Amsterdam Business School
MSc. in Business Administration – Digital Business Track
MASTERS THESIS
The rise of social influencers and the role they play in brands’ digital marketing
campaigns. Do consumers trust influencer’s opinion about endorsed products posted on their
social media channels and has this a direct impact on consumers’ willingness to buy them?
Valeriya Georgieva Popova Student number: 11386932
23rd June 2017 Final Version
Table of contents
Abstract……….3
1.Introduction………...4
1.1 Social Media Phenomenon………...4
1.2 Problem statement………5
1.3 Research Question………....6
2. Thesis structure………..…..7
3. Theoretical Framework………...8
3.1 Web 2.0 and social media………9
3.2 WOM shifting to eWOM………...10
3.3 Who are the digital influencers……….….12
3.4 Hypothesis formation……….16 4. Conceptual framework………..19 5. Research Methodology……….…………..19 5.1 Choice of method………...20 5.2 Sample………21 5.3 Data Analysis………...…..22 5.4 Pre-test……….………...23 5.5 Respondents………...23 6. Research Results……….………24 6.1Factor Analysis………24
6.2 The KMO and Bartlett’s Test……….25
6.3 Multiple Linear Regression Analysis……….…26
6.4 ANOVA output……….….28
7. Discussion and Conclusion………....33
7.1 Conclusion………..33
7.2 Managerial Implications……….…34
7.3 Limitations and future research………..…35
Reference……….37
Appendix 1………..41
Appendix 2………..42
List of figures………...47
Abstract
Living in the 21st century - which is labeled as the century of information by modern society, the business world has stumbled on the buzzing term “influencer marketing”, yet what does it describe and how does it work?
The term influencer marketing is a relatively new form of covert marketing which allows to avoid both consumer’s disinterest and skepticism, both of which were considered to be linked messages in traditional marketing from the past few years. The paramount growth of social media platforms led to brands embracing their influencers, who represent a new type of third party link between companies and consumers. Via their social media channels digital influencers are able to shape consumers’ attitude. This form of influencer marketing consists of companies selecting and identifying themselves with influential individuals who are able to reach a targeted online community worldwide and get in return different types of sponsorship rewards for promoting the brand in creative and authentic way. This shifts the mind of consumers away from traditional forms of advertising and opens doors for researchers to explore a new field of marketing channels. In this research, quantitative research methods were used in order to examine and understand respondents’ opinion on trustworthiness and willingness to buy a specific brand promoted by a digital influencer. Furthermore, it hypothesis
Statement of originality
This document is written by Student Valeriya Popova 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 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.
that multiple publications from those endorsers in a short period could lead to a negative impact on consumers’ willingness to buy. The findings demonstrated that the reliable dimension of trust in influencers opinion has a positive effect on willingness to buy while the multiple publications are mediated by trust in opinion of the influencers and in the contrary, have negative effect on consumers’ willingness to buy.
In the end, this research would contribute to the academic and practical sector in exploring the impact of social media platforms towards influencer marketing.
Key words: Digital Influencers, Influencer Marketing, Social Media, Trust, Willingness to buy
1. Introduction
“Influence is the ability to move a person to a desired action”
-Bob Burg
Over the past couple of years, consumers are more and more into control of how they interact with brands and what content is reaching them. Consumers’ trust in brand-driven advertising is declining. In comparison with advertising such as TV ads, company websites, online advertising and editorial content today’s generation is showing an increasing preference for the opinion of trusted individuals known as “influencers”. (Uzunoglu and Kip, 2014) This is a result of the constant technological advancement and the rise of social media as part of Web 2.0. (Kaplan and Haenlein, 2012) However, more than a decade in, companies are still struggling to come up with a branding model that works in the chaotic world of social media. More than ever, companies nowadays need to shift the focus of their digital efforts to the so called “crowdcultures”. (Holt, 2016) They are digitaly natives who are constantly sharing
know very well when they are being overwhelmed with advertising messages. Being more information savvy and well informed, they cannot be easily controlled or manipulated by organizations. (Holt, 2016) On top of that, they manage to avoid intrusive marketing messages. However, if they feel they are in control of the content seen by them, it is much easier to reach them. Trust is of main importance for them. This is challenging the rules of branding. Thus, today more and more brands have begun to align themselves with social media influencers. (Dimofte et al., 2016) This is as a result of customers expressing increasing preference for 'authentic' opinion over and above the voice of the brand itself. (The Rise of Influencers report “Fashion and Beauty Monitor's” in association with Econsultancy, 2016) A research by the same report has been done into how brands are approaching influencer marketing campagins and with what success. Currently:
“57% of survey respondents say they have an Influencer Marketing strategy in place; additional 21% plan to invest in it over the next 12 months”
This leads to the question: “What exactly is influencer marketing and who are the digital influencers?”
A digital or social influencer is someone whose knowledge and expertise has allowed him to build up a following around social media platforms. (Brown & Fiorella, 2013) Since nowadays we do not go online, but rather live online, it is much more feasible for companies to reach their customers via various social media platforms. What better way to do that, than collaborating with a trusted individual who has created a highly-respected following. However, in order to be considered as “an influencer” they need to combine trust and reach, since each by its own is either word of mouth or just merely advertising. (Claire Cobbledick, 2016) These individuals have the ability to influence the opinions or buying decisions of the brand’s targeted audience, largely in the interest of their social media following. Instead of looking at brands, as they did in the past, consumers now look at each other and at their favorite personalities.
