H5a: Viewers who are exposed to a dual-modality sponsorship disclosure show higher levels
of brand attitudes, compared to viewers who are exposed to a single-modality sponsorship disclosure.
H5b: Viewers who are exposed to a dual-modality sponsorship disclosure show higher levels
of purchase intentions, compared to consumers who are exposed a single-modality sponsorship disclosure.
H6a: Viewers who are exposed to a dual-modality sponsorship disclosure show higher levels
of persuasion knowledge and consequently show lower levels of brand attitude, compared to viewers who are exposed to a single-modality sponsorship disclosure.
H6b: Viewers who are exposed to a dual-modality sponsorship disclosure show higher levels
of persuasion knowledge and consequently show lower of purchase intentions, compared to viewers who are exposed to a single-modality sponsorship disclosure.
However, according to the guidelines of ‘Social Code; YouTube’ (2017), a vlogger does not have to present the disclosure in dual-code. They can choose to either use a visual or auditory sponsorship disclosure in the vlog. Nevertheless, do both forms of disclosure have the same effect on brand attitude and purchase intention?
According to penny (1989), traces in short-term memory of presented information, either verbally or auditory, are different. The code people produce when they silently articulate the visually presented verbal information is called the P (phonological) code. The code produced by people of information that is auditory presented is called A (acoustic) code. The A code is only produced by people when hearing an auditory stimulus and the P code is produced by people both for an auditory and visual stimuli. The A code gives a boost to the short-term memory, and therefore it is easier for people to recall auditory presented information, over visually presented information. This effect is also known as the modality effect. Visually and verbally presented information is processed separately in short-term memory. People cannot resist the automatic storage of auditory information in memory. The P
code of visually presented verbal information is not automatically stored in memory, because it can be disrupted. Therefore, auditory information is superior to visually presented information.
These accounts are grounded by the model of working memory of Baddeley (2003), which postulates that auditory information is directly passed intro the P store, then it can be recalled or the information is ready for rehearsal. Visual information needs to be recoded verbally and then it will be stored in the P loop for rehearsal.
In support of this, Russel (2002) who investigated the effectiveness of product placement in a television show. Showed that audio-only disclosure of product placement is superior to visual-only disclosure of product placement in brand memory. The auditory channel includes the script of a TV show, therefore information presented auditory is more deep-rooted and important than visually presented information. Findings of Tessitore and Geuens (2013), show that recall of a disclosure is a determinant for the effectiveness of product placement disclosure on resistance of the placement. However, the findings of Russel (2002), show that only when a visual product placement disclosure was not in line with the plot of a television program brand attitudes of viewers decreased, because product placements seemed unnatural. Thus, remembering a disclosure does not always lead to attitude change.
Additionally, Boerman et al. (2015) compared the effectiveness of a logo (visual modality) and text (visual modality). Results of this study show that text-only disclosure was more effective than logo-only disclosure in increasing attention and consequently increasing persuasion knowledge of the viewers of a television program. The difference in effectiveness between the two disclosures was viewers visual attention. Additionally, Evans and Hoy (2016), found that in ‘advergames’ a text-only disclosure was more effective in activating
persuasion knowledge than no disclosure. No study has compared the effect of visual disclosures and auditory disclosures to date.
Hence, based on short-term memory and modality effect, the current study expects that auditory sponsorship disclosure leads to more activated persuasion knowledge than visual sponsorship disclosure. Memory and awareness of a visual disclosure depends on viewers’ visual attention, whereas auditory disclosure does not. Besides, auditory disclosure cannot be ignored and so produces better recall of the disclosure, consequently this leads to higher levels of activated persuasion knowledge, resulting in more negative brand attitudes and purchase intentions in contrast to a visual disclosure. The following is hypothesized:
H7a: Viewers who watch a visual sponsorship disclosure show higher levels of brand
attitudes, compared to viewers who hear a auditory sponsorship disclosure.
H7b: Viewers who watch visual sponsorship disclosure show higher levels of purchase
intentions, compared to viewers who hear a auditory sponsorship disclosure.
H8a: Viewers who watch a visual sponsorship disclosure show lower levels of persuasion
knowledge and consequently show higher levels of brand attitudes, compared to viewers who
hear an auditory sponsorship disclosure.
H8b: Viewers who watch a visual sponsorship disclosure show lower levels of persuasion
knowledge and consequently show higher levels of purchase intentions, compared to viewers who hear an auditory sponsorship disclosure.
3.7. Source credibility
The theoretical background of the two main aims of the study have been developed. However, this study explains one important variable that is not the focus of the study, but is nevertheless considered to be of particular important for current study.
The effectiveness of sponsorship disclosure is impacted by source credibility of a vlogger. Source credibility is used to indicate a communicator’s positive characteristics that influences the acceptance of the message of the receiver (Ohanian, 1990). In this study, source credibility is defined as, the degree to which the vlogger is perceived to be trustworthy and has expertise. In particular, expertise or competences refer to the perceived validity of the declarations made by the vlogger, and trustworthiness is the extent to which the viewers perceive the declarations made, to be those the vlogger grants most valid (Hovland, Janis, Kelley, 1953, as cited in Pornpitakpan, 2004).
