Persuasion effectiveness of sponsored content by digital influencers : the influence of physical attractiveness and identification on source credibility as clarification for the effects sponsorship of disclosure

(1)

Persuasion Effectiveness of Sponsored

Content by Digital Influencers

The Influence of Physical Attractiveness and Identification on Source

Credibility as Clarification for the Effects Sponsorship of Disclosure

Kiki Soeters 11400935

23rd of June 2017 Final Draft

MSc. Business Administration – Marketing Amsterdam Business School, UvA

(2)

Statement of Originality

This document is written by Kiki Soeters, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and 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.

(3)

Acknowledgement

First of all, I would like to thank my supervisor, Frank Slisser, for his support and advise during the process of writing this thesis. His feedback and critical view stimulated me to challenge myself. In addition, I would like to thank him for taking the time to guide me in the process of writing my thesis.

Second, my gratitude goes to my family and friends for their on-going support and

understanding, especially during my thesis. Because of their patience and involvement I was able to complete my thesis. Thank you for being there for me.

(4)

3.1 Design ... 44 3.2 Sample ... 45 3.3 Manipulation ... 46 3.3.1 Pre-test ... 48 3.4 Measurement of Variables ... 49 3.5 Data Collection... 52 3.6 Data Analysis ... 53 4 Results ... 55 4.1 Manipulation Check ... 55 4.2 Correlation Analysis ... 56 4.3 Serial Mediation ... 57 4.4 Moderated Mediation ... 60 4.5 Additional Analyses ... 64 5 Discussion ... 68

5.1 The Influence of Physical Attractiveness ... 68

5.2 The Effects of Sponsorship Disclosure... 72

5.3 Theoretical contributions and Practical Implications ... 75

5.3.1 Theoretical Contributions ... 75

5.3.2 Managerial Implications ... 77

5.4 Limitations and Further Research Directions ... 79

6 Conclusion ... 82

References ... 84

Appendices ... 93

Appendix 1: Source credibility scales by Ohanian (1990) ... 93

Physical attractiveness ... 93

(6)

Expertise ... 93

Appendix 2: Identification scale by Basil (1996) ... 94

Appendix 3: Brand attitude scale by Spears and Sing (2004) ... 94

Appendix 4: Purchase intention scale by Xia and Bechwati (2008) ... 95

Appendix 5: Attitude toward the ad scale by MacKenzie and Lutz (1989) ... 95

Appendix 6: Complete Survey ... 95

Appendix 7: Frequency Histograms ... 101

List of Tables

Table 1: The conditions in the 3 x 2 between-subjects design

Table 2: Cross tabulation between sponsorship disclosure conditions Table 3: Means, Standard Deviations, and Correlations

Table 4: Results of process analyses: Physical Attractiveness, Identification, Perceived Source Credibility, Brand Attitude and Purchase Intention

Table 5: Results of process analyses: Sponsorship Disclosure, Identification, Perceived Source Credibility and Brand Attitude

Table 6: Distribution of respondents per condition

Table 7: Results of process analyses: Sponsorship Disclosure, Identification, Perceived Source Credibility and Brand Attitude

List of Figures

Figure 1: Conceptual model

Figure 2: Outcomes of process analyses, serial mediation Figure 3: Outcomes of process analyses, moderated mediation Figure 4: Outcomes of process analyses, additional analyses

(7)

Abstract

With the proliferation of social media usage, marketers are increasingly utilizing digital influencers in advertising practice. Therefore, the effects of sponsored content by digital influencers, which entails compensating digital influencers for communicating information about a company or brand, has obtained increasing interest over the past years. Previous studies confirmed the importance of perceived source credibility with regard to effective persuasion of sponsored content, especially when the sponsorship is disclosed. This study investigates the relation between sponsored content and persuasion effectiveness. More specifically, the influence of physical attractiveness of and identification with the digital influencer on perceived credibility and persuasion effectiveness when disclosure of a sponsorship is involved. A 3 x 2 experimental between subject’s design tested data from a sample of 213 female Instagram users. The findings confirmed that there was a relation between sponsored content by digital influencers and persuasion effectiveness. However, in this study this relationship was fully mediated by physical attractiveness, identification and perceived source credibility in serial. Interestingly, the results revealed a sponsorship

predisposition among the Instagram users. Notably, results regarding Instagram users without the predisposition indicate that sponsorship disclosure affects perceived credibility negatively through identification with the digital influencer. In conclusion, through comprehensive explanation and empirical assessment of sponsored content, this study establishes important contributions by providing a deeper understanding regarding the dependence of persuasion effectiveness of the sponsored content on Instagram users’ perception regarding physical attractiveness, identification and credibility of the digital influencer.

Keywords: source credibility; sponsorship disclosure; digital influencers; Instagram

(8)

1 Introduction

In the age of social media, companies are increasingly seeking advertising opportunities to leverage from the emergence of this communication channel (Kaplan & Haenlein, 2010). The rise of the Internet and Web 2.0 offers marketers various advertising opportunities (Mangold & Faulds, 2009). With the number of global social media users reaching up to 2.34 billion in 2016, which entails approximately a third of the world’s population, social media appears as a viable marketing vehicle (Statista, 2017). In a study on the value of content on Twitter, Swant (2016) found that 49% of Twitter users rely on recommendations of other Twitter users. Besides, companies are developing innovative advertising strategies to cope with the stagnating effectiveness of these traditional media as consumers are turning away from traditional advertising media (Droge et al., 2010; Parker et al., 2008; Wright et al., 2010). A social media channel that is increasingly used as a marketing vehicle to influence consumers is Instagram (Woods, 2016). Instagram is ranked as number 6 in leading social networks worldwide and had over 600 million monthly active accounts in 2016 (Statista, 2017). As a rapidly growing social medium, Instagram offers a platform for digital influencers and other social media users to share their opinion and feelings (Hu & et al., 2014). Although extensive research has been conducted on social networking sites such as Facebook and Twitter, less attention has been given to the photo (and video) capturing and sharing service, Instagram (Hu & et al., 2014).

Another important factor, together with the expansion of social media that opens doors for marketers, is the increase of digital influencers and the influence they exert on social media (Uzunoğlu& Kip, 2014). Digital influencers are individuals that exert influence on the decisions of others by sharing their opinions online, including social media (Uzunoğlu & Kip, 2014). Companies can close deals with digital influencers for communicating company-related information, which is regarded in this study as advertising through sponsored content

(9)

by digital influencers (Chu & Kamal, 2008; Uzunoğlu& Kip, 2014; Chang et al., 2015; Liljander et al., 2015). Leveraging digital influencers as marketing tactic is becoming more and more popular, which is reflected in the increase of money spent on this type of marketing (Weinswig, 2016). The emergence of digital influencers causes necessity to evaluate their influence on consumers. It is important because these digital influencers have the ability to persuade consumers online with attitudinal and behavioural changes as a consequence (Chu & Kamal, 2008; Hsu et al., 2013; Lu et al., 2014).