This emerging form of influence is all about advocacy, review and engagement. (Keller and Fay, 2012) It also helps brands to become more authentic and be relevant to its customers. (Chaudhuri and Holbrook, 2002) Having a massive following across the blogosphere, on YouTube, Instagram, Pinterest, and other platforms they are creating huge opportunities for brands and their digital marketing strategies. (Miles, 2014) However, those opportunities come with some obstacles that currently managers are facing. With the advance of new digital channels influencer marketing is becoming more and more important in order to generate traffic and help drive leads. Based on a recent study “Influencer marketing has been rated as the fastest-growing online customer-acquisition channel, beating organic search, paid search and email marketing”. (Nielsen)
Establishing on what is influencer marketing (IM), I would like to elaborate on the challenges that marketing managers are facing today with incorporating IM to their digital marketing strategies. Touching on that subject clearly shows opportunities for a relevant research gap. First of all, being part of the digital strategies of brands, influencer marketing is a new addition to it and this is causing some problems for managers who are not aware how to deal with this innovation cause by the rise of web 2.0. (Berthon et al., 2012) Currently, “67% of brands find it hard to find relevant influencers” (“50 Influencer marketing statistics, quotes and facts,” 2017) and proof of the actual value (in numbers) of having an influencer marketing strategy as part of their digital marketing. Moreover, some executives are still misled and associate it with “celebrity endorsement”. (Silvera and Austad, 2003) However, influencer marketing is still evolving so brands and influencers have to find a way into how to work together and what content generates best results. This raises the question of my research topic:
Do consumers trust influencer’s opinion and knowledge posted on the social media channels and has this a direct impact on consumers’ willingness to buy the endorsed
There is a clear managerial need to investigate how brands are perceived by the customer based on their influencer marketing strategy and does this reflect their image amongst the targeted customers. Currently, more and more brands are investing in influencer marketing strategies and are creating long term contracts with influencers which is shifting the industry towards something that is more genuine, rather than straight forward business contract. This is marking the first step into being more transparent with customers. (Chaudhuri and Holbrook, 2001) With generally millennials turning to social media for tips, recommendations and advice - Instagram and YouTube are often the first channel they visit before making a purchase. Brands via the influence of bloggers are using those social medias to create digital content promoting awareness and driving engagement around new products. (Miles, 2014) The building block of IM is based on user-generated content. (Fournier & Avery, 2011) This content is provided by the brands’ users and it is seen as a more credible source of information rather that brand-generated content. (Van Noort & Willemsen, 2012; Bambauer-Sachse & Mangold, 2013; Bickart & Schindler, 2001) Furthermore, digital influencers help create credible electronic word-of-mouth. (Bambauer-Sachse & Mangold, 2013). This helps brands engage with their core consumers. A recent study concluded that “as an influencer’s Instagram following increases, the rate of engagement rapidly decreases”. (Markerly, 2015) That proves engaging with the correct influencer is a big challenge for every brand. Whilst taking into consideration the sphere of knowledge, the influencers following and level of engagement with the customers, brands should also consider the kind of relationship they would have like to build with them. In summary, a crucial for effective influencer marketing campaign is the perceived trustworthiness of the digital influencer. (Kapitan and Silvera, 2015)
2. Thesis structure
The structure of this thesis would include two main parts - Theoretical and Empirical research. The theoretical part includes a thorough literature review about who are the digital influencers and how companies are approaching them and improving their digital marketing strategies to make it relevant for its consumers. The main focus is how brands’ strategies that include digital influencers, reflect on brand image, trust and transparency. The theoretical part also reveals information about the phenomenon of web 2.0 - Social media. The empirical part would describe the research methodology, what kind of instruments were used, data collection and sampling. It finishes with conclusions, limitations, and recommendations for future researchers.
3. Theoretical framework
Nowadays, social media is not only an important part of our daily life, but it is playing a big role in the society. Looking into the creator of this phenomena - The Internet and Web 2.0, social media shifts the power from marketers to consumers. (Murphy and Schram, 2014) Customer empowerment is the vital role for success of brands’ marketing strategies which are substantially influenced by social media. (Rosenbaum, 2015) Based on the research of Berthon, Pitt, Plangger & Shapiro (2012) Web 2.0 technologies have caused three effects:
‘(1) a shift in locus of activity from the desktop to the Web’
‘(2) a shift in locus of value production from the firm to the consumer’ ‘(3) a shift in the locus of power away from the firm to the consumer.’
A brand can be defined as “a name, term, sign, symbol, or design, or combination of them which is intended to identify the goods and services of one seller or group of sellers, and to differentiate them from those of competitors”. (Kotler, 1991). Investigating more into the brand’s customer base and having a look into the most profitable segments, it is more and more
evident that brands have to start thinking differently regarding loyalty and the type of experience they would like to provide in order to help drive customer retention, advocacy as well as growth. Currently consumers tend to spend not only more money but dedicate their time to the brands they love. (Muniz and O’Guinn, 2001) In this digital age companies need to understand what influences the customer’s loyalty and perceived brand image. (Magnusson and Zdravkovic, 2011) Sticking with traditional strategies and trial and error could lead to draining profitability and risk of pushing customers away. Nowadays, consumers are tech-savvier and have a non-stop desire for personalized, extra-ordinary experiences that force brands to shift their approach and create new loyalty programs. By implementing new and improved strategies that are more related to the daily routine of consumers, brands create a “life-partner” rather than a person who is marking the relationship as “it’s complicated”.
In order to stay relevant and be competitive companies are incorporating more and more social media channels into their brand channels. (Booth, N. and Matic, 2011) This way customers feel as part of the brand and they could have an interaction with not only the brand but with others sharing their interests. (Muniz and O’Guinn, 2001) Social media is also becoming more and more important for companies. The time when customers were using the Internet only to read, watch and occasionally to buy products and services has past. Today most of the people use the Internet platforms to create and not only, but also modify, share and discuss that content. (Kietzmann et al., 2011) Currently, there are a lot of social media platforms and more new ones are created daily. They each vary in functions, as well as target audience, but the similarity they all have is that they create an environment for not only creating, but maintaining social relationships and sharing relevant content in the form of posts, comments, likes and common interests. (Kim and Han, 2009) One of the reasons that social media is taking so much time of our daily routines is that it creates a real-time, easy accessible environment in which the information spreads much faster. (Tuinsma, 2015) This also results
into a place where connected individuals can influence others much faster rather than offline communication. (Subramani and Rajagopalan, 2003) Online influence is gained by sharing more relevant content and interacting more with social media contacts. (Labrecque, 2014) This increased influence by the social media results into new marketing strategies that brands adopt into their marketing plan. Today “influencer marketing” has various benefits for brands. According to Murphy & Schram (2014), they categorize this strategy as a “unique branded social content from a trusted, influential source”. Thus, it has the ability to target specific customer segments and results into higher click-through rates (CTRs) and lower cost per clicks than traditional display. (Burke, 2016) Usually, influencers create new and fresh content with a new perspective that creates loyal communities, which not only care but share that information. (Muniz and O’Guinn, 2001) Whilst social media acts as a platform where companies can build relationships with customers, influencer marketing helps them to spread their brand-related content via influential and trusted people. Influencers also can shape their followers’ perception which could be both positive and negative due to the lack of control over social media. (Hanna, Rohm and Crittenden, 2011)
Furthermore, numerous studies show that WOM is another powerful tool which should be used accordingly. A McKinsey Study found that using WOM to its full potential could generate twice the sales of paid advertising and has 37% higher retention rates. Moreover, as Weiss (2014) states “WOM is the most trusted source of consumer information and the likeliest to be acted upon”. Looking into the same study, human–to-human contact is more effective than traditional marketing. Thus, with the rise of numerous social media platforms the effectiveness of WOM is shifted to e-WOM, which refers to any statement made by a potential, actual or former customer about a product or company which could be both positive or negative and is available to multiple groups via the Internet. (Hennig-Thurau et al, 2014)