Previous research demonstrated that, in contrast to no disclosures, including a simple disclosure in a blogpost declines the credibility of the source. (Hwang & Jeong, 2016). Consecutively, persuasion knowledge is activated and a consumer may perceive that the brand is biased and trying to persuade them, because they realize the persuasive attempt (Boerman et al., 2012). As a result, users can decline the credibility of the source and oppose the commercial intent of the message, because they experience that the message is unfair or is presented for different intentions than presenting recommendations (Lu et al., 2014).
Furthermore, research has shown that source credibility can affect consumers’ responses (Hwang & Jeong, 2016). If consumers have a positive attitude about the blogger and accept the content of the recommendation post, they will determine to purchase the product (Lu et al., 2014). Thus, the consumer judgment of credibility of the source is considered substantial in the current study, because the credibility determines the acceptance of the persuasive message and ultimately influences consumers’ responses (Lee & Koo, 2012; Lu et al., 2014). However, it is not considered as the main interest of the study, but incorporated as a covariate.
Table 4.1. Conditions experimental design
Vlog Sponsorship disclosure
modality
Brand
1 Dual-Modality disclosure L’Oréal 2 Dual-Modality disclosure PS Pro 3 Visual disclosure L’Oréal 4 Visual disclosure PS Pro 5 Auditory disclosure L’Oréal 6 Auditory disclosure PS Pro 7 No disclosure L’Oréal 8 No disclosure PS Pro
4. Research design
Conceptual framework has been developed and the hypotheses have been formed by providing detailed theoretical background. In the following chapter the research approach, stimuli material and data collection will be discussed, an explanation of how the hypotheses are analysed is followed. In the next chapter the results will be discussed in the discussion.
4.1. Research approach
To test the hypotheses, a 4 (sponsorship disclosure modality: dual, auditory, visual and no disclosure) x 2 (brand familiarity: familiar brand (L’Oréal) vs unfamiliar brand (PS Pro) between-subjects factorial design was used. The control condition was assigned to the no disclosure condition. An overview of the conditions can be found in table 4.1. Thus, an experimental approach was used with eight conditions. Once participants gave their informed consent, they were randomly assigned to one of the eight conditions of the online experiment. The dependent variables were brand attitude and purchase intention. Persuasion knowledge was mediator and familiarity was the moderator in this relationship. The control variables were source credibility, product involvement, experience with watching vlogs, and review positivity
4.2. Stimuli material
To develop eight conditions, eight vlogs were created (Appendix A1). The condition including a visual disclosure, “This vlog is sponsored,’’ was shown 3 seconds prior to the auditory sponsorship disclosure, “It is a sponsored partnership with PS Pro’’ or, “It is a sponsored partnership with L’Oréal.’’ The eight vlogs were of approximately the same length (M: 2.40 minutes). The disclosures followed the requirements of ‘Social Code: YouTube’ (2017). The vlogs were developed with the use of a canon camera, to ensure the quality of the video. The environment in which the vlog was recorded was a living room to keep it similar to other vlogs. The vlogger was a female, age 26 years, and the name of the vlogger was fictitious.
4.2.1. Pre-test
To check for manipulation of brand familiarity, the moderating variable, a pre-test was conducted. 20 women, aged between 18 – 35 participated in the pre-test. The study chose cosmetic brands, because vloggers often write product reviews about cosmetic products. First the participants in the pre-test were asked if they could name 3 cosmetic brands. Afterwards, the participants were asked if they knew L’Oréal or PS Pro (yes or no). Finally, the participants were asked if they were (un)familiar with the two brands of foundations, L’Oréal and PS Pro. The extent to which participants where familiar with the brand was measured on a seven-point scale (1 = “not at all familiar’’ and 7 = “very familiar’’) (Wei et al., 2008).
The results of the pre-test are as follows. 20 participants, all of whom were female with an average age of 25-34, participated in the pre-test. Overall, L’Oréal was named 20%, Maybelline, 16,67%, Max-Factor and Mac 15%, Essence and Chanel 5% and Rimmel, Yves Saint Laurent, Clinique, Hema, Kiko, Tarte, Lancome and Dior, 3,33%. Therefore, L’Oréal was the most familiar cosmetic brand. In addition, all participants who knew L’Oréal and
familiar at all with PS Pro (M = 1,05). Thus, PS Pro was treated as the unfamiliar brand and L’Oréal as the familiar brand for foundation products. Results assure that the conditions will differ on brand familiarity.
4.2.2. Pilot-test
10 participants, all of whom were female, participated in the pre-test of the main experiment. Overall no problems occurred, however, participants found it unusual that during the vlog the next button was not visible. Clear instructions were added to inform the participants to watch the entire vlog, before they could click on the next button.
4.3. Data collection
Qualtrics was used to design the survey and collect the data (Appendix A2). The survey was distributed online through WhatsApp, Facebook, Instagram and Email. Snowball sampling was used, subjects were asked to find future subjects to participate in the experiment. The sample included Dutch women. The reason for most of the sample group being aged between 18 – 34 is that this age range has the highest percentage of people who have ever viewed a vlog, 82%. This is in contrast with people who are aged 50+, where only 45% have ever viewed a vlog (Multiscope, 2016).