In influencing consumers, perceived credibility of the digital influencer is a feature which is becoming more and more important (Metzger, 2010; Westerman et al., 2014). Source credibility perceptions are important because it influences the acceptance of the message (Ohanian, 1990). These credibility perceptions are becoming even more important due to the growing amount of information available online and the diminishing of traditional gatekeepers (Metzger, 2010; Westerman et al., 2014). As a consequence, consumers are designated to assess credibility of information sources themselves (Metzger, 2010; Westerman et al., 2014). This growing importance of source credibility in the online

environment affects the influence digital influencers may exert on consumers (Westerman et al., 2014).

As a result of the increase in marketing spending on sponsored content by digital influencers, guidelines were developed to regulate this type of marketing and to protect consumers (Boerman & van Reijmersdal, 2016). Content shared by digital influencers on Instagram is generally not regarded as advertisements, causing blurring lines between organic content and sponsored content (Liljander et al., 2015; Woods, 2016). Therefore, regulations regarding sponsorship disclosure were introduced to reduce uncertainty of social media users about compensation of digital influencers (Liljander et al., 2015). The aim of these regulations is to protect consumers by creating awareness of the persuasive intent of the message

(10)

(Boerman & van Reijmersdal, 2016).

Regulatory pressure to disclose sponsorship deals might influence the effectiveness of the type of advertising researched in this study, through sponsored content by digital

influencers on Instagram (Carr & Hayes, 2014; Colliander & Erlandson, 2015; Liljander et al., 2015). Since marketing spending and regulations regarding advertising through digital influencers is increasing, it is important to understand how sponsorship disclosures affect credibility perceptions of digital influencers and under which conditions these effects are mitigated.

In recent years, extensive research regarding the relationship between source credibility and persuasion effectiveness in advertising has been conducted (Ohanian, 1990; Pornpitakpan, 2004). Regarding commercial settings persuasion is considered as effective when it causes attitudinal and behavioural changes regarding brand attitude and purchase intention (Petty & Cacioppo, 1981). Studies focusing on social media show that product recommendations on blogs and electronic word-of-mouth (WOM) on social networking sites can increase persuasion effectiveness (Prendergast et al., 2010; Colliander & Dahlén, 2011; Wang et al., 2012; Lu et al., 2014; Dehghani & Tumer, 2015).

Till date, no research has been conducted on persuasion effectiveness of digital influencers on Instagram (Hu et al., 2014). Although, digital influencers influencing

consumers on other media, such as blogs and social networking sites, has been widely studied (Prendergast et al., 2010; Colliander & Dahlén, 2011; Wang et al., 2012; Lu et al., 2014; Dehghani & Tumer, 2015). Therefore, Instagram as medium for persuading consumers through sponsored content by digital influencers is a differentiating approach in this study.

The addition of physical attractiveness as constituent for measuring source credibility on social media is another aspect which is differentiating in this study. Previous research on perceived credibility on social media has only included expertise and trustworthiness as

(11)

constituents for measuring source credibility (Brown et al., 2007; Chu & Kamal, 2008; Chu & Kim, 2011; Wu & Wang, 2011; Hsu et al., 2013; Spence et al., 2013). Instagram is a photo-sharing application which is visually oriented, reason to include physical attractiveness as constituent of source credibility in this study. Since the focus of Instagram is on pictures, physical attractiveness might be even more influential than expertise and trustworthiness compared to other, less visually oriented media.

Recent studies show that the introduction of sponsorship disclosure guidelines affects credibility perceptions and persuasion effectiveness of digital influencers, with mixed results (Liljander et al., 2015; Boerman & van Reijmersdal, 2016). The majority of these studies found negative effects of disclosing the sponsorship on credibility perceptions and persuasion effectiveness (Campbell et al., 2012; Carr & Hayes, 2014; Liljander et al., 2015). Most of these studies focused on measuring the influence of sponsorship disclosure on source credibility and persuasion effectiveness separately (Campbell et al., 2012; Colliander & Erlandson, 2015; Liljander et al., 2015; Hwang & Jeong, 2016). Only one study researched the effect of perceived credibility on perceived WOM influence as a consequence of disclosing a sponsorship (Carr & Hayes, 2014). However, in this study, the influence of different types of sponsorship disclosure on persuasion effectiveness was not included. As credibility perceptions of a digital influencer will influence brand attitude and purchase intention (Chu & Kamal, 2008; Hsu et al., 2013), this study will include the combined effects of disclosing a sponsorship on perceived credibility and persuasion effectiveness. Perceived source credibility has not been studied as explanation for the mainly negative effects of disclosing a sponsorship on persuasion effectiveness yet. This entails another aspect which is added in this study.

The effects of disclosing a sponsorship on social media entail a relatively unexplored body of research since regulations and guidelines regarding sponsorships are relatively new

(12)

(Boerman & van Reijmersdal, 2016). Conditions under which the mainly negative effects of sponsorship disclosure on consumer responses mitigate remained unstudied (Boerman & van Reijmersdal, 2016). Therefore identification with the digital influencer concerning

sponsorship disclosure has been included in this study. Furthermore, celebrity endorsement literature shows that identification with a celebrity strengthens the influence a celebrity may exert on an individual (Basil, 1996; Fraser & Brown, 2002; Jin & Phua, 2014). One could assume that the negative effects of sponsorship disclosure on credibility perceptions and persuasion effectiveness might be mitigated by identification with the digital influencer. Therefore, identification with a digital influencer is a construct new in the body of research regarding source credibility and sponsorship disclosure. This thesis will seek to answer the following question:

‘How does identification and perceived source credibility, as well as physical attractiveness specifically, influence the effects of disclosing a sponsorship on persuasion effectiveness of sponsored content of digital influencers on Instagram?’

Practitioners and researchers can benefit from the findings of this study as this study will gain insights into the persuasion effectiveness of a newly emerging type of advertising, through sponsored content by digital influencers on Instagram. The outcomes of this study can be used to improve advertising effectiveness as it will show the impact of source

credibility and more specifically, the influence of physical attractiveness. Also, this study will demonstrate how disclosing a sponsorship of sponsored content by digital influencers affects credibility perceptions and persuasion effectiveness, separately as well as combined.

Furthermore, a new condition influencing persuasion effectiveness of online sources will be introduced, identification with the source, which entails, in this case, the digital influencer.

(13)

The results of this study will help practitioners to select digital influencers that will provide effective persuasion. The inclusion of the identification theory will demonstrate new insights and will extend which is literature currently available on persuasion effectiveness regarding source credibility and sponsorship disclosure.

In the following chapter, the previously mentioned concepts and the relationships between these concepts will be explained in more detail. After that, the conceptual model developed from the reviewed literature will be presented. Also, this chapter will address the hypotheses based on the literature. The method section will explain the collection and

measurements of the data, after which the results will be presented. Followed by a discussion in which the results, implications, limitations and suggestions for future research will be considered. Finally, a general discussion will be drawn which will highlight the most prominent findings.

(14)

2 Literature Review

This chapter will provide a review of current literature about the key concepts in this study. Based on the reviewed literature a research question will be composed. First, concepts concerning the type of advertising studied in this thesis will be clarified. Followed by an explanation of persuasion effectiveness, which will elaborate on advertising and electronic word-of-mouth. Consecutively, the literature on sponsorship disclosure and the effects of disclosing a sponsorship will be reviewed. Hereafter, source credibility and the influence on persuasion effectiveness will be discussed. The influence of source credibility on social media will be discussed as well. Finally, identification theory and its connection with the other concepts will be analysed.

