Social ties on SNS and how they influence the perception of social advertising when using friends as reference

Social ties on SNS and how they influence the perception of social advertising when using friends as reference.

Master Thesis

Author: Daniel Bodera

Student number: 5974208 Supervisor: Joris Demmers

Abstract

The rise of Internet, followed by intense popularization of social network sites has changed the way consumers react between each other, towards products and towards advertising. It has also changed the way companies advertise their products and do research. This Master’s thesis studies the social ties on SNS and how they influence the perception of social advertising when using friends as a reference. Review of existing literature has shown a need to examine how people react in this newly formed online environment. After examining current studies on the topic, hypotheses were created in relation to the research question. The sample size consisted of 196 valid results. Respondents were presented with a scenario about a Facebook profile and later on asked to answer questions in a survey. There were 8 different scenarios and two types of products: digital camera representing search goods and cough medicine representing credence goods. Participants were asked to evaluate their intention to rely on an advice about the digital camera or cough medicine from a weak or a strong tie source. The study finds that the consumers of either type of product category are more likely to rely on a social ad from a strong tie, rather than a weak tie reference. Furthermore, despite of the strength of the tie, perceived expertise of a social tie will positively influence the intention to rely on a social ad reference.

Table of Contents:

1. Introduction 4

2. Literature Review 720

3. Theoretical Framework 18

4. Methodology 212

5. Results and Analysis

2632

6. Conclusions and Discussion 31

7. Limitations and Directions for Future Research

3342

References

3547

Appendix I: Questionnaire

4052

Appendix II: Tables 44

Appendix III: Profiles 48

1. Introduction

The world we live in today travels at extremely fast-growing speed - people are educated, computers are a common feature in nearly every household and Internet gives access to unlimited amount of information. Today, the online display advertising is the fastest growing advertising segment, reaching $25 billion in 2012 (Nielsen, A., 2011). Recently born, Social Network Sites (SNS) have brought even larger arena for exchange of information, interests, business and pleasure. SNS are the most visited web 2.0, where individuals construct a personal profile within a bounded system and share information with friends (Boyd & Ellison, 2007). The largest and most well-known SNS platform was launched in 2004, called Facebook (Wasserman, 2012)

SNS can provide a new and potentially powerful type of social advertisement through a “friend” (Chang, Chen, Tan, 2012). A lot of research has been produced throughout recent years, concentrated on social networks and more specifically social advertising on SNS. Many authors investigated the way participants act and behave on SNS, their personal connections with friends, attitudes towards companies and advertisements. Companies followed the trend as well trying to keep business profiles up-to-date in order to reflect the growing movement. Marketers saw an opportunity to advertise their products and services on SNS, which turned out to be a very efficient and cost-effective approach. Boyd and Ellison argue that recent SNS has the power to enable companies to create advertisements that best serve their promotional objectives, due to large number of options SNS has for individuals and organizations.

Word of mouth (WOM) has been a very important term in marketing and sales for many years allowing people to communicate and interact between each other (Li and Hitt 2008). More specifically, WOM of a SNS is combining the traditional and modern word of mouth in a way that is allowing users to electronically communicate with friends and relatives, but at the same time have real physical contact at different occasions. Companies nowadays should refocus their strategies to this new trend, have a solid understanding of the new media opportunities and what makes them so attractive to consumers (Hennig-Thurau et al., 2010).

The strength of relationships between users of social networking sites lies on a continuum ranging from strong to weak tie(Obal, Burtch, Kunz 2012). Weak- ties such as relationships with colleagues, for example, are used for information seeking behaviours while strong- ties, related to family and friends, are used mostly in decision-making behaviour (Brown and Reingen, 1987; Granovetter, 1973). Researchers should give significant effort, reconsider the effect of the two different types of ties and examine in more detail how consumers behave in this new environment (Obal, Burtch, Kunz 2012). More specifically, research is needed to define how consumers determine the usefulness of social ties on SNS and the information they offer. This Master’s thesis examines the contexts where consumers rely on information from social ties in order to make consumption decisions about goods and services.

After the initial enthusiasm for growth from researchers and marketers within the advertising industry, studies in the past years begun to indicate unexpected low sales on SNS (Chang, Chen, Tan, 2012). Relatively new, these platforms still have a lot to uncover and explore. Deeper insights might lead to more effective advertising strategies. This Master’s thesis will build on prior existing literature and try to enrich it by focusing on examining social ties and how they influence the perception of social advertising when using friends as reference. Offering a research, designed with scenarios analysis, this paper will focus on two categories of advertised products, namely – search and credence goods. For a search product, the author used an example of digital camera, while for a credence product – cough medicine. Additionally, expertise will be used to check how users will react towards a source having expertise and experience as opposed to a source without.

This Master’s research thesis will make several contributions by addressing the attitudes and behaviour of people towards social advertising. The author begins with going through the stages of development of marketing – from its roots to its digital evolvement nowadays. Furthermore, he develops a conceptual framework presenting the structure of the current studies, which will eventually lead to investigate the social ties and how they influence the perception of social advertising when using friends as reference. The author suggest that consumers will rely more on an advice from a strong, rather than a weak-tie source no matter the type of product or service they are

interested in. Furthermore, he tries to find out if consumers of search goods will be more likely to rely on weak-ties since these products are easier to evaluate. This study goes even deeper by checking if expertise is present and will individuals have more intention to rely on social advertising on SNS. Finally this master thesis enlarges the existing literature on social network sites and advertising, trying to investigate how social ties influence the perception of social advertisements.

2. