“The influence of consumer usage of
a company´s Facebook page on the
level of brand involvement”
“The influence of consumer usage of a
company´s Facebook page on the
level of brand involvement
“
By
Jelte Bekius
Rijksuniversiteit Groningen
Faculty of Economics and Business
Master Thesis Business Administration
Marketing Management
Groene Velden 34 8211 BA Lelystad (06)13727716 J.Bekius.2@student.rug.nl Student number 1918087 June, 2012
MANAGEMENT SUMMARY
PREFACE
This master thesis has been written to finish my Master of Science in Marketing Management at the University of Groningen. It is the end of a great period in my life as a student in which I have learned a lot. I am looking forward to the next step in my life at the Dutch Marine Corps, which will start in August 2012. I hope I can take advantage of the seven years of study that are belonging to the past now. I would like to thank my supervisor prof.dr. Janny C Hoekstra for her feedback, support and time during the entire process. Finally, thanks to my second supervisor MSc. Daniela N Naydenova who gave interesting feedback to make final improvements to this thesis.
Jelte Bekius
TABLE OF CONTENTS
1 INTRODUCTION ... 7
1.1 NEW MEDIA ... 7
1.2 NEW COMMUNICATION CHANNELS ... 8
1.3 FACEBOOK AND COMPANIES ... 9
1.4 CHALLENGES AND OPPORTUNITIES OF SOCIAL MEDIA ... 10
1.5 PROBLEM STATEMENT ... 11
1.6 ACADEMIC AND MANAGERIAL RELEVANCE ... 11
1.7 STRUCTURE ... 12
2 THEORETICAL FRAMEWORK & HYPOTHESES ... 13
2.1 INTERNET &NEW MEDIA DEVELOPMENT ... 13
2.2 ONLINE BRAND COMMUNITIES ... 14
2.3 OUTCOMES OF BRAND COMMUNITIES ... 15
2.4 SOCIAL MEDIA ... 16
2.5 SOCIAL NETWORKING ... 17
2.6 FACEBOOK ... 17
2.6.1 Importance of a corporate Facebook page ... 18
2.6.2 Creating value at Facebook ... 19
2.6.3 Facebook’s interactivity ... 21
2.7 CONSUMER BRAND INVOLVEMENT ... 22
2.7.1 The Involvement construct ... 22
2.7.2 Discovering the involvement construct in the Facebook perspective ... 23
2.8 FACEBOOK INTENSITY ... 24
2.9 WORD-OF-MOUTH ... 25
2.10 ELECTRONIC WORD-OF-MOUTH ... 26
2.11 PERCEIVED VALENCE OF INFORMATION ... 27
2.12 USER-GENERATED CONTENT SENSITIVITY ... 28
2.13 PERCEIVED USER-GENERATED CONTENT ... 29
2.14 CONCEPTUAL MODEL ... 30 3 RESEARCH METHODS ... 31 3.1 SURVEY ... 31 3.2 MEASUREMENT OF CONSTRUCTS ... 31 3.3 FORMATIVE SCALES ... 35 3.4 SAMPLING ... 36
3.5 USAGE FORMATIVE OR REFLECTIVE SCALE ... 37
3.6 PLAN OF ANALYSIS ... 37
3.6.1 Simple Regression ... 37
3.6.2 Multiple Regression and Interaction terms ... 38
3.6.3 Multicollinearity ... 38 3.6.4 General assumptions ... 39 4 RESULTS ... 40 4.1 SINGLE REGRESSION ... 40 4.2 MULTIPLE REGRESSION ... 40 5 DISCUSSION ... 42
5.1 CONCLUSION MAIN EFFECT ... 42
5.2 CONCLUSIONS MODERATORS ... 43
5.3 LIMITATIONS AND RECOMMENDATIONS ... 45
5.4 THEORETICAL IMPLICATIONS ... 46
5.5 MANAGERIAL IMPLICATIONS ... 46
5.5 CONCLUSION ... 46
REFERENCES ... 48
APPENDIX1 QUESTOINAIRE ... 54
APPENDIX2 MULTICOLLINEARITYCHECK ... 60
1
INTRODUCTION
“In 1995, when the first notable social networking website, Classmates.com, was launched, few might have guessed that 15 years later, social networking sites would have tens of millions of users and would be valued at billions of dollars” (Trusov et al., 2010).
T
imes are changing. This is probably a very common statement to start a research paper introduction, but very applicable to this research. The above sentence about classmates.com is a good example of a technological change with huge impact. As technology continues to break new ground and advance in leaps and bounds, its role in our lives is, without question, felt in every little way. Technology seems to change more rapidly than ever before, and allows us to connect and communicate with each other like never before the twentieth century. ‘In the current marketing environment, sustaining a competitive advantage on the basis of product differentiation often is a fatiguing race to a constantly shifting finish line. Any lead in the race is eroded quickly by replication or even by superior technology from competitors. One way to avoid this treadmill, is to redefine the terms of competitive advantage’ (McAlexander et al., 2002). “Successful companies should anticipate to (constantly) changing marketing conditions, be reactive to consumer (changing) needs and compete by offering superior value,” (MSI Research Priorities, 2012). Competitive advantage can be gained through several channels in terms of customer relationship management and different forms of strategic marketing orientations (Rust et al., 2002; Shah et al., 2006; Treacy and Wiersema, 1993; Kumar and Reinarts., 2006; Mittal et al., 2002; Langerak and Verhoef, 2003). One way to become a successful company is to be the first one in adapting new technologies. “This kind of successful companies foster an entrepreneurial atmosphere and can commercialize new technologies quickly,” (Treacy and Wiersema, 1993). Furthermore, these successful companies must develop a coherent marketing plan that delivers value to customers and captures value for shareholders (MSI Research Priorities, 2012). It might seem easy to become a (relationship) marketing champion, however, reality shows that it is hard to establish the corporate structure and culture to become one.1.1 New Media
et al., 2002). This addresses the need for companies to quickly discover new technologies and respond to them. ‘The rise of New Media channels such as Facebook, YouTube, Google and Twitter, has enabled customers to take a more active role as market players and reach (and be reached by) almost everyone, anywhere and anytime’ (Hennig-Thurau et al., 2010). ‘The digital innovations of the last decade made it effortless, indeed second nature, for audiences to talk back and talk to each other’ (Deighton and Kornfeld, 2009). Consumers no longer require their computer to communicate online. Through high-tech mobile phones, portable computers and portals such as Twitter, real-time information exchange has become an integral element of consumer behavior anywhere and anytime (Hennig-Thurau et al., 2010). This growth of New Media has changed and is still changing the entire marketing environment (Mathwick et al., 2008). Today, the flow of information about a brand has become multidirectional, interconnected, and difficult to predict. Marketers have lost control over their brand(s). In the current market, companies are participating in a conversation about their brand (Deighton and Kornfeld, 2009). Furthermore, New Media has empowered consumers to promote and distribute their own offers. Consumers serve as retailers on eBay, media producer on YouTube, authors on Wikipedia, and reviewers on Amazon and Tripadvisor for example. Consumers do all of this and more on Facebook and MySpace (Hennig-Thurau et al., 2010). Facebook and MySpace are absorbing functions from other channels and start delivering the ´total New Media package´ (Kietzmann et al., 2011). Playing music from Spotify and movies from YouTube on Facebook are examples of these integrated (take-over) features. This New Media phenomenon means for companies that they: are facing more competition, more product comparisons and maybe the most challenging occurrence, more electronic word-of-mouth and user-generated content. New Media offers companies numerous ways to reach consumers, communicate with them, and measure their communication, browsing or purchase-related behaviors (Hennig-Thurau et al., 2010).