Looking into statistical evidence in a survey that ComScore has conducted “84 percent of European Internet users belong to at least one social network.” This highlights the importance of incorporating social media into companies’ brand channels. The internet - something that a couple of years has been seen as a luxury has become mainstream in everyday transaction and communications. The following could be related to “influencer marketing”, it is becoming a widely-discussed topic. The millennial demographic of today is more and more turning to peer recommendations which is amplified by social media platforms. Furthermore, digital influencers are able to change behavior and cause effect and often are misinterpreted with reach and popularity. (Linkedin: WOMMA influencer guidebook 2013) Brands include digital influencers into their new strategies mainly because according to Nielsen, 90% trust peer recommendations and only 32% trust advertising, so influencers are seen as trendsetters and people often go to their platforms to get inspired. Usually, content creators write and post about one consistent subject which makes it easier for a company to target that niche segment. That would be a perfect opportunity for brands to generate more traffic to their website. Therefore, creating an effective way to expand company’s reach and get more exposure and receptive audience. However, regarding the reach of the influencer, authors of different researches have different opinions in whether a higher following lead to greater exposure or au contraire the more niche the influencer the more personal is the relationship they have created with their following group. (Business Insider)
To sum it up, influencers:
1. Attract your target audience
2. Create brand awareness and exposure 3. Are perceived as trustworthy experts
A main support of this shifting strategy is the survey conducted by Nielsen showing that “23% feel loyal to organizations that partner with social influencers, such as bloggers and vloggers. 42% are loyal to brands that their family and friends do business with.” On top of the three main reasons mentioned above, eMarketer found that companies use social influencers to tackle the rising concerns about ad blocking and ad avoidance. The statistics done by Omnicom Media Group’s July 2016 report show that there are “69.8 million ad blocker users in 2016, which is a rise of 34.4% from 2015.” (based on American users)
This new emerging influencer community is shifting a significant power over the perceptions of brands and companies. There a shift in the traditional one-way communication nowadays, it is transitioned to multi-dimensional involving peer-to-peer communication. (Berthon et al., 2008) This could be explained by the rapid expansion of social media over the past years. Social media channels are influencers main ways of communicating with their followers. As Booth and Matic (2011) have stated in their research on leveraging influencers in social media to shape corporate brand perceptions: “The “nobodies” of the past are now the new “somebodies” demanding the attention of communication professionals who seek continuous engagement with targeted consumers throughout the var ious channels of the social web”. Following on that research the impact on brand equity from consumers is higher than ever before. In order to stay relevant brands, need to adapt their corporate marketing to be able to control an environment that is out of their hands. The authors argue that the ownership of the brand has never been in the company’s hands, but rather has been controlled by the consumer. On the other hand, once consumers believe in your brand they would become your greatest advocate. As an example of that we can have a look at Apple, who for its customers nowadays cannot do any wrong, nevertheless, they might have some issues with its products.
In Malcolm Gladwell’s book “The Tipping Point” influencers are described as people who have a lot of knowledge about a topic. Currently, when a brand identifies influencers, they
often look at only numbers (following). However, influence is not just having a lot of followers. Most importantly, it is expertise and credibility on subject matter that builds the relationship between the influencer and his or her followers. Researchers have moved from the connection of only brand-consumer to the triad of consumer-brand-consumer. (Muniz and O’Guinn, 2001) However, a loyal customer base is built by creating a trustworthy relationship. Three of the most important key messages that customers are looking for are: relevancy, originality and genuine message. (Akbar and Parvez, 2009) These are key points that would result into customers using more of brand’s products and services as well as recommending it to others and acting as brand’s advocates. Additionally, consumers form “brand communities” that directly affect brand associations and brand awareness, tend to strengthen brand loyalty as well. Members of those communities contribute with more behaviors rather than mere repurchases and include other consumers too. (Muniz and O’Guinn, 2001) Thereby, creating a strong brand community could be a critical step in building robust relationship marketing. (James, 2001) Socially embedded and entrenched loyalty, brand commitment, hyper loyalty are some of the advantages they are granting to companies. (Keller, 2001). Having influencers connect with that community could be an added value for the goals of the marketing campaigns the firm is working on. However, there other factors that incite brand performance amongst consumers. Being relevant, original and genuine helps the brand build a more authentic image amongst the customers. Even though, that there is not yet a characterized definition for brand authenticity in the marketing literature, today “authenticity has overtaken quality as the prevailing purchasing criterion, just as quality overtook cost, and as cost overtook availability”. (Gilmore and Pine, 2007) From the consumers’ perspective, brand authenticity has been explained as a combination of different associations. Some of them include quality commitment, sincerity, heritage (Napoli, Dickinson, Beverland, and Farrelly, 2014), originality, reliability, naturalness (Bruhn et al., 2012), individuality, consistency (Schallehn, Burmann and Riley, 2014),
integrity, credibility, symbolism (Morhart, Malär, Guèvremont, Girardin and Grohman, 2014). Furthermore, every customer could decide on their own whether a brand is authentic or not. In his research, Grant (1999) has clearly declared that “authenticity is the benchmark, against which all brands are now judged”. Moreover, a brand is seen to be authentic when diverse stakeholder groups truly experience what they are promised (Fisher-Buttinger and Vallaster, 2008). A brand needs to be seen as real and honest, which would have to be act upon with genuine and transparent campaigns. This would result into creating loyal customers and creating awareness amongst them. (Keller, 2001) When consumers have a certain mind-set about a brand, about both brand recognition and brand recall, they can reward it with paramount profits. (Keller, 2001) According to Keller once again, those profits could be increased value sales, consumers themselves recommend the brand to others resulting into reduced budged for promotional campaigns and choosing the brand over the competitors, resulting into higher volume sales. There is often a relation that researchers make between influencer marketing and brand placement. Both being cost-effective, targeted to a specific group, based on the content and the message that is shared. (Booth and Matic, 2011) Previous studies about social media marketing have for example showed how brand related user-generated content differs between Facebook, Twitter and YouTube (Smith, Fischer and Yongjian, 2012) and that there are different outcomes of consumer’s action for the different marketing channels. (Weinberg and Pehlivan, 2011) However, there is little knowledge about how nowadays brands interact with influencers when incorporating them into their digital marketing campaigns. Since, more and more brands are using bloggers to communicate with their target customers, from a managerial point of view this research would be beneficial in order to examine how to deal with trust, transparency and brand authenticity between the sponsored content (interaction) and the customer of the brand. The term ‘sponsored content’ is relatively new and consists of hidden advertising in which the commercial message is embedded in an editorial content. This is done
on social media platforms such as Blogs, Youtube and Instagram (Kietzmann et al., 2011). In a study by Nielsen around 71.2% of the influencers stated that their audience is most happy when they are “honest, funny, open, willing to call it like I see it”. (Appendix 1) Regarding measuring success of a publication, influencers believe traffic to their platforms is the most important, however marketers state that engagement with content is how they measure return on investment. Even though both creators and marketers consider authenticity to be the building block for a successful campaign, usually brands would like to have more control over the message and approach influencers for sponsored content. Furthermore, trust is considered to be one of the most important variables relating to creating a bond between the customer and the brand. (Murphy and Schram, 2014)A study by Kapitan and Silvera (2015) has found that many consumers consider digital influencers more trustworthy than celebrities and additionally 84% of millennials “report that user generated content created by the digital influencers, influence their purchasing decisions” (p. 12). Accordingly, this is as a result of the Kelman’s (1961) internalization process. This phenomenon can be described as when consumers are persuaded by an endorsed message and gradually get persuaded to adopt the endorser’s beliefs. (Kapitan & Silvera, 2015) This is a two-step process consisting of consumers identifying with the digital influencer (source) using the mechanisms of “familiarity or attractiveness” and then “weigh an endorser’s authenticity and adopt the message as if it were their own”. This research has been conducted by the same authors. Furthermore, to explore the sponsored content on social media, research has shown that digital influencers share is much more powerful than the content shown in advertisements (Hall, 2010; Woods, 2016) Social media empowers communication among consumers and create a source for eWOM which is more likely to be free from manipulation compared to advertisements. (Uzunoğlu & Kip, 2014) This way the content created under the form of endorsement is more effective due to the “perceptions of high source credibility”. Key inputs for consumers are expertise, honesty and trustworthiness.