2.1 Social Media as Marketing Tool

The emergence of the Internet and the growth of social media users has ensured that social media has become a valuable marketing tool (Kaplan & Haenlein, 2010). This chapter will start with the discussion of the marketing possibilities that social media offers. Thereafter, particular attention will be given to Instagram as social media application and digital influencers as social media users. This results in an explanation of advertising through

sponsored content by digital influencer on Instagram, a marketing opportunity enabled by the emergence of social media.

Before discussing advertising on Instagram through sponsored content by digital influencers, it is important to clarify the concept of social media and the promotional opportunities social media offers. Kaplan and Haenlein (2010) define social media as: ‘a group of Internet-based applications that build on the ideological and technological

(15)

(p. 61). Examples of content are post videos, photos, music, blogs, wikis, personal profiles and personal web sites (Kaplan & Haenlein, 2010)

Blogs and social networking sites are regarded as separate social media applications and will be discussed in the following paragraphs (Kaplan & Heanlein, 2010). The difference in media richness and social presence is the primary distinction between these social media application, whereby social networking sites incorporate a higher degree of media richness and social presence (Kaplan & Heanlein, 2010). The focus of this study is on Instagram, which is a visually oriented medium with a high level of self-disclosure as it aimed at sharing photos (Kaplan & Heanlein 2010; Woods, 2016). Other social media applications with fewer opportunities for self-disclosure and self-presentation will, therefore, be excluded for the purpose of this study. (Kaplan & Heanlein, 2010).

Several advertisement opportunities can be distinguished related to blogs (Droge et al., 2010). Companies can create blogs, offer products to bloggers for free, provide blog

sponsoring, pay bloggers for steering readers to company websites and pay bloggers to advertise brand information (Dorge et al., 2010; Chu & Kamal, 2011; Liljander, 2015). Research about blog marketing shows the influential possibilities of this type of marketing. Onishi and Manchanda (2012) found that blogging data in the movie market could be used as a predictor for sales. They found a positive relation between blogging activity and market implications, such as increased turnover rates and sales, indicating the impact of being mentioned on blogs.

Regarding social networking sites three marketing possibilities, similar to blog marketing, can be identified. Social networking sites offer possibilities for a brand fan page, sponsored advertisements and WOM opportunities, of which the effects of WOM is most commonly studied (Maurer & Wiegman, 2011; Prendergast et al., 2010; Wang et al., 2012; Dehghani & Tumer, 2015). Studies on brand fan pages and WOM on social networking sites

(16)

have proven its effectiveness regarding an increase in customer equity, such as brand attitude and purchase intention (Prendergast et al., 2010; De Vries et al., 2012; Wang et al., 2012; Dehghani & Tumer, 2015).

The majority of research on effects of marketing activities on social networking sites focuses on Facebook, Twitter and YouTube (Erkan, 2015). A less commonly studied social network sites type is Instagram, which will be discussed in the following chapter.

2.1.1 Instagram

This chapter will briefly explain Instagram as a social media application followed by a literature review concerning Instagram.

The features of the relatively new and fast growing social media application Instagram will be discussed (Hu & et al., 2014). The application provides users with the opportunity to capture and share photos and videos with other users (Hu et al., 2014; Erkan, 2015). Users can follow other Instagram users with similar interests or commonalities, which are called

“followers” (Hu et al., 2014). On these photos and videos, several filters can be placed through which the content appeals differently (Erkan, 2015). Instagram users can

communicate with each other by leaving comments on the content updated (Erkan, 2015). As a result of the following and commenting features, Instagram can be distinguished as a social media application (Kaplan & Heanlein, 2010).

Current research conducted on Instagram merely focuses on characteristics of content shared by users (Hochman & Schwartz, 2012; Bakhshi et al., 2014; Hu et al., 2014). A study concerning content on Instagram concluded that photos containing faces received more responses regarding likes and comments compared to photos without faces (Bakhshi et al., 2014). Another research analysing content on Instagram identified five different types of Instagram users based on eight common photo categories (Hu et al., 2014). Hochman and

(17)

Schwartz (2012) recognized cultural differences in colour patterns, number of images viewed in a specific time period and differences in image attributes such as colour tone and brightness during a certain day between content on Instagram in two different cities.

However, research on Instagram as a marketing tool is still limited (Erkan, 2015). One study on marketer generated content on Instagram found sectorial differences in engagement ratios regarding customer responses and WOM (Erkan, 2015). These different responses varied per sector due to product orientation differences in content per section and differences in information required about types of products (Erkan, 2015).

To date, no research has been conducted concerning the marketing possibilities of digital influencers on Instagram. Before discussing these marketing opportunities, the concept of digital influencer will be clarified in the following chapter.

2.1.2 Digital Influencers

This chapter will define digital influencers according to the literature. In marketing literature, commonly used concepts entail influencers and social media influencers. This section

analyses the similarities, differences and nuances between influencers, social media influencers and digital influencers. The concept of digital influencer will be defined in the context of this study.

In literature, several overlapping concepts regarding digital influencers can be found (Kiss & Bichler, 2008; Freberg et al., 2011; Uzunoğlu & Kip, 2014). These concepts, influencers and social media influencers, will be analysed to describe and define digital influencers.

A broad concept commonly used to describe the social impact of individuals regarding conversations, is influencers (Kiss & Bichler, 2008; Li et al., 2011). ‘Influencers supply the authority that allows a message to be conveyed quickly and reliable through WOM

(18)

techniques’ (Kiss & Bichler, 2008, p. 234). The marketing implication of these influencers is the possibility to reach a large number of people rapidly and to attain preferred behavioural responses (Kiss & Bichler, 2008).

A more specific concept entails social media influencers, which Freberg and colleagues (2011) define as ‘a new type of independent third party endorser who shape audience attitudes through blogs, tweets, and the use of other social media’ (p. 90). The persuasive power of these social media influencers is bounded to social media (Freberg et al., 2011). Persuasion of influencers through WOM is not restricted to social media as other channels for communication are attainable as well (Kiss & Bichler, 2008; Li et al., 2011).

In literature, the following definition of digital influencers is described. ‘Digital influencers can mediate messages and affect communities in the digital environment, where messages can be disseminated rapidly and easily with a potentially viral effect’ (Uzunoğlu & Kip, 2014, p. 592). One could derive the conclusion that this definition overlaps with the definitions of influencers and social media influencers. The scope of influencers’ persuasive power is broader compared to digital influencers as the concept of digital influencers is bounded to the digital environment. By contrast, the scope of social media influencers is smaller as these kinds of influencers are limited to social media in exerting influence. This study will incorporate digital influencers as it encompasses all conversations online.

From the analyses of influencers and social media influencers, digital influencers will be regarded in this study as individuals that exert influence on the decisions of others by sharing their opinions online, including social media. The following chapter will describe how companies can leverage digital influencers for advertising purposes.

(19)

2.1.3 Advertising through Sponsored Content by Digital Influencers

The social media application Instagram offers opportunities for advertising through digital influencers. Social media applications, including Instagram, enable digital influencers to exert influence on social media users by sharing their opinions (Kaplan & Haenlein, 2010; Wright et al., 2010; Uzunoğlu & Kip, 2014; Woods, 2016). The ability of digital influencers to influence Instagram users offers opportunities for advertising (Uzunoğlu & Kip, 2014).