Before subjects could participate in the experiment they needed to sign an informed consent. Only women could participate in the experiment, therefore the first question topic was gender. If any men started the experiment they were directed to the end of the experiment and thanked for their time. Afterwards, the participants proceeded to the next question, instructions were shown that the participants should watch the vlog and that the next button would appear after they had watched the vlog. A time slot was included to ensure that participants watched the vlog and were exposed to the disclosures. The vlog was uploaded on YouTube to ensure reliability, as most vlogs are watched via YouTube (Gibson et al., 2008).
To filter out participants that were randomly clicking through the experiment, a random question was added to ensure attention. If participants gave the right answer they proceeded, otherwise they were directed to the end of the experiment and thanked for their time. To make sure that participants were familiar with L’Oréal or unfamiliar with PS Pro, brand familiarity was checked. If participants did not know L’Oréal or knew PS Pro, they were excluded from the data set (see, 4.3.1. Data cleaning). The survey included the following measures respectively: brand attitude, purchase intention, conceptual and attitudinal persuasion knowledge, product involvement, source credibility, positivity of the review, frequency of watching vlogs, income, age, education, nationality. IBM SPSS statistic version 24 was used to analyse the data.
4.3.1. Data cleaning
In total 320 participants participated in the experiment of whom all were female. Before conducting the main analysis, errors were checked within each variable, by executing a frequency analysis. A few errors were detected, one participant did not finish the experiment and was deleted from the data set. Five participants knew PS PRO and were deleted, because participants should not know PS PRO. In addition, three participants did not know or were not familiar with L’Oréal, and were deleted from the dataset as well, because participants should know or needed to be familiar with L’Oréal. Furthermore, no unexpected or unusual values were identified. The dataset was checked for outliers, by standardization of the scores and conducting a frequencies analysis of the standardized variables. There were no possible outliers, cases with z > |3, identified.
4.3.2. Manipulation check
It was examined if manipulation of the independent variable, modality of the sponsorship disclosure, was correctly executed. In the no disclosure condition, 84.83%
Table 4.2. Sponsorship disclosure type recall No disclosure Verbal disclosure Textual disclosure
Dual disclosure Total
No disclosure 62 0 0 0 62
Disclosure Total 74 67 45 52 238
Textual disclosure 22 4 44 5 75
Verbal disclosure 19 46 8 9 82
Dual disclosure 11 17 22 31 81
disclosure, 58.67% correctly recalled the modality of sponsorship disclosure. In the condition of an auditory disclosure, 56.10% correctly recalled the modality of sponsorship disclosure. Finally, in the condition of dual-modality disclosure condition, 38.27% correctly recalled the modality of sponsorship disclosure. Overall, 56.27% correctly recalled the modality of sponsorship disclosure.
11 participants wrongly concluded that they were exposed to a disclosure, while they were assigned to the no disclosure condition. These participants were excluded from further analysis, because they could manipulate the results of the baseline condition. Participants who wrongly concluded that they were in a no disclosure condition were not deleted from the data set, as these are important results to determine the effect of a sponsorship disclosure.
Overall, there was a significant difference in the participants who were exposed to a disclosure, but recalled that they were not exposed to a disclosure (21.84%) and consumers who correctly recalled that they were exposed to a disclosure (84.93%), X2 (2) = 87.624, p <0.001. The manipulation of sponsorship disclosure is succeeded. In table 4.2, an overview is given of participants recall of sponsorship disclosure type.
4.3.3. Final dataset
A final dataset of N=311 participants, all of whom were Dutch-speaking women, was left for further analysis. The average age was between 18-24, which accounted for 89% of the participants. To better understand the relationship between the dependent and independent
variables, covariates, source credibility, product involvement, frequency of watching vlogs and positivity of the vlog were all included in the analysis.
4.4. Measurement scales
The measurement scales of dependent variables, brand attitude and purchase intention, the mediation variable, persuasion knowledge, the control variables and manipulation checks, are described in this section.
Dependent variables
Brand attitude
The dependent variable brand attitude was measured with five items on a seven-point semantic differential scale (Dekker & Reijemersdal, 2013). Participants were asked whether they found the brand “Unappealing/Appealing’’, “Bad/Good’’ ,“Unlikable/Likable’’, “Unpleasant/Pleasant’’ and “Unfavorable/Favorable’’. Furthermore, brand attitude was measured by asking participants to indicate to what extent (1: strongly disagree – 7: strongly agree) they agreed with the four below statements (Yoo, Donthu & Lee, 2000). The X stands for either the brand L’Oréal or PS Pro
(1) “It makes sense to buy X instead of any other brand, even if they are the same. (2) “Even if another brand has the same features as X. I would prefer to buy X (3) “If there is another brand as good as X. I prefer to buy X
(4) “Even if another brand is not different from X in any way, it seems smarter to purchase X. The mean score of brand attitude was measured to create a single measure. Reliability analysis showed the scale to be reliable (M=3.77, SD=1.16, α= 0.82).