(Kapitan & Silvera, 2016) Same research shown that positive feeling towards the endorsement and the influencer are created by a content that consumers perceive as believable and credible, which on its hand deems the endorser authentic. Ohanian (1990) summarizes trustworthiness as “the listener's degree of confidence in, and level of acceptance of, the speaker and the message” (p.41). Moreover, it is perceived as “when someone is consistently honest or truthful”. Important to acknowledge is that there must be “continual consistency” in influencer’s opinions and showings. (Scott, 2010) In order to be perceived as trustworthy, the sender of the message has to be accepted by the receiver, this way it results into believable endorsements and therefore digital influencers are only influencing those who accept and trust them. To summarize and state the conclusions that lead to the first hypothesis of this research, an essential aspect to endorsement effectiveness is the perceived trustworthiness of the digital influencer. This has direct effect on consumers’ intention to buy and the following hypothesis is drawn, in order to answer the research question stated in beginning:
H1: Trust in influencer’s opinion would have a positive effect on consumers’ willingness to buy.
In the stated hypothesis 1 willingness to buy is evaluated as dependent variable that is directly affected by trust as the independent variable. Researchers define willingness to buy as “likelihood that the buyer intends to purchase the product”. (Dodds et al. 1991) The main goal of every marketing campaign is to facilitate a strong bond between the consumer and the brand as well as persuade consumers to make the first-time purchase. (Hiscock 2001) Usually, of paramount importance is the relationship of trust between them. This shapes consumer’s buying behavior and customer loyalty, and can then leads to repeat purchase. Building trust and creating relationships between the online influencer and their followers is a very important process. Often, online influencers are perceived as a ‘real friend’ because of these strong relationships with their audiences. (Labrecque, 2014) Additionally, the increased usage of
social media and the tools that social media platforms have is allowing followers to see the influencer’s daily life via pictures and video stories. This nowadays considered social interaction, leads to consumers being more likely to be influenced by the received message and believe that the influencer’s opinion is a more reliable and trusted source, rather than the brands’ advertisement. (Hwang & Jeong, 2016)
Nowadays there are multiple platforms that brands and influencers could join efforts into creating content for their consumers. As one of the fastest growing mobile photo-sharing platforms and the main focus of marketers – Instagram, for example allows companies to reach consumers based on their interests. (Miles, 2014) This is resulting into the constant search performed by brands in order to find the right influencer who can promote their product via those trusted relationships and actively create interesting content for them in order to promote products, generate positive influence and brand awareness. (Dimofte et al., 2016) Furthermore, people believed they are engaged in a direct two-way conversation which feels like the performer, in this case the digital influencer is talking directly to them rather than the traditional media environments such as radio or television. (Rubin, Perse & Powell, 1985) This all is due to the improvements of new technologies and the creation of the Internet- making online media, especially blogs and social media platforms, a place for direct conversation. (Labreque, 2014) This direct communication strengthens online engagement and builds social relationship by giving the opportunity to gain insights into the daily life of influencers once again through their personal pictures and video stories. (Colliander & Dahlén, 2011) Users engaging with the content created by the influencer builds the relationship even stronger. Thus, making it a point of interest for further research of the outcomes when influencers post sponsored content multiple times per day about the same topic and leads to hypothesize:
H2: Multiple publications in short time period would have negative effect on consumers’ willingness to buy which would be mediated by trust in influencers
opinion/knowledge
As stated above both marketers and influencers believe that being authentic is the factor that drives their content. That would lead us to establish direct connection to how posting several similar in content publications could possibly jeopardize whether consumers’ beliefs that the influencer is sharing their genuine thoughts, creating negative trust in both the brand and the content creator. According to a study by Kang and Hustvedt (2013) transparency has strong effects on both trust and general attitude. The results also show that transparency indirectly affects their intentions to purchase from the brand and spread positive WOM about the corporation. As “trust” is explained by Chaudhuri and Holbrook (2001) it is consumer’s belief that brands will keep their promises and act on its consumers’ interests. Based on trustworthy relationships between customer and brands, this would have a significant impact on loyalty, retention, purchase intention, willingness to buy and act, and overall market performance. (Erdem and Swait, 2004) For this reason, companies should be focusing building strong relationships with their consumers (Knowles, 2003). Despite extensive studies on brand trust, there is little investigated into creating digital marketing strategies with influencers that would build trust and result into higher purchase levels.
The following two hypothesis would lead us to the current conceptual model created to get a visual representation of what would be further tested into my thesis research. According to Keller’s model (1993) a customer has to be primarily exposed to the brand in order to gain brand awareness and further identify its elements. After this people form attitudes about the brand and create their perception of its image. The statement above would be relevant to digital influencers as well.
4. Conceptual Framework
Primary conceptual model would be drawn in the figure below to show the hypotheses stated in the previous chapter. This aims to facilitate the readers understement visually and present a clear table that would serve as a guidance for the relevant research that would be conducted next.
Figure 1
5. Research Methodology
The structure of the thesis research starts from more general, but goes to more specific. In the beginning, I have examined theories of different authors about the phenomenon of social media which further in the analysis have linked to the new concept of digital marketing and companies’ strategies of collaborating with social media influencers to create content for their brands. Narrowing it down, my approach was to establish specific research questions. In order to be easier when analyzing the data, two hypotheses were developed and then tested. Following the extensive literature review I have conducted, the developed hypothesis would be essential for deriving the final results and finding out the relationships between the social media influencers, consumers and brands. Thus, for the purposes of the current thesis I would
derive to the proper hypothesis and test them using suitable statistical tools. A deductive approach would be done so that it would be of a great value to find correlations between different variables through statistical calculations. The methods used by the researchers differ according to the way data that is collected and then analyzed. In the quantitative methods, researchers usually use techniques for collecting and analyzing data which at the end generate numerical data i.e. convert statistical information into descriptive information.
5.1 Choice of method
The research strategy that is following in my thesis is survey. I have chosen this method as it is one of the most popular research strategies among social and business researchers. One of the reasons is that it can present answers to different type of questions such as what, who, where and how. On the other hand, the survey method gives us the possibility of gathering enough reliable primary data. I am interested in gathering original data and conducting my own survey which would allow me to collect large number of data at a relatively low price. (Saunders et.al, 2009) I am following the most common used technique of a structured questionnaire. As previous researchers have established, questionnaires are set of written questions on a particular topic where the opinion of group or groups of people is explored. (Sommer & Sommer, 1997) With this method I will collect information about facts, opinions, activities, level of knowledge or simply attitudes. In essence, the questionnaire would enable me to transform my research to research questions that the data would provide answers to. Each of the structured hypothesis include one dependent and one independent variable - the relation between them also called cause. Dependent variables are those that the researcher is trying to explain and independent are those that cause the change or explain the dependent ones. (Auriat, 2005)
5.2 Sample
For the purposes of this research, multiple sampling methods were applied. First of all, the
purposive sample was used. Based on the objectives of the study it is a non-probability sample also known as “judgmental, selective, or subjective sampling”. (Walliman, 2006) Moreover, this method was chosen since it can be applied when you need to reach a targeted sample quickly. In this situation sampling for proportionality is not going to be vital for the results of the survey. However, this sampling method does not ensure representativeness, researchers state that still useful information can be supplied. (De Vaus, 1996) The second sample method that was applied was the snowball. It refers to a technique where the next participants are reached through other respondents (Babbie, 2011) by forwarding it to relatives and friends (Zhou & Sloan, 2011). This method is suitable, although, it has several disadvantages, such as not enabling to gain a representative sample, however it presents the opportunity to reach a large number of participants within a short amount of time (Möhring & Schlütz, 2010). The last method I applied is referred to reaching those persons as participants who are instantly accessible - convenience sampling. (Walliman, 2006) Again, it generates large number of participants, thus applicable in this study. Since the survey was conducted online, the respondents were also recruited online. All of them have received a link of the webpage and a short text asking them for their collaboration, following with a further request to forward the link to their networks. This approach has presented the opportunity to gain access to a population that is beyond my own personal peer group. However, it was difficult to trace if it has actually generated significant number of respondents beyond my network group.
5.3 Data Analysis
For the purpose of testing the hypothesis drawn in chapter three, the type of software used is IBM SPSS Statistics 24. They would allow me to perform different types of analysis, data
transformation and would present valuable output for the purpose of answering the research question. (Arkkelin, 2014) Three types of analysis are performed to test both hypothesis. First, factor analysis was applied. It is one of the most widely used exploratory tool that reduce the dimensionality of multivariate data. (Bartholomew 1980) Researchers perform factor analysis in order to explore a content area, structure a domain, map unknown concepts, classify or reduce data, illuminate causal nexuses, screen or transform data, define relationships, test hypotheses, formulate theories, control variables. The second step of analyzing the dataset is to perform a regression analysis with which I would sort out those variables that indeed have an impact and it would provide answers to questions such as: “Which factors matter most? Which can we ignore? How do those factors interact with each other? And, perhaps most importantly, how certain are we about all of these factors?” (HBR article regression analysis) Defining since the beginning of the hypothesis the dependent variable — the main factor that I am trying to understand or predict and the independent variables — the factors supposable have an impact on your dependent variable would be a helpful step leading to the final part of the analysis. Lastly, as regression analysis, correlation analysis deal with relationships among variables as well. It indicates only how or to what extent variables are associated with each other. The correlation coefficient would measure only the degree of linear association between the two variables. Depending on the three analysis, the a priori postulated relationships among the variables can be supported or rejected. (Byrne, 2013; Raykov & Marcoulides, 2000)
5.4 Pre-test
The online survey questionnaire used in this thesis was conducted on Qualtrics. Before the actual survey was distributed amongst the respondents, the quality, functionality and grammar of the formulated questions was tested in a pretest (Möhring & Schlütz, 2010; Zhou & Sloan, 2009). For this purpose, three people received the link to the survey and provided thorough
feedback that gave valuable insights on the comprehension of the questions as well as they provided suggestions for improvement where needed. The data collection took place between 15 and 21 May 2017. In this time period, a total of 320 people participated in the survey. Of these participants, 284 finished the questionnaire, making the response rate high. At the end demographic data like age, gender, nationality and degree of education was measured as control variables.
5.5 Respondents
Of the 284 respondents that were included in the data analysis, 60,5 % are female (N=172) and 39.4 % male (N=112). Further, the respondents age range in the groups from first group being under 18 to last group 45-54 years old. However, most of the respondents are in the age group of 18-24 (N=181), following with 25-34 (N=95). Regarding the educational level of the participants, it can be noticed that they are rather highly educated. Only 1.06 % (N=3) do not have any kind of high school diploma. Most of the respondents, however, claimed to have master’s degree 41% (N=117) and further 38 % have already obtained a bachelor’s degree (N=108). In addition to this, three of the respondents have already finished a professional degree, one of which has a PhD. One interesting fact that can be noticed from the results of the survey is the multiple nationalities that participated in the survey, however, the predominant group are Bulgarians, which would be further elaborated in the limitations and recommendation part of this thesis.
6. Research Results
6.1 Factor Analysis
In order to identify the latent variables, I have performed factor analysis as a first step of the analysis of the results. Though it has multiple uses, I would be applying it to simplify the dataset I have collected, meaning, reducing the number of variables within the regression model. Correlation of the variables is reported using the correlation matrix. The same was used to test for any presence of multicollinearity. The recommended variance is 0.5 plus variance, meaning that I used the communality table to drop items that account for less than that value. The variable Influencer’s Opinion was developed from 4 items defined by questions: Q29, Q33, Q17 and Q22. These questions have been used to rate the impact of influencer opinion on the willingness to buy. To reduce dimensionality, we have done a factor analysis on the 4 variables as follows: Correlation Analysis Correlation Matrix Influencer_opini on1 Influencer_opini on3 Influencer_opini on2 Influencer_opini on4 Correlation Influencer_opinion1 1.000 .536 .769 .533 Influencer_opinion3 .536 1.000 .458 .428 Influencer_opinion2 .769 .458 1.000 .559 Influencer_opinion4 .533 .428 .559 1.000 Table 1
The correlation between the four variables is moderate. However, Item 1 had a high correlation with Item 2, this may be an indication that the items were measuring a similar construct. Since all variables had a correlation greater than – 0.8 and less than 0.8, we conclude that there was no presence of multicollinearity.
6.2 The KMO and Bartlett’s Test
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .755
Bartlett's Test of Sphericity
Approx. Chi-Square 472.712
df 6
Sig. .000
Table 2
The KMO and Bartlett’s tests show that there is a sampling adequacy of 0.755. This is a good measure as it is greater than 0.6 which is considered to be the least acceptable. Since the Bartlett’s Test of Sphericity has a small p-value, we conclude that the item vary significantly and that the correlation matrix is not an identity matrix.
Communalities Communalities Initial Extraction Influencer_opinion1 1.000 .785 Influencer_opinion3 1.000 .526 Influencer_opinion2 1.000 .758 Influencer_opinion4 1.000 .588 Extraction Method: Principal Component
Analysis.
Table 3
The communality table shows how much of the variance should be considered for further analysis, ideally any variable that is to be considered for further analysis should have a communality value of more than 0.5. As such all the items with less than 0.4 communality value shall be dropped. A high value indicates that the variable is highly related to the other variables and as such measures the same construct. A low value may be due to an inadequate sample. Poor understanding of the question or external factors affecting the responses. Since all items had a communality value greater than 0.5, we shall drop no item at this point.
Total Variance Explained
Total Variance Explained
Component Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % 1 2.656 66.408 66.408 2.656 66.408 66.408 2 .600 14.996 81.403
3 .522 13.043 94.447 4 .222 5.553 100.000 Extraction Method: Principal Component Analysis.
Table 4
From the total variance table, only one component had an Eigen value greater than 1, thus I use this component for further regression analysis.
6.3 Multiple Linear Regression Analysis
The goal with this analysis is to assess the association between two or more independent variables and a continuous dependent variable. The equation used is expressed as follows:
𝑌′= 𝑏0+ 𝑏1𝑋1+ 𝑏2𝑋2+ ⋯ + 𝑏𝑝𝑋𝑝
Where Y’ is the predicted value of the dependent variable, X1 through Xp distinct independent or predictor variables.
Statistical Analysis
Hypothesis 1: Trust in influencer’s opinion would have a positive effect on willingness to buy
Assumptions Testing/ Normality Test
Multiple linear regression assumes that variables have a normal distribution. This means that the errors are normally distributed and the residual will approximate a normal curve. I have tested this using a Histogram superimposed with a Normal curve in SPSS.
Figure 2: Histogram superimposed with a Normal Curve
The Interpretation from this figure leads us to conclude that the normal curve takes a bell shape. We can then conclude that the residuals approximate a normal distribution and thus the model satisfies the normality test. To test for homoscedasticity, I will use a scatter plot with standardized residual values and see how they vary. Multiple linear assumes homoscedasticity, this means that variances along the best fit of line remain similar as you move along.
The homoscedasticity assumption is satisfied since the variance from the scatter plot is approximately uniform on both sides.
Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .470a .221 .218 .88418770 Table 5
From the model summary information, the R Square is equal to 0.221. This implies that trust in influencer explains 22.1% variation in the willingness to buy.
6.5 ANOVA output
The following section presents the ANOVA analysis which would allow me to establish the direction and strength of the relationship between the variables. Furthermore, we can clearly observe the possibility of interaction effect. The following were derived: estimates of effect size, observed power and homogeneity tests - all means for descriptive statistics.
ANOVAa
Model Sum of Squares df Mean Square F Sig.
1
Regression 62.318 1 62.318 79.712 .000b
Residual 219.682 281 .782 Total 282.000 282
a. Dependent Variable: trustInfl
b. Predictors: (Constant), Influencer_Opinion
Table 6
From the ANOVA model, there is a significant relationship (P < 0.05), thus we reject the null hypothesis of non-significant relationship and conclude that influencer opinion is a significant predictor of buyer’s willingness to buy. This confirms Hypothesis 1 of the research.
Next analysis step that is performed is the Regression Coefficient: Coefficientsa
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) -1.521 .178 -8.531 .000 Influencer_Opinion .492 .055 .470 8.928 .000 a. Dependent Variable: trustInfl
Table 7
From the coefficient table, we get the following resulting regression equation:
𝑊𝑖𝑙𝑙𝑖𝑛𝑔𝑛𝑒𝑠𝑠 𝑡𝑜 𝑏𝑢𝑦 = −1.521 + 0.492 𝐼𝑛𝑓𝑙𝑢𝑒𝑛𝑐𝑒𝑟 𝑂𝑝𝑖𝑛𝑖𝑜𝑛
From that statistic, we can conclude that this predictor is statistically significant (p < 0.000). In the interpretation, we conclude that one unit increase in Influencer Opinion, increases the buyer’s willingness to buy by 0.492 units. Thus, we can confidently confirm our hypothesis that Trust in Influencers opinion has a positive Impact on the buyers’ willingness to buy.
To test this second hypothesis we shall fit a multiple linear regression with multiple publications and Influencer’s opinion as the predictor variables. Noting that, it does not require factor analysis since multiple publications are represented by a single independent variable.
H2: Multiple publications in short time period has a negative effect on willingness to buy mediated by trust in influencer’s opinion (knowledge)
The bell shaped normal curve indicate that the normality assumption is not violated. The residuals assume a normal distribution.
Figure 4 Assumptions Testing: Normality
Multicollinearity
Multicollinearity refer to the existence of perfect collinearity or highly correlated variables that may influence the analysis. We use VIF to test for the presence of Multicollinearity. Multicollinearity constitutes a threat which can jeopardize the effective estimation of the type of structural relationship used commonly in the regression analysis. (Farrar and Glauber, 1968) Research states that if there is any value above 3 would indicate presence of multicollinearity.
Unstandardized Coefficients
Standardized
Coefficients Collinearity Statistics B Std. Error Beta Tolerance VIF (Constant) -2.184 .188
In your opinion how likely is an Influencer honest when repetitively posting about a product or a service?
.233 .032 .361 .944 1.060
Table 8
Analyzing and looking at the model all VIF values were less than 3, thus we conclude absence of multicollinearity. Next step is to check the homoscedasticity. Meaning that the assumption of the variance around the regression line is the same for all values of the predictor variable Looking again at the scatter plot, the variance is approximately uniform on both side and thus the homoscedasticity assumption is satisfied.
Figure 5
Model Summary
Model Summaryb
Model R R Square Adjusted R Square
Std. Error of the Estimate 1 .586a .344 .339 .81298275
a. Predictors: (Constant), In your opinion how likely is an Influencer honest when repetitively posting about a product or a service? Influencer_Opinion
b. Dependent Variable: trustInfl
Table 9
The model has an Adjusted R square of 0.339, this implies that the model explains 33.9 % variation in trust of influencers.
Model Output
ANOVAa
Model Sum of Squares df Mean Square F Sig.
1
Regression 96.937 2 48.468 73.332 .000b
Residual 185.063 280 .661 Total 282.000 282
a. Dependent Variable: trustInfl
b. Predictors: (Constant), multiple_publications, Influencer_Opinion
Table 10
From the ANOVA model, the p-value indicates statistical significance (P < 0.05). Thus, I can reject the null hypothesis and conclude that at least one of the predictors is statistically significant. Coefficients Table Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta (Constant) -2.184 .188 -11.631 .000 Influencer_Opinion .402 .052 .384 7.712 .000 Multiple_publications .233 .032 .361 7.237 .000 Table 11
From the resulting table I conclude the following regression equation: 𝑊𝑖𝑙𝑙𝑖𝑛𝑔𝑛𝑒𝑠𝑠 𝑡𝑜 𝑏𝑢𝑦
= −2.184 + 0.402 𝐼𝑛𝑓𝑙𝑢𝑒𝑛𝑐𝑒𝑟 𝑂𝑝𝑖𝑛𝑖𝑜𝑛 + 0.233 𝑀𝑢𝑙𝑡𝑖𝑝𝑙𝑒 𝑃𝑢𝑏𝑙𝑖𝑐𝑎𝑡𝑖𝑜𝑛𝑠 From the t statistic, all the predictor variables were statistically significant (P < 0.000). In this case, the interpretation of the effect of Multiple Publication when mediated by trust in
influencer opinion. In this case one unit increase in Multiple Publications, increases consumer willingness to buy by 0.233 units.
Consequently, this would lead to reject the proposed hypothesis and conclude that multiple publications do not have a negative impact on willingness to buy.
7. Conclusion and Discussion
7.1 Conclusion
The goal of the following research was to investigate how effective can digital influencers be into persuading consumers to buy and test different products they advertise via their social media platforms. To address the research question, two hypotheses were concluded after thorough literature review, leading to answers tested via different statistical tools. The general aim of this study was to distinguish who are todays digital influencers, what is their role into company’s digital marketing strategies and how the created content influence the purchase power of consumers. Base on that, different variables were examined and a thorough research on trustworthiness was conducted in order to confirm the hypothesis that trust in influencer’s opinion has an increasing effect on consumers’ intentions to buy. Moreover, to contribute to the current gap in the literature, this study also aimed to explain the concept of the unlimited ways social media platforms provide for consumers to communicate, create, advocate and share information about brands and their products. Hence, there should be a collective effort between brand-created and influencer and user-generated communications that would be creating multiple opportunities for increasing all brand equity metrics. The assumptions based on theories and existing literature, were tested in SPSS and analyzed in the empirical part of this thesis. Based on these findings, recommendation for future research and theoretical and practical implications can be suggested.
In conclusion, nowadays social media has established huge popularity and have strong influence in the modern society which is affirmed by this research. Companies and individuals have become more lawful into influencing different users’ opinions via social networks. Usage and engagement levels of social media are reaching higher levels and have created better user’s experience, associations and awareness which leads to more positive brand attitude for companies and the main factor that has permitted this change is the creation of content from digital influencers.
7.2 Managerial Implications
From these conclusions, it is possible to draw several implications for practice. The current study’s findings will have the aim to provide valuable insights and implications and serve as advice and guidelines for both brands and digital influencers. Summarized the results provide important managerial insights for brands such as how to best fit influencer marketing into their digital strategy, gain recognition and use the social media outlets to their advantage. Furthermore, it helps digital influencers understand consumers’ trust and the relevant strategy for disclosing information. As more social media platforms arise and others continue to grow, their influence on the market would increase, important for future research would be to study more in depth the phenomena and continue focus on sources of endorsements to digital influencers by the brands. This would present empirically beneficial results for consumers, markets, brands and endorsers. The results provided by this study could have practical implications for professionals in corporate communication. They could be used as benchmark for influencer marketing campaign. Communication experts should aim to focus on measurements of a successful influencer campaign such as sentiment of comments, buzz and
mentions and ROI. However, focusing on engagement rate could be misleading and would lead to an unsuccessful influencer marketing campaign.
7.3 Limitations and future research
The conducted research was established by using a sample of respondents consisting of personal friends, family, fellow students and my own social network. Therefore, the sample I have gathered consists of people that are rather same age as mine or younger, mostly already highly educated, who use social media on a regular basis, since there was no respondent that has stated he/she does not use social media channels – which was the first question of the conducted survey. However, the sample does not fully represent the majority of social networks users since nowadays all age groups use social media network sites. As a consequence, the results could not be applicable to a large extent of people, however since most of them were millennials it would be applicable for the age group of 18-34, which is a big sample of today’s users, who have great influence in their networks. Moreover, the age group that participated in the study consist of emerging adults that spend more time online than doing any other activity (Coyne, Padilla-Walker and Howard, 2013), which can conclude it is representable group for the purposes of this thesis.
Even though, most of the respondents are based in Europe which could be a limitation in regards of cultural views, it would present an opportunity for future research including a broader selection of nationalities including America and Asia, where influencer marketing has a strong influence in companies’ marketing strategies. Since in my research question I have focused more on the trust and willingness to buy that is influenced by digital influencers, an interesting topic for future research would be to examine all different types of sponsorship disclosure in the field of influencer marketing on social media. An example of how sponsorship
disclosures and types of tags affect consumer’s trust and opinions about the brand in general. Different social media platforms, such as blogs, vlogs and videos on YouTube, Snapchat and Instagram could be examined. Despite some of the shortcomings, it should be summarized that digital influencers are powerful online communities that can influence their audience and are becoming a highly credible source for many. This thesis showed interesting insights on how influencing works in digital marketing. Furthermore, influencers are here to stay and provide a lot of positive opportunities for brands to reach their target market. Thus, embracing influencers should be marketing managers’ future goal. Moreover, this research is a great starting point for the still needed further investigation.