This research will examine a new, emerging way of advertising, through sponsored content by digital influencers on Instagram, which entails communicating brand-related information through digital influencers (Uzunoğlu& Kip, 2014; Wood, 2016). Marketers or companies can compensate digital media influencers for disclosing information about a

company, product or service (Hsu et al., 2013; Liljander et al., 2015; Kapitan & Silvera, 2016; Woods, 2016). Similar concepts addressed in literature are sponsored content, paid product recommendations, branded content and product placement (Chu & Kamal, 2008; Boerman & van Reijmersdal, 2016). In blog marketing, sponsored product recommendation is often applied to describe blogs posts that are sponsored by a third party (Lu et al., 2014). In this study, the manner of advertising which entails compensating digital influencers for communicating information about a company or brand on Instagram will be referred as sponsored content by digital influencers on Instagram. Thus, with this type of advertising the digital influencers are compensated by a brand or company for the placing sponsored content (Uzunoğlu& Kip, 2014; Woods, 2016).

Companies are increasing their investments in compensating digital influencers for disclosing company-related information (Uzunoğlu& Kip, 2014; Woods, 2016). It is important to study whether this advertising method is effective. Therefore, persuasion effectiveness of these digital influencers will be discussed in the following chapter.

(20)

2.2 Persuasion Effectiveness

This chapter will discuss in which situations persuasion attempts can be regarded as effective. Thereafter, persuasion effectiveness will be discussed in more detail regarding advertising and electronic WOM. Persuasion can be defined as actively trying to influence people’s behaviour or beliefs by overtly anticipating on their reason or emotion (Lee & Xia, 2011). In general, persuasion is considered effective when the attempt to influence results in an attitudinal and behavioural change. This change in attitude and behaviour can differ in persistence,

depending on the level of involvement (Petty & Cacioppo, 1986). Cues processed under higher levels of involvement have a stronger impact on persuasion effectiveness, regarding the persistence of attitudinal change (Petty & Cacioppo, 1986). Besides involvement, several other factors influencing persuasion effectiveness specifically aimed at advertising can be identified,.

2.2.1 Persuasion Effectiveness in Advertising

In literature, persuasion of an advertisement is considered as effective when it results in an attitudinal change concerning brand attitude and purchase intention (Petty & Cacioppo, 1981; Petty et al., 1983). According to, effective persuasion occurs when brand attitude and

purchase intention are influenced through the advertisement cue (Petty et al., 1983). The most commonly studied features that influence persuasiveness of an advertisement message are content, medium and source (Petty & Cacioppo, 1981;

Pornpitakpan, 2004). Source credibility is most frequently researched concerning persuasion effectiveness of an advertisement due to its importance in evaluating a message

(Pornpitakpan, 2004). Moreover, source credibility is becoming increasingly important because of the amount of information available online (Metzger, 2010; Westerman et al., 2014). In the online environment, traditional gatekeepers are becoming absent, with the

(21)

consequence that consumers of content online are designated to assess the credibility of the source themselves (Metzger, 2010; Westerman et al., 2014). Due to the increasing importance of source credibility, especially in the social media environment, an extensive review on source credibility will be conducted in another chapter. Before discussing source credibility, previous research on persuasion effectiveness of marketing activities on social media will be discussed.

2.2.2 Persuasion Effectiveness on Social Media

This chapter will elaborate on persuasion effectiveness of electronic word of mouth on social media. A study on offline and online WOM mentioned that 50% of WOM recipients base their purchase decision on the WOM information they receive (Keller, 2007). These results indicate the persuasive power WOM has.

For the purpose of this study, the focus will be on electronic word of mouth because Instagram is a social network operated online. eWOM is often defined as ‘any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet’ (Hennig-Thurau et al., 2004, p.39). Another term frequently used concerning WOM is product recommendations (Liljander et al., 2015).

Following the definition of eWOM, advertising through sponsored content by digital influencers can be regarded as a type of eWOM, because digital influencers communicate company-related statements (Hennig-Thurau et al., 2004; Uzunoğlu & Kip, 2014). Therefore, it is important to consider how eWOM affects attitudes and purchase intention.

Studies on eWOM usually focus on one social media application, such as a specific social networking site such as Facebook (Wallace et al., 2011). Because blogs and social networking are distinct social media applications they will be discussed separately (Kaplam &

(22)

Haenlein, 2010). First, the influence of eWOM on persuasion on blogs will be discussed, followed by eWOM on social networking sites.

Literature about persuasion on blogs mainly focuses on blog sponsoring (Chu & Kamal, 2008; Hsu et al., 2013; Lu et al., 2014; Liljander et al., 2015). Multiple studies confirm persuasion effectiveness of blog sponsoring, and thus a changes in attitude towards the brand and an increase in purchase intention (Chu & Kamal, 2008; Hsu et al., 2013; Lu et al., 2014). Several variables influencing this relationship are taken into account. Perceived blogger trustworthiness has a positive influence on purchase intentions, and this relationship is mediated by perceived usefulness of the recommendation and argument quality (Chu & Kamal, 2008; Hsu et al., 2013). A study focusing on the credibility of product

recommendations mentioned that blogger recommendations of search goods as well as goods with high brand awareness, resulted in an increase in perceived recommendation credibility. Attitude towards the sponsored recommendation and purchase intentions increased when the product recommendation was perceived as credible (Lu et al., 2014). Colliander and Dahlén (2011) compared publicity on online magazines with blogs and concluded that blogs have more impact on brand attitudes and purchase intentions.

Another important variable influencing persuasion effectiveness of blog sponsoring is sponsorship disclosure (Liljander et al., 2015). Because of the blurring lines between genuine product recommendations on social media and sponsored product recommendation and governmental pressure on disclosure regulations, sponsorship disclosure is becoming more important (Woods, 2016). Sponsorship disclosure will be discussed in another chapter due to the growing influence sponsorship disclosure may exert on persuasion effectiveness.

Research shows that blog sponsoring can yield positive results for companies in terms of purchase intentions, due to the influence of sponsorship disclosure, recommendation credibility and perceived usefulness of the recommendation (Chu & Kamal, 2008; Hsu et al.,

(23)

2013; Lu et al., 2014; Liljander et al., 2015). Hereafter, the second social media application for persuasion, social networking sites, will be discussed.

Electronic WOM on social networking sites tend to exert a positive influence on attitudinal change and purchase intention (Prendergast et al., 2010; Wang et al., 2012; Dehghani & Tumer, 2015). Peer communication on social networking sites can influence purchase intentions directly through conformity and indirectly by increasing product involvement (Wang et al., 2012). A study confirming the positive influence of eWOM on purchase intention noticed that purchase intentions are affected through similarity. When the topics discussed are similar to the reader’s interested, online WOM has a direct influence on attitude, which in turn influences purchase intentions directly (Prendergast et al., 2010). Another study examined consumers’ opinion on Facebook advertising, which is a type of eWOM since a statement about the advertisements is made. This study showed a positive effect of consumers’ opinion on brand attitude and the positive effect of brand attitude on purchase intentions as well (Dehghani & Tumer, 2015).