1.2 New communication channels
even help other consumers to solve product-related problems for free, which reduces service costs and increases quality (Mathwick et al., 2008). Brands are widely discussed on the different New Media channels. Imagine an angry post from a consumer of the Dutch NS railways in the morning because of a delayed train, or the message from a friend about the delicious cup of coffee he ordered the same morning at Starbucks. The effects of these kinds of postings still remain mostly undiscovered (MSI research priorities, 2010-2012). When Facebook started, companies and brands were no part of it. Companies were not able to react or anticipate to consumer postings. The introduction from corporate Facebook pages in 2006 brought change (Facebook, 2012). From the year 2006, companies were able to be proactive and to respond on Facebook.
1.3 Facebook and companies
word-of-mouth and user-generated content concerning a company or brand on Facebook are not properly researched yet. Both constructs are discussed in many studies, however not in combination with Facebook. This ‘free consumer-service’ delivered to consumers by consumers is an attractive trend for companies which requires attention in the marketing literature. This trend might be interesting for companies and brands in a way that companies should encourage and support this form of ‘free-service’ to gain better financial results and more loyal customer relationships (Hennig-Thurau et al., 2010).
1.4 Challenges and opportunities of Social Media
in (Trusov et al., 2010). Friends of the posting member at Facebook read these updates. The postings are assumed to have an effect on other consumers (Adjei et al., 2010). The corporate postings are assumed to have an effect to the ‘Likers’ too (Hennig-Thurau et al., 2010). Another question is to find out what the effect is of consumer postings at a corporate Facebook page concerning the company’s activities, products, events and the brand in general on other consumers. Because of the fact that consumers might influence other consumers (Adjei et al., 2010), the answer to that question is relevant for companies and brands.
1.5 Problem Statement
The previous paragraphs serve as an introduction to the essence of this study. The study is build around two key variables: (1) Consumer usage of a company’s or brands Facebook page and (2) Consumer brand involvement. These two constructs will be described extensively in the theoretical part of this study. In addition to the proposed relation between the independent and the dependent variable, five moderators are added to the (tested) model: (1) Electronic word-of-mouth, (2) Perceived user-generated content, (3) User-generated content sensitivity, (4) Perceived valence of information and (5) the Facebook Intensity level. The following problem statement is formulated:
What is the effect of the intensity level of the usage of a company’s or brands Facebook page on the level of the consumer brand involvement and what are the moderating effects of E-WOM, Perceived UGC, UGC sensitivity, Facebook Intensity and Perceived valence of information?
1.6 Academic and managerial relevance
importance. The MSI discussed the importance of the implications of New Media and channels (social and mobile) for managing customer relationships, marketing communications and branding. They mentioned the fact that companies still doubt how to use New Media to interact with, and improve relationships with the customer. Furthermore, the MSI mentioned that the role of the brand is changing because of Social Media. Brands do more than just promise functional and emotional benefits: they are now facilitators of conversations about personal and social issues (for example: identity and sustainability). The question regarding the approach of Social Media is still an important one in the marketing literature (MSI, 2012). This study tries to contribute to the current literature regarding that question. The study of Hennig-Thurau et al., (2010) illustrated that the rise of New Media is disruptive for the management of relationships with customers. However, Hennig-Thurau et al., (2010) do not mention the effects of the usage of a company’s or brands Facebook page on relationships outcomes. Their framework illustrates that New Media requires a shift in marketing thinking. Consumers have become highly active partners, serving as customers as well as producers and retailers, being strongly connected with a network of other consumers (Hennig-Thurau et al., 2010). This study will hopefully contribute to the research of Hennig-Thurau et al., (2010) in a way that it tries to build further on their existing framework.
1.7 Structure
2
THEORETICAL FRAMEWORK & HYPOTHESES
This chapter starts with an overview of the Internet and New Media landscape. Social Media and online brand communities are explained afterwards. This is followed by a discussion of the seven constructs used for this study: Consumer usage of a company’s Facebook page, Consumer involvement with the brand, Facebook Intensity, Electronic word-of-mouth,
Perceived valence of information, Perceived user-generated content and User-generated content sensitivity. The relations between the constructs are based on the findings in existing literature. The hypotheses are summarized in the conceptual model, which is the overview of this study.
2.1 Internet & New Media development
engage, and respond to them. As stated by Homburg et al., (2000) there is growing evidence in the business press that the ways firms are organizing their marketing activities is subject to major changes. Many companies have changed their organizational structures to become more responsive to customer needs (George et al., 1994). Oktemgil and Greenley (1997) conducted research into the consequences of high and low adaptive capability of companies. They concluded that higher adaptive capability is still likely to lead to higher return on investment. Thus, it is relevant for marketing science to understand which factors are influencing success in New Media and to develop the skills to adapt to those determined factors.
2.2 Online Brand Communities
“Social networking practices are those that focus on creating, enhancing, and sustaining ties among brand community members, these include welcoming, empathizing, and governing” (Schau et al., 2009). Schau et al. (2009) stated that ‘this trio of practices highlights the homogeneity of the brand community, or the similarities across brand community members and their normative behavioral expectations of themselves and one another’. The three practices are focusing on the intangible domain and emotions and create a more social bond within the community (Schau et al., 2009). The study of Schau et al., (2009) challenged the assertion that brand communities are communities of limited liability as concluded by Muniz and O’Guinn’s (2001). Schau et al., (2009) suggested that social networking practices evolve and move past brand boundaries. If a company gives the consumers the opportunity to construct brand communities and the freedom to modify their products, they will (Schau et al., 2009). Therefore, companies should supply the consumers with the necessary tools to participate in an online brand community. Facebook is a sufficient channel to accomplish this (Hennig-Thurau et al., 2010). Schau et al., (2009) argued that ‘if more practices lead to a stronger brand community, marketers should strive to encourage greater diversity in practices because these multiple opportunities serve to cultivate these markets’. Companies that want to encourage co-creation should foster a broad array of practices, not merely customization, as is (mostly) the current focus (Schau et al., 2009).