References:
Akbar, M. M. & Parvez, N. 2009. Impact of Service Quality, Trust, And Customer Satisfaction Engender Customers Loyalty. ABAC Journal, 29.
Arkkelin, Daniel. (2014). "Using SPSS to Understand Research and Data Analysis" Psychology Curricular Materials. Book 1
Auriat, N., & Saniscalco, M. (2005). Quantitative research methods in educational planning. International Institute for Educational Planning
Bambauer-Sachse, S. and Mangold, S. (2013). Brand equity dilution through negative online word-of-mouth communication. Journal of Retailing and Consumer Services, Vol 18, pp. 38-45.
Bartholomew D. (1980) Factor Analysis for Categorical Data. Journal of the Royal Statistical Society. Series B (Methodological), Vol. 42, No. 3, pp. 293-321
Berthon, P.R., Pitt, L. and Campbell, C. (2008). “Ad lib: when customers create the ad”, CA Management Review, Vol. 50 No. 4, pp. 6-31.
Berthon, P. R., Pitt, L. F., Plangger, K. and Shapiro, D. (2012). Marketing meets Web 2.0, social media, and creative consumers: Implications for international marketing strategy, Business Horizons, vol. 55, no. 3, pp. 261-271
Bickart, B. and Schindler, R. M. (2001). Internet forums as influential sources of consumer information. Journal of Interactive Marketing, Vol 15: pp. 31–40.
Booth, N. and Matic, J. (2011). Mapping and leveraging influencers in social media to shape corporate brand perceptions. Corporate Communications: An International Journal, Vol. 16 Issue: 3, pp.184-191
Brown, D., and Fiorella, S. (2013). Influence marketing: How to Create, manage, and measure Brand Influencers in Social media marketing. Que Publishing.
Bruhn, S., Schafer, D. and Heinrich D. (2012). “Brand Authenticity: Towards a Deeper Understanding of its Conceptualization and Measurement”, Advances in Consumer Research, Vol. 40, pp. 567-576
Burke, C. (2016). How the cost of Influencer marketing compares to paid ads, in 5 charts. Retrieved February 9, 2017, from http://www.mavrck.co/5-ways-to-measure-digital-and-influencer-marketing-campaigns/
Chaudhuri A. and Holbrook, M. (2001). The Chain of Effects from Brand Trust and Brand Affect to Brand Performance: The Role of Brand Loyalty. Journal of Marketing: April 2001, Vol. 65, Number. 2, pp. 81-93.
Chaudhuri A. and Holbrook, M. (2002). Product-class effects on brand commitment and brand outcomes: The role of brand trust and brand affect. The Journal of Brand Management, Vol. 10, Number 1, pp. 33-58(26)
Cobbledick, C. (2016). IAB South Africa’s press office 2017, from Bizcommunity, http://www.bizcommunity.com/Article/196/459/141022.html
Colliander, J. and Dahlén, M. (2011). Following the fashionable friend: The power of social media. Journal of Advertising Research, Vol 51(1), pp. 313-320
De Vaus, D. (1996). Surveys in Social Research. Vol 5 of Social research today.
Dickinson J., Beverland, S. and Farrelly, F. (2014). Measuring consumer-based brand authenticity. Journal of Business Research, Vol. 67(6), pp. 1090-1098.
Dodds, W. and Kent M. (1991). "Effects of Price, Brand and Store Information on Buyers' Product Evaluations," Journal of Marketing Research, Vol 28, pp. 307-19
Erdem, T., & Swait, J. (2004). Brand credibility, brand consideration, and choice. Journal of Consumer Research, Vol 31(1), pp. 191–198.
Farrar and Glauber. (1967). Multicollinearity in Regression Analysis. The Review of Economics and Statistics, Vol. 49
Fisher-Buttinger C. and Vallaster, C., (2008). Connective branding: Building brand equity in a demanding world John Wiley & Sons, London
Gilmore, J. H. and Pine, J. B. (2015). Authenticity: What consumers really want. United States: Harvard Business School Press.
Gladwell, M. (2000). The tipping point: How little things can make a big difference. Boston: Little, Brown and Company.
Grant, J. (2000). The new marketing manifesto: Building successful brands in the 21st century. London: Orion Business (an Imprint of The Orion Publishing Group Ltd).
Hall, T. (2010). 10 Essential Rules for Brands in Social Media. Retrieved April 25, 2017, from http://adage.com/article/digitalnext/10-essential-rules-brands-social-media/142907/
Hanna, R., Rohm, A. & Crittenden, V. L. (2011). We’re all connected: The power of the social media ecosystem, Business Horizons, vol. 54, no. 3, pp. 265-273
Hennig-Thurau, T., Wiertz, C., & Feldhaus, F. (2014). Does Twitter matter? The impact of microblogging word of mouth on consumers’ adoption of new movies. Journal of the Academy of Marketing Science, Vol 43(3), pp. 375–394
Hiscock, J. (2001). "Most Trusted Brands," Marketing, March 1st, 32-33
Holt, D. (2016). Branding in the age of social media., https://hbr.org/2016/03/branding-in-the-age-of-social-media
Hwang, Y. & Jeong, S. (2016). “This is a sponsored blog post, but all opinions are my own”: The effects of sponsorship disclosure on responses to sponsored blog posts. Computers In Human Behavior, Vol 62, pp. 528-535
Hye-Ryeon L., Hye Eun Lee, Jounghwa Choi , Jang Hyun Kim & Hae Lin Han (2014) Social Media Use, Body Image, and Psychological Well-Being: A Cross-Cultural Comparison of Korea and the United States, Journal of Health Communication, pp. 1343-1358
Inc, M. (2015). Instagram marketing: Does Influencer size matter? – Markerly Blog, from http://markerly.com/blog/instagram-marketing-does-influencer-size-matter/
Jacoby, J. and Chestnut, R. (1979). Brand Loyalty, Measurement and Management. Journal of advertising, pp. 120
McAlexander J., Schouten J. and Koenig H. (2002) Building Brand Community. Journal of Marketing. Vol. 66, No. 1, pp. 38-54
Kang, J., and Hustvedt, G. (2013). Building trust between consumers and corporations: The role of consumer perceptions of transparency and social responsibility. Journal of Business Ethics, Vol. 125(2), pp. 253–265.
Kapitan, S. and Silvera, D. (2016). From digital media influencers to celebrity endorsers: attributions drive endorser effectiveness. Marketing Letters. pp.1-15
Kaplan, A. and Haenlein, M. (2012). The Britney Spears universe: social media and viral marketing at its best”, Business Horizons, Vol. 55 No. 1, pp. 27-31