Previously described studies show the positive influence of eWOM, although negative online WOM can exert negative influence on brand attitude and purchase intentions (Kozinets et al., 2010). Negative eWOM will not be discussed as it is irrelevant for this study as digital influencers will not engage in negative eWOM because they are compensated for

communicating company-related information.

Research that included both social networking sites and blogs studied the impact of banner ads and advice from others on purchase decisions and therefore persuasion

effectiveness (Wallace et al., 2011). Results indicated that for all the social networks included in the study, eWOM has a much larger impact on persuasion effectiveness, regarding

(24)

Therefore, it can be concluded that eWOM on social media has the ability to exert a positive influence on persuasion effectiveness. As this study is about compensating digital influencers for communicating information about a company or brand on Instagram, which is a type of eWOM, it important to study factors that influence persuasion effectiveness. For this reason, the following chapters will elaborate on variables that might influence the intended change of the advertisement through sponsored content by digital influencers on Instagram in brand attitude and purchase intention.

2.3 Sponsorship Disclosure

This chapter reviews the current literature regarding sponsorship disclosure. First, the concept of sponsorship disclosure and its increasing importance will be explained. Followed by an exploration of the different effects sponsorship disclosure elicited. Finally, the influence of sponsorship disclosures on persuasion effectiveness in the online environment will be discussed.

Within marketing, a distinction between overt and covert marketing can be made (Liljander et al., 2015). With overt marketing “the consumer is aware of the commercial interest of the communicator” (Liljander et al., p. 613, 2015). Overt marketing means disclosing the sponsorship, meaning that the audience is aware of compensations for the promotion. With covert marketing, the sponsorship is not disclosed (Liljander, 2015). Thus, when the digital influencer, who is compensated for communicating information about a company or brand, reveals the collaboration with the company or brand, the sponsorship is disclosed.

The effects of sponsorship disclosure are becoming increasingly important because of regulatory pressure to inform consumers about sponsorships (Boerman & van Reijmersdal, 2016; Woods, 2016). In a growing number of countries, regulations are introduced requiring

(25)

disclosure of sponsorships (Hwang & Jeong, 2016). According to the guidelines of the

Federal Trade Commission of the United States, consumers need to be aware of a sponsorship when the message has a persuasive intention (Woods, 2016). Concerning the Internet and social media, specific guidelines were developed. These guidelines require sponsorship

disclosure of sponsored content by digital influencers in the United States (Woods, 2016). Not disclosing a sponsorship infringes the right of the consumers to be aware of the advertising intent (Boerman & van Reijmersdal, 2016). Having mentioned this, there is still a grey area regarding these guidelines as the regulations concerning sponsorship disclosure differ between countries and types of medium (Woods, 2016).

Attention to sponsorship disclosure in literature has increased, since the development of guidelines requiring sponsorship disclosure (Boerman & van Reijmersdal, 2016). Most studies about the effects of sponsorship disclosure focused on traditional media, such as television, print and radio (Boerman & van Reijmersdal, 2016; Hwang & Jeong, 2016). Research regarding the effects of digital influencers disclosing a sponsorship is still lacking and limited to blog marketing, leaving other social media applications omitted (Hwang & Jeong, 2016). As regulations concerning sponsorship disclosure and advertisements by digital influencers are increasing, the effects of disclosing a sponsorship should be taken into

consideration.

In the following chapter, the effects of sponsorship disclosure on consumer responses will be discussed. Followed by a more in depth focus on the effects of sponsorship disclosure on persuasion effectiveness.

(26)

2.3.1 Effects of Sponsorship Disclosure

This chapter will discuss the effects of sponsorship disclosure. Since the aim of this study is to evaluate persuasion effectiveness regarding attitudinal changes, the focus will be on the effects of disclosure on consumer responses. Consumer responses, such as brand attitude, brand liking and purchase intention, indicate the effectiveness of the persuasive intent as it implies attitudinal changes (Petty & Cacioppo, 1981; Petty et al., 1983).

The majority of the literature focused on consumer responses to sponsored content on traditional media since the guidelines requiring disclosure for sponsorship on social media are relatively new (Boerman & van Reijmersdal, 2016). Studies concerning traditional media, including television, movies, radio and video news releases, mainly show negative effects of disclosing a sponsorship on consumer responses regarding brand attitude and purchase intention (Wei et al., 2008; Tessitore & Geuens, 2013; Boerman et al., 2015). Therefore, the conclusion can be drawn that for traditional media persuasion effectiveness decreases when the sponsorship is revealed.

Besides consumer responses on sponsorship disclosure concerning persuasion effectiveness some positive effects sponsorship disclosure elicits can be found as well. Most studies concerning traditional media, including television and movies, found that disclosing a sponsorship has a positive effect on brand memory and brand recall (Tewksbury et al., 2011; van Reijmersdal et al., 2013; Boerman et al., 2015). Mentioning the sponsorship increases the likelihood that the advertised brand will be remembered (Boerman et al., 2015).

Thus in general, sponsorship disclosures elicit negative effects when considering brand attitude and purchase intention, but the effects concerning brand recall and memory are generally more positive.

Although the majority of studies concerning the effects of sponsorship disclosure found negative results, under certain conditions the negative effects of the disclosure might be

(27)

mitigated (Boerman & van Reijmersdal, 2016). Moreover, effects of disclosure on social media might differ from the results found in studies focusing on traditional media. Therefore, the next chapter will discuss the effects of sponsorship disclosure online.

2.3.2 Effects of Sponsorship Disclosure Online

Research conducted on the influence of sponsorship disclosure in the online environment has focused especially on blogs and advergames (Boerman & van Reijmersdal, 2016). The influence of sponsorship disclosure on blog marketing effectiveness is commonly researched (Carr & Hayes, 2014; Colliander & Erlandson, 2015; Liljander et al., 2015). Most studies concerning sponsorship disclosure on blogs researched the influence of the disclosure on source credibility perceptions (Carr & Hayes, 2014; Liljander et al., 2015; Hwang & Jeong, 2016). Since the following chapter will elaborate on source credibility, including the influence of sponsorship disclosure, this chapter will focus on the other effects of disclosing a

sponsorship.

Studies exploring the effects of disclosing a sponsorship deal concluded that disclosure affected various consumer responses (Campbell et al., 2012; Colliander & Erlandson, 2015; Liljander et al., 2015). Colliander and Erlandson (2015) focused on blog credibility and attitude towards the blog and found that consumer response regarding the blog suffered from the disclosure. Campbell and colleagues (2012) found a negative influence of disclosing a sponsorship on brand recall. In line with these results, another study concluded that when a sponsorship was suspected, purchase intention and the intention to revisit the blog were affected by the disclosure (Liljander et al., 2015). The study of Carr and Hayes (2014) confirms that suspicion of a sponsorship decreases the influence a digital influencer exerts.

Although research mainly shows the negative effect of disclosure on consumer responses, some studies did not found this negative influence on persuasion effectiveness

(28)

(Colliander & Erlandson, 2015; Hwang & Jeong, 2016). With differing disclosure and receiver characteristics, persuasion effectiveness does not suffer from revelation of the

sponsorship deal (Boerman & van Reijmersdal, 2016). The study of Colliander and Erlandson (2015) show that para-social interaction, which entails the illusion of a friendship with a digital influencer, explained why persuasion after disclosing the sponsorship was unaffected. Another study only found a decrease in brand attitude when the sponsorship was mentioned after the product placement, disclosing the sponsorship before the product placement did not affect brand attitude (Campbell et al., 2012).

Thus, research concerning traditional media shows that disclosing a sponsorship negatively influences persuasion effectiveness since the disclosure resulted in lowered brand attitudes and purchase intentions (Wei et al., 2008; Tessitore & Geuens, 2013; Boerman et al., 2015). Studies regarding the influence of sponsorship disclosure on brand attitude and

purchase intention in the online environment found mixed results (Liljander et al., 2015). These mixed results can be assigned to differing disclosure and receiver characteristics (Campbell et al., 2012; Colliander & Erlandson, 2015). Not taking these differing disclosure and receiver characteristics into account, most studies found a negative relationship between disclosing the sponsorship and consumer responses (Campbell et al., 2012; Colliander & Erlandson, 2015; Liljander et al., 2015). Hence, the following hypotheses have been established.

H1 Sponsorship disclosure of sponsored content by a digital influencer on Instagram has a direct negative influence on brand attitude

(29)

2.4 Source Credibility

In this chapter, a literature review on source credibility will be conducted. First, the concept of source credibility will be explained, followed by an exploration of the influence of source credibility on persuasion effectiveness. Thereafter, source credibility on social media will be studied. Finally, literature on the effects of disclosing a sponsorship on credibility perceptions will be discussed.

Source credibility can be defined as: ‘a term commonly used to imply a

communicator’s positive characteristics that affect the receiver’s acceptance of the message’ (Ohanian, 1990, p. 41). Because source credibility is a rather broad concept, this study will focus source credibility in commercial settings.

Until the development of a measurement scale to measure celebrity credibility the dimensions of source credibility were unclear for advertising and commercial settings (Pornpitakpan, 2004). Ohanian (1990) identified a tri-component construct, including

expertise, trustworthiness and physical attractiveness, for measuring persuasion effectiveness of celebrity endorsers. Expertise is ‘the perceived ability of the source to make valid

assertions’ (McGracken 1989, p. 311). A source is perceived with high expertise when a source is perceived as knowledgeable, experienced, qualified and skilled (Ohanian, 1990). Trustworthiness refers to ‘the perceived willingness of the source to make valid assertions’ (McGracken 1989, p. 311). The extent to which a source is perceived as dependable, honest, reliable and sincere determines the trustworthiness (Ohanian, 1990). A source can be regarded as physically attractive when perceived as classy, beautiful, elegant and sexy (Ohanian, 1990).

Source credibility is a major factor that influences persuasion effectiveness; online this influence is becoming even more important (Pornpitakpan, 2004; Metzger, 2010; Westerman et al., 2014).

(30)

In the online environment credibility assessments are becoming increasingly important due to the growth of information available online (Westerman et al., 2014).

Therefore, credibility assessments have become inevitable (Metzger, 2010; Westerman et al., 2014). Another factor increasing the importance of credibility assessments is the diminution of traditional gatekeepers (Metzger, 2010). Gatekeepers present at traditional media channels are less present in the online environment, as a consequence, online information recipients are appointed to execute credibility assessments themselves (Metzger, 2010).

The influence of source credibility on persuasion effectiveness might increase as a consequence of the growing importance of source credibility in the online environment. As the focus of this study is on the social media application Instagram, source credibility is an important factor influencing the effectiveness of the advertisements through sponsored content by digital influencers. For this reason, the influence of source credibility on persuasion effectiveness will be discussed in the following chapter.

2.4.1 Source Credibility and Persuasion Effectiveness

Extensive research about the relation between source credibility and persuasion effectiveness in advertising has been conducted. Findings suggest that source credibility is an essential element of effective persuasion regarding attitudinal and behavioural change (Ohanian, 1990; Pornpitakpan, 2004; Chu & Kamal, 2008; Hsu et al., 2013;). Although, most studies suggest that a source with greater credibility results in higher persuasion effectiveness, a few studies are contradicting these findings (Pornpitakpan, 2004).

The three dimensions identified by Ohanian (1990) have been used in multiple studies confirming their influences on persuasion effectiveness. Amos and colleagues (2008) found that celebrity trustworthiness, celebrity expertise and celebrity physical attractiveness are the

(31)

most detrimental sources determining attitude towards the advertisement, brand attitude and purchase intention.

Physical attractiveness of the source has been shown to influence persuasion

effectiveness (Chaiken, 1979; Baker & Churchill, 1977; 1977; Kahle & Homer, 1985; Patzer, 1983). Physical attractiveness of the communicator can increase persuasion effectiveness, both on verbal and behavioural measures (Chaiken, 1979). Another study confirming the effectiveness of physical attractiveness on persuasion found a difference in persuasion effectiveness between male and female models. This study showed that for female subjects only attractive male models enhances purchase intentions (Baker & Churchill, 1977). Other variables influencing persuasion effectiveness of a physically attractive source, regarding purchase intention, are involvement with the product and communicator liking (Kahle & Homer, 1985; Patzer, 1983). Joseph (1982) found that physical attractiveness significantly influences opinion change. However, when other cues, such as expertise are present, this relationship weakened, which indicates that physical attractiveness alone is not enough for persuasion.

Perceived expertise can exert influence on attitude, where an expert influences

persuasion more compared to a non-expert (Maddux & Rogers, 1980). Source expertise has a larger impact under a low involvement level compared to a higher involvement level, but for both levels of involvement an expert source influences attitudinal change positively (Petty & Cacioppo, 1986). Maddux and Rogers (1980) concluded that under certain conditions an attractive source needs to be complemented with perceived expertise or supporting arguments to be persuasive. This confirms the different dimensions a source needs to possess for

effective persuasion.

Although most research confirms the positive influence of trustworthiness on persuasion effectiveness, not all research on source credibility supports the effectiveness of

(32)

source trustworthiness (Pornpitakpan, 2004). Priester and Petty (2003) argue that an

untrustworthy endorser might have more impact on attitude formation because a message of an untrustworthy endorser will be more scrutinized, which ensures an attitude change that is more profound. However, most of the studies researching trustworthiness found a positive relationship between trustworthiness and persuasion effectiveness (Pornpitakpan, 2004).

Therefore the conclusion can be drawn that the three constituents of source credibility consisting of expertise, trustworthiness and physical attractiveness can influence persuasion effectiveness of an advertisement positively (Ohanian, 1990; Pornpitakpan, 2004; Chu & Kamal, 2008; Hsu et al., 2013;). The next chapter will elaborate on the importance of the three constituents of source credibility in the context of social media.

2.4.2 Source Credibility on Social Media

Regarding social media, the influence of source credibility on persuasion effectiveness is commonly studied. These studies indicate the influence a credible source can exert on attitude and purchase intentions when recommending products (Chu & Kamal, 2008; Hsu et al., 2013).

In studies with regard to persuasion on social media applications, trust and expertise are most commonly included in measuring source credibility (Brown et al., 2007; Chu & Kamal, 2008; Chu & Kim, 2011; Wu & Wang, 2011; Hsu et al., 2013; Spence et al., 2013). In assessing online source credibility, physical attractiveness has not been included since eWOM does not contain visual images of the communicator, causing physical attractiveness

irrelevance (Chu & Kamal, 2008).

Several studies researched the effects of perceived source credibility on persuasion effectiveness, including brand attitude and purchase intention (Chu & Kamal, 2008; Wu & Wang, 2011; Hsu et al., 2013). eWOM through bloggers’ product recommendation can

(33)

influence intentions to online shopping positively through a trusting effect (Hsu et al., 2013). Perceived trustworthiness of the blogger has a significant influence on the perceived

usefulness of the recommendation. This perceived usefulness, in turn, has a significant effect on attitude and online purchase intentions (Hsu et al., 2013). Chu and Kamal (2008) confirm the importance of perceived trustworthiness of bloggers. In combination with argument quality, the perception of a credible source, which entails trustworthiness and expertise, has a greater impact on brand attitude compared to lower credibility perceptions (Chu & Kamal, 2008). Another research regarding product involvement, confirmed the positive influence of source credibility on brand attitudes in eWOM on social networking sites (Wu & Wang, 2011). In this study trustworthiness and expertise were included as the determinants of source credibility as well (Wu & Wang, 2011).

Thus in literature, trustworthiness and expertise are most commonly used as

constituents for measuring the influence of source credibility in the context of social media, including blogs and social networking sites (Brown et al., 2007; Chu & Kamal, 2008; Chu & Kim, 2011; Wu & Wang, 2011; Hsu et al., 2013; Spence et al., 2013). Previous studies about source credibility on social media, confirm the positive influence source credibility with perceived expertise and trustworthiness as constituents, has on persuasion effectiveness regarding attitudinal changes (Chu & Kamal, 2008; Wu & Wang, 2011; Hsu et al., 2013).

Several studies, concerning source credibility in endorsement and advertising literature, including celebrity endorsers, take physical attractiveness in consideration

(Chaiken, 1979; Baker & Churchill, 1977; 1977; Kahle & Homer, 1985; Patzer, 1983; Amos et al., 2008; Kapitan & Silvera; 2016). These studies confirmed the importance of physical attractiveness as a constituent of source credibility to cause changes in attitude and purchase intentions. Till date, physical attractiveness has not been included as a constituent in assessing perceived source credibility on social media. Since Instagram is a visually oriented medium,

(34)

as it mainly consists of pictures, in this study physical attractiveness will be included as a constituent of source credibility. Therefore, the influence physical attractiveness exerts on perceived source credibility will be examined in this study.

In assessing source credibility, it is important to recognize that the three constituents will influence each other (Pornpitakpan, 2004). Perceived physical attractiveness influences the whole credibility perception of the source and therefore the other constituents of source credibility (Maddux & Rogers, 1980; Joseph, 1982). Thus, the positive influence of physical attractiveness on the credibility perception of the digital influencer could partly be explained by the influence of physical attractiveness on the other constituents.

Based on previously discussed literature this study expects a direct relationship between physical attractiveness and perceived source credibility, in such a way that an attractive digital influencer will be perceived more credible compared to an unattractive digital influencer.

H2 Physical attractiveness of the digital influencer has a positive influence on perceived credibility

Also, based on previous research the conclusion can be drawn that physical attractiveness will influence persuasion effectiveness (Chaiken, 1979; Baker & Churchill, 1977; 1977; Kahle & Homer, 1985; Patzer, 1983). This would indicate that a direct relationship between physical attractiveness and brand attitude is expected. The following hypothesis has been composed.

(35)

H3 Physical attractiveness of the digital influencer has a positive influence on brand attitude

The other constituents, expertise and trustworthiness, of perceived source credibility, have been widely studied in the context of blogs. These studies showed that expertise and trustworthiness influence source credibility perceptions; higher perceptions of expertise and trustworthiness resulted in higher perceptions of credibility (Brown et al., 2007; Chu &

Kamal, 2008; Chu & Kim, 2011; Wu & Wang, 2011; Hsu et al., 2013; Spence et al., 2013). In the context of traditional media and social media research showed that credibility perceptions of the source has a positive influence on persuasion effectiveness (Ohanian, 1990;

Pornpitakpan, 2004; Chu & Kamal, 2008; Wu & Wang, 2011; Hsu et al., 2013). Therefore the expectation is, that persuasion will be more effective when the digital influencer is perceived as credible.

H4 Perceived source credibility of the digital influencer has a positive influence on brand attitude

2.4.3 Source Credibility and Sponsorship Disclosure

Studies researching the effects of sponsorship disclosure on blogs commonly include source credibility as a variable (Carr & Hayes, 2014; Colliander & Erlandson, 2015; Liljander et al., 2015). Disclosing a sponsorship influences the credibility perception of the source as the sponsored content could be viewed as no longer subjective (Carr & Hayes, 2014). While the subjectivity feature of a digital influencer is exactly what makes the digital influencer so influential (Uzunoğlu & Kip, 2014; Woods, 2016). Therefore, the effects of sponsorship

(36)

disclosure and the influence on source credibility are often combined in studies concerning blogs and social media (Carr & Hayes, 2014).

Research about the effects of disclosing a sponsorship on source credibility suggests mixed results (Campbell et al., 2012; Carr & Hayes, 2014; Colliander & Erlandson, 2015; Liljander et al., 2015; Hwang & Jeong, 2016). Liljander and colleagues (2015) found that disclosing the sponsorship influenced purchase intentions and the intention to revisit the blog negatively. Not mentioning the sponsorship, did not affect purchase intentions and intention to revisit. For both types of sponsorship, disclosure and no disclosure, no influence on perceived source credibility was found (Liljander et al., 2015). A possible explanation could be that source credibility remained unaffected because disclosure only affects the perceived trust of the blogger, not the other constituents (Liljander et al., 2015).

A study differentiating on timing of the sponsorship disclosure found similar results concerning persuasion effectiveness in their study (Campbell et al., 2012). Not mentioning the sponsorship and disclosing the sponsorship before placing the advertisement did not influence brand attitude. However, mentioning the sponsorship after the placement of the advertisement affected brand attitude negatively (Campbell et al., 2012). Concerning both disclosure types, brand recall was lower compared to no disclosure (Campbell et al., 2012).

Contradictory results can be found in the study of Hwang and Jeong (2016). In this study, a distinction has been made in the manner of disclosing the sponsorship. In general, disclosing a sponsorship negatively affected source credibility. Although, when mentioned that the opinions were honest together with the sponsorship, the negative effect disappeared. Brand attitude and behavioural intentions were not affected by both types of sponsorship disclosure (Hwang & Jeong, 2016). This means that mentioning the sponsorship will

influence source credibility but not persuasion effectiveness, which contradicts the results of the study previously mentioned.

(37)

Another study on sponsorship disclosure showed that disclosing a sponsorship did influence the attitude towards the blog and credibility of the blog negatively (Colliander & Erlandson, 2015). In this study disclosing the sponsorship did not influence brand attitude and purchase intentions, in line with the results of previously mentioned study.

The research of Carr and Hayes (2014) is the only study evaluating the influence of source credibility on persuasion effectiveness, regarding WOM influence. The studies

previously described assessed the influence of disclosure on source credibility and persuasion effectiveness separately (Campbell et al., 2012; Colliander & Erlandson, 2015; Liljander et al., 2015; Hwang & Jeong, 2016). In assessing the effects of sponsorship disclosure on perceived blogger credibility Carr and Hayes (2014) make a distinction between explicitly mentioning the absence of a sponsorship, explicitly disclosing, implicitly disclosing and not disclosing a sponsorship. Consistent with the results of the study of Liljander and colleagues (2015) no difference was found in perceived source credibility between explicitly and not mentioning the sponsorship. With explicitly mentioning involvement of a sponsorship the source was perceived less credible and with implicit disclosure least credible (Carr & Hayes, 2014). With regard to the influence of the source, they showed that the less credible the source is perceived, the less influence the source exerts via eWOM. Accordingly, the influence a digital influencer exerts through eWOM affects persuasion effectiveness. Summarizing the results, the majority of the studies discussed found a negative

relationship between sponsorship disclosure and credibility perceptions (Carr & Hayes, 2014; Colliander & Erlandson, 2015; Hwang & Jeong, 2016). Furthermore, disclosure influences the subjectivity perception of the source negatively (Carr & Hayes, 2014). Since subjectivity is the most important feature of a digital influencer’s ability to exert influence, disclosing the sponsorship of a sponsored content by digital influencers may affect persuasion effectiveness (Uzunoğlu & Kip, 2014; Woods, 2016). Therefore, the following hypothesis is presented.

(38)

H5 Sponsorship disclosure has a negative influence on perceived credibility of the digital influencer

Literature previously mentioned, studied the effects of disclosing a sponsorship on perceived credibility and persuasion effectiveness separately (Campbell et al., 2012;

Colliander & Erlandson, 2015; Liljander et al., 2015; Hwang & Jeong, 2016). The decrease in persuasion effectiveness as a result of disclosing a sponsorship might be a consequence of perceived source credibility since disclosure of a sponsorship affects perceived credibility. Thus, the negative influence of disclosing a sponsorship of sponsored content by a digital influencer on persuasion effectiveness of an advertisement can be partially explained by perceived source credibility.

H6 Perceived source credibility will mediate the negative relation between sponsorship disclosure and brand attitude

2.5 Identification Theory

In this section identification theory will be described first. Thereafter, the inclusion of identification theory in this study will be explained. Followed by a short review of the literature regarding the effects of identification with celebrities. Finally, a link will be established between sponsorship disclosure, physical attractiveness, source credibility, persuasion effectiveness and identification.

Identification has been widely studied, but these studies have experienced difficulties in conceptualizing identification. Therefore, a comprehensive conceptualization of

(39)

definition is the different applications of the theory in a variety of contexts (Basil, 1996; Cohen, 2001). As literature on identification with digital influencers and a comprehensive conceptualization of identification theory is still lacking, this review will focus on studies concerning identification with celebrities.

Theoretical foundations of identification with celebrities can be found in social identity theory (Jin & Phua, 2014). According to social identity theory people identify themselves with others and groups to derive self-concept in an effort to make sense of their surroundings (Tajfel, 1982). As social identity theory focuses predominately on groups, Kelman’s (1961) theory of opinion change is more suitable regarding this study due to the fact that this theory has often been applied as theory of identification in advertising and mass communication as well as its focus on individuals (Basil, 1996; Jin & Phua, 2014). Therefore, this study will follow the theory of identification of Kelman (1961).

According to Kelman, identification occurs when ‘individuals accept influence from another person or group in order to establish or maintain satisfying self-defining relationship to the other’ (2006, p. 3). Persuasion occurs as individuals try, to be like this other person, in order to maintain their relationship (Kelman, 1961). To maintain this relationship individuals adopt similar attitudes and behavior from this other person (Kelman, 1961). Thus, as

individuals adopt attitudes and behavior, persuasion can occur through identification. In the case of digital influencers, individuals who identify themselves with digital influencers may adopt their attitudes and opinions because they desire to be as much as possible like the digital influencer.

Since studies on identification with digital influencers are still lacking, the effects of identification with celebrities will be discussed. Studies elaborating on Kelman’s (1961) identification theory confirm the ability of celebrities to exert influence on individuals’ attitude and behaviour (Basil, 1996; Fraser & Brown, 2002; Jin & Phua, 2014). Basil (1996)

(40)

found that identification with the celebrity explained celebrity effects, in terms of adopting attitude and behaviours of a celebrity. The more an individual identifies himself or herself with the celebrity, the more the individual changed his or her attitude and behaviour

according to the behaviour of the specific celebrity (Basil, 1996). Fraser and Brown (2002) extend Kelman’s (1961) identification theory to a mediated world. They concluded that individuals develop strong identification by continuously mimicking the value and lifestyle of the celebrity in order to be the same (Fraser & Brown, 2002). A study, combining

identification and purchase intention found that identification with a celebrity explained the relationship between a celebrity type and purchase intention (Jin & Phua, 2014). This indicates the influence of celebrity identification in endorsed product advertisements (Jin & Phua, 2014).

Identification theory is included in this study due to the ability of digital influencers to exert influence. Digital influencers have been defined as individuals that exert influence on the decisions of others by sharing their opinions online, including social media. Identification with these digital influencers might explain or strengthen this influence, which is the case with celebrities (Basil, 1996; Fraser & Brown, 2002; Jin & Phua, 2014). Since this study aims to measure persuasion effectiveness of digital influencers, in terms of brand attitude and purchase intention, identification theory is included as it has the possibility to influence persuasion effectiveness. Therefore, the following hypothesis has been composed.

(41)

Also, individuals have the tendency to trust other individuals whom they

perceive to be comparable to themselves (McCroskey et al., 1975). Also, Colliander and Erlansson (2015) found that para-social interaction with a blogger influences the perceived credibility of the blog. Para-social interaction entails the illusion of a friendship with the blogger. This illusion affects the perceived credibility in such a way that with a higher level of para-social interaction the credibility perceptions will increase. Therefore, there is expected that the more an individual identifies with the source, the more credible the source is perceived.

H8 Identification with the digital influencers has a positive influence on perceived source credibility of the digital influencer

Regarding sponsorship disclosure, one could expect that the disclosure of a

sponsorship by a digital influencer on Instagram will have a negative influence on persuasion effectiveness. According to Kelman’s (1961) identification theory, individuals adopt attitudes and behaviour from others with whom they identify themselves, in order to maintain the relationship. In the case of celebrity endorsements, individuals adopt attitudes and behaviours of celebrities they identify themselves with (Basil, 1996; Fraser & Brown, 2002; Jin & Phua, 2014). Celebrities are often perceived as less related compared to digital influencers, as celebrities are perceived as higher-class society (Woods, 2016). Therefore, the effects of identification may be even stronger with respect to celebrities. In this study, there is expected that individuals who identify with a digital influencer will accept the influence from these digital influencers, due to the adoption of attitudes and opinions, similar to celebrity identification. As a consequence identification with a digital influencer will decrease the negative influence of disclosing a sponsorship of sponsored content by a digital influencer on

Persuasion effectiveness of sponsored content by digital influencers : the influence of physical attractiveness and identification on source credibility as clarification for the effects sponsorship of disclosure