2.3 Outcomes of Brand Communities
Given that communication within communities can affect brand perception, the question is how ‘general’ communities such as Facebook can be used for brand management. Regarding the relationship outcomes, existing research has shown that referral in online communities enhances customer acquisition (Trusov et al., 2009). The study of Algesheimer et al., (2005) described how identification with the brand community leads to positive consequences, such as greater community engagement, and negative consequences, such as normative community pressure and (ultimately) reactance. ‘Brand communities are about to learn consumer evaluations of new offerings, competitive actions, and so forth, and to maximize opportunities to engage and collaborate with highly loyal customers’ (Franke and Shah 2003).
communities show three traditional principles of community: shared consciousness, rituals and traditions, and a sense of moral responsibility (McAlexander et al., 2002). McAlexander et al., (2002) concluded that benefits for a firm, which are arising from a brand community are many and diverse. Community-integrated customers serve as brand missionaries, carrying the marketing message into other communities. They are more forgiving than others of product failures or lapses of service quality. ‘They are less adaptive to switch brands, even when confronted with superior performance by competing products. The members are motivated to provide feedback to corporate ears and they constitute a strong market for licensed products and brand extensions’ McAlexander et al., (2002). In addition, the study of McAlexander et al., (2002) concluded that in many cases, loyal customers making long-term investments in a company's stock. Customers who are highly integrated in the brand community are emotionally invested in the welfare of the company and desire to contribute to its success (McAlexander et al., 2002). A community is made up of its member entities and the relationships among them. Numerous companies are using social networking sites to support the creation of brand communities (Muniz & O’Guinn, 2001) or for marketing research (Kozinets, 2002). To promote a new flavor of their ice cream, Ben & Jerry’s launched an online competition facilitated mainly by Facebook (September, 2011).
2.4 Social Media
significantly lower costs and fast delivery-especially through technology, such as the Internet’, stated by Trusov et al., (2009). Firms are at the crucial point whether to get into this Social Media ‘hype’ or not (Kietzmann et al., 2011). Kaplan & Haenlein (2010) stated that the Social Media revolutionary new trend should be of interest to companies operating in online space or any space.
2.5 Social Networking
The Marketing Science Institute (2006) selected “The Connected Customer” as its top research priority. The core of a social network site is a group of user profiles where registered members can place information they want to share with other users, mostly friends or followers (Trusov et al., 2010). For the most part, users are involved in two kinds of activities on the site: (1) they create new content by editing their profiles by for example: adding pictures, uploading music, writing blogs and messages, or (2) they consume content that others create by for example: looking at pictures, downloading music, reading blogs, and messages (Trusov et al., 2010). ‘On most of the social network sites, users can add other users to their networks of friends. Usually, one user initiates the invitation, and the other user accepts or rejects it. When accepted, the two profiles become linked’ (Trusov et al., 2010). When two profiles are linked, the users can see the total present information, which the other user (now called a ´friend´) has uploaded. Facebook, the core of this research, enables its users to present themselves in an online profile, build up a network of real and (real) online ‘friends’ who can post comments on each other’s pages, and view each other’s profiles. In addition, Facebook members can join virtual groups based on common interests and learn each other’s hobbies, interests, musical tastes, and romantic relationship status through the profiles’ (Ellison et al., 2007). Social network sites can be oriented towards: (1) work-related contexts (e.g., LinkedIn.com), (2) romantic relationship initiation (the original goal of Friendster.com), (3) connecting those with shared interests such as music or politics (e.g., MySpace.com), (4) or the college student population (the original idea of Facebook.com)’ Ellison et al., (2007). The success of Internet social networking sites depends on the number and activity levels of their user members (Trusov et al., 2010).
2.6 Facebook
year’s facts: less than ten years ago, Facebook did not even exist. In January 2012, the Facebook network has more than 800 million registered users and more than 50% of the active users log on to Facebook in any given day (facebook.com/press, 2012). Compared to the total world population, Facebook is the third largest community listed after China and India respectively and listed above the United States, representing the fourth place (Socialnomics, 2012). The average user on Facebook has 130 friends, there are more than 900 million objects that people interact with (pages, groups, events and community pages), the average user is connected to 80 community pages, groups and events and on average, more than 250 million photos are uploaded per day. Besides these self-explanatory facts, analyzing the growth rates, Facebook will, without question, soon be the largest community in the world compared to offline communities (Socialnomics, 2012). Facebook is an up and coming, but definitely already established business and ´society´ which requires attention in the broadest sense from entrepreneurs who want to be innovative and create competitive advantage throughout product innovation.
2.6.1 Importance of a corporate Facebook page
brands create external competitions and elections with prices for the participating ´Likers’. To increase the amount of Facebook users who are ‘liking’ the AJAX Facebook page, AJAX is giving away free tickets to popular football matches to encourage the consumers to ‘Like’ the page. However, the effects of more ‘Likers’ are unknown. Facebook continues to establish itself as a valuable business platform (Hof, 2011). If managers are trapped in the mindset that Facebook can only be used as a social instrument to connect with friends, it is time to reconsider this (Malhotra et al., 2011). When organizations did not create a business page yet, there are some facts that might make those organizations reconsider. More than 1.5 million organizations have a business page and every day, 20 million people become fans of those Facebook pages. ´They tie their personal identity and online affinity to a brand, that's powerful´ (Malhotra et al., 2011). Many companies introduce a Facebook link to the corporate website and vice versa. Companies are actively developing social media identities, enabling the firms to enhance brand attractiveness and attract consumer attention (Kane et al., 2009). Social networking sites function as channels for interaction between a company and their customers, helping to convey company brand value. Kane et al., (2009) stated that many companies have built a Facebook fan page to create online social communities and establish closer ties with their customers. Marketing communication platforms established on social networking sites not only allow enterprises to promote themselves and their products, but also provide customers with a channel through which inquiries and comments can be posted. This, in turn, allows enterprises to receive customer feedback and improve future operations (Li and Bernoff, 2008). Social capital is an important resource for establishing communities, with the potential to provide extensive benefits for customers and businesses alike.
2.6.2 Creating value at Facebook
Bernoff (2008) suggested that social networking sites (fan pages) could help corporations to transmit their brand name and message.
2.6.3 Facebook’s interactivity
The goal of operating fan pages is not only to convince fans to click the Like button, but too convince them to want to own or use targeted brand products.
2.7 Consumer brand involvement
The second key (and dependent) variable in this study is ‘Consumer brand involvement’. The previously discussed (independent) variable ‘Consumer usage of a company’s Facebook page’ is predicted to have a relation with the level of consumer brand involvement. In this paragraph, the construct itself and the relation between both variables will be discussed.
2.7.1 The Involvement construct
Involvement with products and services, including leisure, has been explored extensively within marketing and consumer behavior disciplines (Slater and Armstrong, 2010). Despite the fact that there is no universal definition of the involvement construct, commonality exists amongst researchers who posit that involvement with a product or service comprises personal relevance, interest, or importance (Slater and Armstrong, 2010). This outcome is the first bridge to the ‘Consumer usage of a company’s Facebook page’ construct. As stated before, consumers are attracted to a (corporate) Facebook page because of their personal relevance, interest and importance, it is their real-life counterpart (Schau et al., 2009). This is the first relation between the two variables: both constructs are personal and build up around levels of interest and importance. This is the reason to state that, a higher level of one variable, will lead to a higher level of the other variable. For instance, more usage of the AJAX Facebook page, will lead to more AJAX consumer brand involvement, because the level of personal relevance is higher and stimulated more regularly than when the Facebook page is not used. This leads to the question what brand involvement is.
relevance of the product to the individual. Involvement is connected to an individual’s self-concept. Similarly, Park and Young (1986) state that “most researchers agree that the level of involvement can be understood by the degree of personal relevance or importance.” In addition, Richins and Bloch (1986) define involvement in terms of the consumer’s degree of interest or arousal for a given product. On the basis of these previous studies, Malär et al., (2011) defined product involvement as the personal relevance of the product, which is determined by the extent to which the product is interesting and important to the consumer. As stated in paragraph 2.6.1, Facebook is the perfect platform for facilitating brand awareness and promoting brand engagement, because its core consists of social networking and relationship building (Tuten, 2008). Companies are able to use Facebook features to be more important and more interesting for consumers. When interesting stories, product updates and other relevant information are placed at the corporate Facebook page, it is likely that the level of brand involvement will increase because the consumer is exposed to more relevant information than when the consumer is not making use of the Facebook page. When personally relevant knowledge is activated in memory, a motivational state is created that energizes or drives consumers’ cognitive behavior (attention, comprehension, information search, Celsi and Olson 1988) or affective responses (emotions, Park and Young 1986). Although, there are products that on average (across various consumers) are high or low involvement, the actual level of involvement is still defined individually. “In other words, involvement resides within the consumer but is influenced by the product.” (Park and Young 1986).
2.7.2 Discovering the involvement construct in the Facebook perspective
interesting) stories (Ben & Jerry’s, 2012). Perceived service quality is a construct, which is visible at Facebook as well. The possibility to post comments (Electronic word-of-mouth) about service quality is assumed to have an influence on other consumers as well. Kapferer and Laurent (1985a, 1985b) and Zaichkowsky (1985) remain prominent in discussions regarding the measurement of involvement due to their significant contributions, including the Consumer Involvement Profile (CIP) and Personal Inventory Scale (PII) respectively. These scales have been used, adapted, and found to be valid in many research contexts including services and leisure (e.g. Brakus et al., (2009). It can be concluded that the level of consumer involvement with a product, service or brand is likely to increase when a consumer is making more use of this product, service or brands Facebook page. The presence of interesting, informing and entertaining content on the Facebook page will have a stimulating role for the level of brand involvement. In line with the theories mentioned in this paragraph, the first hypothesis is formulated as follows:
H1: The intensity level of consumer usage of a company’s Facebook page has a positive
influence on the level of consumer brand involvement. 2.8 Facebook Intensity
sees more endorsement news feeds of the ‘Liked’ companies than a consumer with a lower Facebook Intensity level, the level of brand involvement is assumed to be higher. The consumer with a high Facebook Intensity level is more active on Facebook, and for that reason, is probably more active in discussions and product evaluations about the company, which will increase the level of brand involvement. The Facebook Intensity level is assumed to have a positive moderating effect on the relation between the Consumer usage of a company’s Facebook page and the level of Consumer brand involvement. In line with the theory described, the following hypothesis is formulated:
H2: The Facebook intensity level is positively moderating the effect of consumer usage of a
company’s Facebook page on the level of consumer brand involvement. 2.9 Word-of-mouth
the most important rule of WOM marketing is to be interesting and that nobody talks about boring companies, boring products, or boring ads’. Word-of-mouth is called the worlds most effective, yet least understood marketing strategy. Marketers are particularly interested in better understanding WOM because traditional forms of communication appear to be losing effectiveness (Nail 2005). Trusov et al., (2009) find that referrals have a strong impact on new customer acquisition. In addition, Trusov et al., (2009) state that electronic-WOM communication is a critical factor for firms seeking to acquire new customers and that WOM can have larger and longer lasting effects than traditional marketing activity.
2.10 Electronic word-of-mouth
brand awareness, are indicators that E-wom has influence on consumer behavior and outcomes (Dellarocas and Narayan 2006). It is assumed that E-wom has a moderating effect on the relation between consumer usage of a company’s Facebook page and the level of consumer brand involvement. Posts from consumers on the corporate Facebook page will probably have a positive effect on the level of brand involvement. In addition, it is assumed that, the more a consumer is posting about a company or brand on Facebook, the higher the level of brand involvement will be. When a consumer is posting more brand or company related messages than another consumer on Facebook, this consumer is assumed to have a higher level of brand involvement. The effects from E-wom are proven to have a positive effect to fellow users and consumers of a certain product when they are positive, but also when the message is less positive (Trusov et al., 2011). When a consumer is spreading the word about a brand or company, it is assumable that this consumer is more attracted, interested and enthusiastic about this company. The previously discussed characteristics of brand involvement, personal relevance and interest, play an important role in this. In line with the mentioned theories and previous studies, this study proposes that the effect from the consumer usage of a company’s Facebook page on the level of consumer involvement with the brand is positively moderated by Electronic word-of-mouth initiated by the consumer itself. Brand advocating activities, like spreading positive messages on the corporate Facebook page are assumed to have a positive effect on the relation between the dependent and the independent variable. A consumer, who is posting regularly about the brand or company on Facebook, and who is involved in brand or company related discussions, is probably facing a stronger effect from the independent to the dependent variable than another consumer who is posting less to none messages on Facebook. In line with the mentioned theories, the following hypothesis is formulated:
H3: Electronic word-of-mouth, initiated by the consumer himself, is positively moderating the
relation between consumer usage of a company’s Facebook page and the level of consumer brand involvement.
2.11 Perceived valence of information
consumers, who faced poor customer service or products, probably want to share their criticisms. The second message will be of different character than the first one. Valence refers to the extent to which the information exchanged reflects positively or negatively on the product or brand in question (Adjei et al, 2010). The study of Adjei et al., (2010) followed the existing theories about the asymmetric effects of the valence of information on customer evaluations (Sirdeshmukh et al., 2002). Furthermore they used the theories from Ba and Pavlou (2002) and Wyer and Gordon (1982) who concluded that negative information often shows a stronger effect than positive information, since negative information has a more distinctive coding in memory. Contrary, the results of Adjei et al., (2010) concluded that,
although negative information can be shared among community members, the impact of the negative information is not as strong as the benefits of positive information. ‘So, maintaining a brand community that allows customers to know the firm more intimately through consumer-to-consumer conversations will work in the firm’s favor, even if negative information is shared’ (Adjei et al., 2010). In accordance with the theory and results of Adjei et al., (2010), this study proposes that when the valence of information perceived by the consumer is positive, it will have a stronger positive moderating effect on the relation between the consumer usage of a company’s Facebook page and the level of consumer brand involvement, than when the perceived valence is more negative. In line with the discussed theory in this paragraph, the following hypothesis is formulated:
H4: When the perceived valence of information by the consumer is positive, it has a stronger
moderating effect on the relation between consumer usage of a company’s Facebook page and the level of consumer brand involvement than when the perceived valence of information is negative.
2.12 User-generated content sensitivity
why consumers are attracted to these new media and how they influence consumer’s affect and behaviour (Hennig-Thurau et al., 2010). Little is known about consumer behaviour with regard to new multimedia products (Kaplan and Haenlein, 2009b). The UGC construct is not being studied in various ways. Facebook is almost fully created by UGC and many effects from Facebook are studied as mentioned before in several ways. The precise effect from UGC on consumers is lacking in the current literature. As stated before, consumers are different and this is why consumers are likely to have different attitudes towards UGC. Some consumers probably evaluate UGC as relevant and reliable, where others might think that, UGC is unreliably and meaningless because the reader has no idea about the author. Consumers do not know whether the author is an experienced user, an expert or a consumer without any knowledge or expertise. More involved with and more sensitive consumers to UGC are probably more influenced by UGC and it is predicted that the consumers’ level of brand involvement changes by reading UGC. It is proposed that the effect of consumer usage of a company’s Facebook page on the level of consumer involvement with the brand will be more positively moderated when the consumer involvement with UGC is high than low. Following these theories about UGC, the following hypothesis is formulated:
H5: User-generated content sensitivity has a positive moderating effect, such that higher
user-generated content sensitivity will increase the effect from consumer usage of a company’s Facebook page on the level of consumer brand involvement.
2.13 Perceived user-generated content
more personal interesting and relevant due to the perceiving of this extra information. It is proposed that when a consumer perceives more (amount of) UGC, it will positively moderate the effect from consumer usage of a company’s Facebook page on the level of consumer brand involvement. In line with the theory mentioned, the following hypothesis is formulated: H6: The perceived amount of user-generated content has a positive moderating effect, such
that more perceived user-generated content will increase the effect from consumer usage of a company’s Facebook page on the level of consumer brand involvement.
2.14 Conceptual model
The hypotheses and relations between the constructs as described in this chapter can be displayed in the following conceptual model:
Facebook Intensity
Consumer usage of a company’s Facebook page
Consumer brand involvement User-‐generated content
sensitivity Perceived valence of information
3
RESEARCH METHODS
In this chapter, the research methods that were used to collect and analyse the data are presented. The form and execution of the survey are described extensively in paragraph 3.1, followed by explaining the measurement of the used constructs in paragraph 3.2. An explanation of the formative scales is given in paragraph 3.3. The sample is described in paragraph 3.4 and the choices between a reflective or a formative scale are explained in paragraph 3.5. The final section, paragraph 3.6, is thoroughly describing the conducted analyses. The data is collected from Dutch students who have a Facebook profile and are, at least ‘Liking’ one company or brand on Facebook. The methods for the data collection as well as the construct measurements are explained. This chapter ends with the plan of analysis. 3.1 Survey
The data collection took place by means of a questionnaire. The questionnaire was a self-administered survey, distributed in two ways. The first method was through sending an email to a sample of Dutch students with a link to the online questionnaire. The second method was by placing hyperlinks to the questionnaire at different students Facebook profiles. Three hundred of undergraduate and graduate Dutch students were sent an email invitation from the author of this study, with a short description of the study, information about confidentiality and a link to the survey. Two reminder emails were sent to those who had not responded in the first place. The survey was hosted on Thesistools (www.thesistools.com), an online survey-hosting site and was fielded in December 2011. The survey distributed to this sample consisted of measures for Consumer usage of a company’s Facebook page, Consumer involvement with the brand, Perceived user-generated-content, User-generated-content sensitivity, Facebook Intensity, Perceived valence of information and Electronic word-of-mouth. To get a clear overview of the sample, some introducing questions about demographics are asked in the questionnaire as well.
3.2 Measurement of Constructs
latent variable result in changing values of all reflective indicators (Christophersen and Konradt, 2010). High correlations between the indicators are likely and can be understood as a criterion for high internal consistency. The formative measurement is in the opposite direction of the causal relationship between the latent variable and the manifest indicators. In this situation, the indicators cause the latent variable (Christophersen and Konradt, 2010). There might be some high correlations between the formative indicators but this is not expected. The factor analysis and Cronbachs Alpha are not applicable for evaluating a formative measure as they are for a reflective measurement (Christophersen and Konradt, 2010). A reflective scale can give different outcomes than a formative scale because the construct is measured in a different way and perspective as declared above. For that reason it is decided that the following constructs are measured by both a formative and a reflective scale: Consumer usage of a company’s Facebook page, Facebook Intensity and User-generated-content are measured using both reflective and formative scales. For an example of the used questionnaire, see appendix 1. The scale items, the corresponding sources (literature) and the Cronbach’s alpha’s (α) are shown in Table 11. The scales are reliable when the
Cronbach´s alpha´s are higher than 0.6, which indicates the generally accepted sufficient level of reliability (Malhotra, 2006). The level of Consumer usage of a company’s Facebook page is measured in two ways: the first method is based on the study of Cooper and Schindler (2008) and is build up around a formative scale. The participant is asked to answer three questions about the usage level of the company’ s Facebook page on an 8-point scale ranging from (1) more than one time a day to (8) infrequently or never. This construct is named: Consumer usage of a company’s Facebook page (F). The second way is a reflective method to measure the level of consumer usage of a company’s Facebook page and is based on the prior work of Hur et al., (2011), Jang et al., (2008) and Algesheimer et al., (2005). The participant is asked to rate three statements about the usage of a company’s Facebook page on a 7-points Likert scale, the Cronbach’s Alpha is .786 which is sufficient (>.6). This construct is named: Consumers usage of a company’s Facebook page (R). The level of Consumer brand involvement is measured using the existing work of Zaichkowsky (1985) and Brakus et al., (2009) using their brand involvement scale. The participant is asked to rate six statements about the company or brand on a 7-points Likert scale (Cronbach’s Alpha 830). Prior work of Ellison et al., (2007) is used to measure the Facebook Intensity level. The Facebook Intensity scale was created in order to obtain a better measure of Facebook usage than frequency or
duration indices. This measure includes two self-reported assessments of Facebook behavior, designed to measure the extent to which the participant was actively engaged in Facebook activities: the number of Facebook ‘friends’ and the amount of time spent on Facebook on a typical day. Both are added together to create a formative scale representing the Facebook Intensity level. This construct is named: Facebook Intensity (F). In addition, this measure includes a series of 7-points Likert scale attitudinal questions designed to measure the extent to which the participant was emotionally connected to Facebook and the extent to which Facebook was integrated into the daily activities of the participants (Cronbach’s Alpha .850). This construct is named Facebook Intensity (R). To measure the Electronic word-of-mouth construct, the three-item scale proposed by Verhoef et al., (2002); Reichheld (2006) and Hur et al., (2011) is used. The participants are asked to answer three statements about Electronic word-of-mouth on a 7-points Likert scale (α = .862). To capture the perceived valence of information exchanged, a scale is used based on the conceptualization put forth by Berger (1987) and Adjei et al., (2010) which measures the general positive or negative tone of messages and/or replies. The participants are asked to answer six statements on 7-points semantic differential scale (Cronbach’s Alpha .733). The work of Zaichkowsky (1994) is used to measure User-generated content sensitivity. In this study, they used the revised personal involvement inventory theory. Zaichkowsky (1994) measured involvement using ten items that should be rated on a 7-points scale. This metric is chosen, ‘because it measures both cognitive and affective involvement, and is therefore able to give a comprehensive view on personal involvement with user-generated content’ (Zaichkowsk, 1994). The participant is asked to answer 10 statements on a 7-points semantic differential scale (Cronbach’s Alpha
.932). Perceived user-generated content will be measured using the work of Cooper and Schinlder (2008) and represents a reflective scale. The participant is asked to answer three questions about the perceiving of UGC on an 8-point scale ranging from (1) more than one time a day to (8) infrequently or never.
TABLE 1 Constructs and operationalization
Construct Operational definition/measures Literature Cronbach’s Alpha Consumer usage of a
company’s Facebook page (F)A
1.How often do you visit this company’s/brand Facebook page? 2.How often do you post messages and or reactions at this company/brand Facebook page?
Cooper and Schindler (2008)
3.How often do you receive messages from this company/brand on your personal Facebook page?
Consumer usage of a company’s Facebook page (R)B
1. I will visit this company’s/brand Facebook page continuously. 2. I will exchange information and opinions with the members of this company’s/brands Facebook page. 3. I will collect information through this company’s/brands Facebook page.
Hur et al., (2011), Jang et al.,(2008) & Algesheimer et al.,(2005)
.786
Consumer brand
involvement 1.Unimportant to me - Important to me 2. Of no concern to me - Of concern to me 3. Relevant to me - Irrelevant to mer 4. Means a lot to me - Means nothing to mer
5. Useless to me - useful to me
6. Significant to me - Insignificant to mer
Zaichkowsky (1985) & Brakus et al., (2009)
.830
Facebook Intensity (F) 1.About how many friends do you have at Facebook?
1 = 10 or less, 2 = 11–50,3 = 51–100,4 = 101–150, 5 = 151–200, 6 = 201–250, 7 = 251–300,8 = 301–400, 9 = more than 400 2. In the past week, on average,
approximately how many minutes per day have you spent on Facebook?
1 = less than 10, 2 = 10–30, 3 = 31–60, 4 = 1–2 hours, 5 = 2–3 hours, 6 = more than 3 hours
Ellison et al., (2007) Formative scale
Facebook Intensity (R)B Please answer the following statements:
1. Facebook is part of my everyday activity.
2. I am proud to tell people I’m on Facebook.
3. Facebook has become part of my daily routine.
4. I feel out of touch when I haven’t logged onto Facebook for a while. 5. I feel I am part of the Facebook community. 6. I would be sorry if Facebook shut down.
Ellison et al., (2007) .850
Electronic word-of-mouthB 1. I often tell others about the company/
brand.
2. I recommend the company/brand to others.
3. I will leave positive comments about the company/ brand on community sites.
Verhoef et al., (2002) , Reichheld (2006) and Hur et al., (2011) .862 Perceived valence of
informationC I would say, that collectively the messages I received from the company/ brand on
Facebook are:
1. Positive ……… NegativeR
2. Pleasing ………... DispleasingR
3. Upsetting……….. Not upsetting
Berger (1987) and Adjei et al., (2010)
I would say, that collectively the replies and messages I read from the other company’s/brands Facebook members are: 1. Positive ……… NegativeR 2. Pleasing ………... DispleasingR
3. Upsetting……….. Not upsetting
User-generated content
SensitivityC To me user-generated content is: 1.Important ………... unimportantR
2. Boring………... interesting 3. Relevant………... irrelevantR
4. Exciting………... unexcitingR
5. Means nothing………... means a lot to me
6. Appealing………... unappealingR 7. Fascinating………... mundaneR
8. Worthless………... valuable 9. Involving………... uninvolvingR 10. Not needed………... needed
Zaichkowsky (1994) .932
Perceived user-generated contentA
1. How often do you read ideas from other consumers on the company’s Facebook page?
2. How often do you read creations from other consumers on the company’s/brand Facebook page?
3. How often do you read product suggestions from other consumers on the company’s/brand Facebook page?
Cooper and Schindler (2008)
Formative scale
RItem was reverse-coded
A 8-Point scale ranging from 1 = More than one time a day to 8 = Infrequently or never B7-point likert scale ranging from 1 = strongly disagree to 7 = strongly agree
C7-point semantic differential scale
3.3 Formative Scales
TABLE 2 Correlation Matrices Consumer usage of a company’s FB page (F)
Facebook Visit Post Messages Receive Messages
Facebook Visit 1,000 ,426** ,000
Post Messages ,426** 1,000 ,010
Received Messages ,000 ,010 1,000
Perceived user-generated content
Ideas Creations Suggestions
Ideas 1,000 ,674** ,492**
Creations ,674** 1,000 ,681**
Suggestions ,492** ,681** 1,000
**. Correlation is significant at the 0.01 level (2-tailed).
For the construct consumer usage of a company’s Facebook page (F), the correlation coefficients are partly accepted. However, because of the correlation coefficient of .426, it is better to continue the analysis with the reflective scale. For the second construct, Facebook Intensity (F), the correlation level is acceptable with a correlation coefficient of, 284. The third formative scale, Perceived user-generated content, showed three high levels of correlations compared to the acceptance level of .35. Therefore, it is decided to take one variable for this construct and use that value for further calculations and analyses2. The chosen variable is: ‘How often do you read creations from other consumers on the
company’s/brand Facebook page?’ This question is chosen because it has the most general
and broad assumption of the perceived user-generated content in it. The question is asking broadly about creations where the other two questions are more selective measuring ideas and product suggestions.
3.4 Sampling
The population used in this study are Dutch graduated and under graduated students. The target population are all the Dutch students who have a registered profile at the online social networking platform Facebook and have at least ‘liked’ one company or brand at Facebook. It is difficult to make a reliable estimation of the population due to the fact that there are no studies and figures published about the amount of average company and brand likes for students. However, due to the huge amount of Facebook users and the growing amount of
companies and brands on Facebook, the population is expected to be very large. For that reason, a sample (subgroup) of the population will be selected for this study. A convenience sample was used and the respondents had a time period of three weeks to submit their survey. The sample size (N=174) consists of 58% males and 42% females, the average age is 23.46 (SD 4.3) and almost 90% has finished or is following higher education (HBO, WO). On average, in the sample size, most of the students are using Facebook for 0.5 to 2 hours (55%) a day and 70% has between the 250 to 500 friends. The students like on average 1-3 companies or brands on Facebook (43%), followed by 20.7% liking 4-6 companies or brands. 3.5 Usage formative or reflective scale
The constructs: Consumer usage of a company’s Facebook page and Facebook Intensity have two scales: a formative and a reflective one. First, a single regression analysis is executed to determine which scale provides the best explanation of variance. The scale with the strongest explanation of variance is used for the multiple regression analysis. The results of the regression analysis showed that for the construct: Consumer usage of a company’s Facebook page, the reflective scale is more powerful with an explained variance of 21% versus an explained variance of 10.9% using the formative scale. For that reason the reflective scale will be used for the multiple regression analysis and the calculation of the interactions. The results of the regression analysis showed that for the construct Facebook Intensity, the reflective scale is more powerful with an explained variance of 54% versus an explained variance of 9% using the formative scale. For that reason the reflective scale will be used for the multiple regression analysis and the calculation of the interactions.
3.6 Plan of Analysis
In this section the statistics tools that are used to analyse the data are discussed. In short, the statistical analyses that will be used in order to test the proposed hypotheses are described. The data analysis will be performed using the statistical software program SPSS.
3.6.1 Simple Regression
CI = β0 + β1CU+
ε
3.6.2 Multiple Regression and Interaction terms
To test the other hypotheses, a multiple regression analysis is conducted. The dependent variable is the same as in the simple regression model: Consumer brand involvement (CI). The independent variable is the Consumer usage of a company’s Facebook page (R). In addition, there are five moderating variables in this multiple regression model: (1) Facebook Intensity (R), (2) Electronic word-of-mouth, (3) Perceived valence of information, (4) User-generated content sensitivity and (5) Perceived user-User-generated content. To measure the moderating effects, five new variables are created, called the interaction terms. The regression model will test the direct effect of the moderators and the interaction effects. In this study, only the interaction effect will be discussed. The equation of the multiple regression line is as follows:
CI = β0 + β1CU+ β2FIB + β3Ewom + β4PV + β5UGCS + β7PUGC + β8CU*FI+ β9
CU*E-wom+ β10CU*PV+ β11CU*UGCS+ β12CU*PUGC+
ε
CU = Consumer usages of a company’s Facebook page (R) FI = Facebook Intensity (R)
E-wom = Electronic word-of-mouth PV = Perceived valence of information UGCS = User-generated content sensitivity PUGC = Perceived user-generated content
3.6.3 Multicollinearity
tolerance score should be higher than .1 and the Variance of Inflation should be lower than 10. Because of the mean-centered variables it can be concluded that the variables have a low degree of multicollinearity, the VIF scores are all < 2.958 and the tolerance scores are >.799.
3.6.4 General assumptions
4
RESULTS
In this chapter the results of the regression analyses are presented and discussed. The formulated hypotheses are tested and the results are used to accept or reject the hypotheses. 4.1 Single Regression
In this section, the result of the single regression analysis is discussed. The relation between the independent variable Consumer usage of a company’s Facebook page (R) with the dependent variable Consumer brand involvement is tested. The single regression model shows that the model is significant with an F-value of 45.745 (p=.000). Support is found for H1 since
the intensity level of the consumer usage of a company’s Facebook page, significantly increases Consumer brand involvement (β=.342, t=6.764,p <.001). The amount of variance explained by consumer usage of a company’s Facebook page is 21%.
4.2 Multiple Regression
In this section, the results of the multiple regression analysis are discussed. The relation between the independent variable: Consumer usage of a company’s Facebook page (R), and the moderating effects of Facebook Intensity (R), Electronic word-of-mouth, Perceived valence of information, User-generated content sensitivity and Perceived user-generated content with the dependent variable Consumer brand involvement are tested. The results are used to adopt or reject the proposed hypotheses. The multiple regression table 3, shows that the model is significant with an F-value of 6,082 (p=.000). The amount of variance explained by the independent variable and the interaction terms is 29.2%. For the second time, (first with the single regression model) there is support found for H1, since the intensity level of the
consumer usage of a company’s Facebook page (R), significantly increases Consumer brand involvement (β=.298, t=2,710,p <.01).
In addition, the other five hypotheses were tested and the results are displayed in Table 3. The results show that a higher Facebook Intensity level is not significantly influencing the positive effect from consumer usage of a company’s Facebook page on the level of consumer brand involvement. For that reason H2 is rejected. In addition, there is not enough significant
evidence to accept H3. Electronic word-of-mouth, initiated by the consumer himself, does not
moderating effect from the positive Perceived valence of information on the positive relation between consumer usage of a company’s Facebook page and consumer brand involvement is rejected. There was no significant evidence to accept H4. H5, regarding User-generated
content sensitivity is rejected too. There was not enough significant evidence to accept the fifth hypothesis. A significant relation between the interaction term of Perceived user-generated content and the dependent variable Consumer brand involvement is found, however, the relationship appears to be negative (β=-.181, t=-2,405,p <.05). Thus, when the perceived amount of user-generated content increases, the level of Consumer brand involvement decreases. Therefore H6 is also rejected.
TABLE 3 Multiple Regression
β t
Consumer Usage (R) ,298 2,710**
Facebook Intensity (R) ,064 ,807
E-WOM ,183 1,717***
Perceived valence of information -,015 -,185
UGC - sensitivity ,116 1,444
Perceived UGC ,077 1,086
FB Intensity * CU (R) -,024 -,255
E-WOM * CU (R) ,003 ,037
Perceived Valence of info * CU (R) ,016 ,168 UGCSc * CU (R) ,089 1,034 PUGCc * CU (R) -,181 -2,405* R2 ,292 Adjusted R2 ,244
a. Dependent variable: Consumer Involvement with the brand b. All the independent variables are mean-centered
5
DISCUSSION
The goal of this study was to discover whether there is a significant relation between consumer usage of a company’s Facebook page and the level of consumer brand involvement. Five constructs: (1) Facebook Intensity, (2) Electronic word-of-mouth, (3) Perceived valence of information, (4) User-generated content sensitivity and (5) Perceived user-generated content were proposed to have a positive moderating effect on the relationship between the consumer usage of a company’s Facebook page and the level of consumer brand involvement. The following problem statement was formulated:
● Problem Statement: What is the effect of the intensity level of the usage of a
company’s or brands Facebook page on the level of the consumer brand involvement and what are the moderating effects of E-WOM, Perceived UGC, UGC sensitivity, Facebook Intensity and Perceived valence of information?
This study started with an extensive literature study to find out what was already known and written about the dependent and independent variables and the used moderators. Based on the findings in the literature study and findings from the empirical study, this chapter will answer the problem statement. Finally the limitations of this study are given and the recommendations for further research are discussed.
5.1 Conclusion main effect
The problem statement can be divided into two parts, the first part is about the direct relation between the independent and the dependent variable, the second part describes the moderating effects. The first part of the problem statement is quite similar to H1. The following answer
well. Probably the main reason lays in the fact, that involvement with a product or service comprises personal relevance, interest, or importance (Slater and Armstrong, 2010) and this is exactly what is stimulated by the company’s Facebook usage. Consumers are attracted to a (corporate) Facebook page because of their personal relevance, interest and importance, it is their real-life counterpart (Schau et al., 2009). However, one important point, which needs to be discussed in this conclusion, is the causality problem. In this study, the causal relationship is proposed between consumer usage of a company’s or brand Facebook page and the level of consumer brand involvement. The results have shown that there is a significant relation between these two constructs. However, the causal relation between the independent and the dependent variable might also be the other way around. In that way, higher brand involvement will lead to more usage of a company’s or brands Facebook page. It seems reasonable that when a consumer’s level of brand involvement is high, he will make more use of the company’s or brands Facebook page than a consumer with a lower brand involvement level. Presumably because of the fact that the company or brand is more in the consumers mind and the consumer is more attracted to the brand or company.
5.2 Conclusions moderators
The second part of the problem statement addresses the five moderating effects. H2 to H6 were
to state that these three constructs have a moderating effect. The E-wom construct is not positively moderating the relation in contrast to the prediction based on the theory from Dellarocas and Narayan (2006). Social Media has potential for strong brand building, innovative product development and better quality assurance (Dellarocas, 2003), these where arguments to propose that more E-wom (initiated by the consumer himself), should lead to higher brand involvement, even when the postings were not positive (Trusov et al., 2010). Previously proven results, as more brand loyalty, increase in sales and brand awareness, are indicators that E-wom has an influence on consumer behavior and outcomes (Dellarocas and Narayan 2006), however, on Facebook this effect is not directly moderating the relation between the dependent and the independent variable. When a consumer is posting more messages about a brand or company, it does not directly mean that the level of brand involvement is higher compared with a consumer with less comments and posts. One reason for the rejection of H3 regarding, E-wom initiated by the consumer himself, could be that,
consumers are not triggered (enough) to posts comments and to participate in conversations about the brand. So, consumers with a high brand involvement level, might not be triggered to post positive content, while consumers with a lower brand involvement level are more triggered because of the more negative distinctive coding in memory (Adjei et al., 2010). The Perceived valence of information does not have a moderating effect. This could be declared by the theory from Trusov et al., (2010) which stated that the effects from negative postings are also beneficial for the company. Consumers might, in accordance with this theory, not mention, or experience a difference between positive and negative posts at the corporate Facebook page. Striking, is the result that user-generated content sensitivity does not have a moderating effect on the relation between the dependent and independent variables. Consumers do evaluate user-generated content in a different manner, but probably, the differences are to small to show some significant results. In contrast to H6, the perceived
5.3 Limitations and Recommendations
Unfortunately, the conditions for this study were not perfect. One of the limitations is the used sample. Because of time limitations, a limited amount of respondents could be reached. However the fact that the respondents were randomly chosen, most of the respondents were around the age of 23 and were following or had finished higher education. This is not surprising, because most of the respondents were close to the author of this study. The results are not very differentiated for that reason. Further research can be interesting with a larger and more differentiated sample. Especially with a more distributed age and education level. Furthermore, interesting for future studies is to focus at one specific company or brand. If the respondents Like the same brand or company, the results are more valid and reliable. When a future study is conducted in collaboration with a company or brand, this information can be gathered in a more convenient way. Without collaboration it is unfortunately not possible to gather data from one specific company or brand on Facebook. When one company or brand is used, postings can be coded to make the study more reliable. The respondents reading, commenting and measuring the same posts, which make the results more generalizable. Splitting up Facebook Intensity into a more detailed construct is interesting as well. It is interesting to know for each respondent what his motivations are to use Facebook and especially corporate Facebook pages. An additional interesting point for further study is to split the study into service and product company’s or brands. Slater and Armstrong (2010) stated that involvement is typically discussed in terms of 'products and services', suggesting that the terms are similar, in spite of the apparent difference between them, as noted by previous studies. The results from Slater and Armstrong (2010) support the notion that involvement in services is distinct from involvement in products and therefore should be used in a different approach. Emerging from recent academic marketing literature in the past decade is the acceptance that services are different from products, and, as such, research has and should continue to consider them separately to contribute to future knowledge.
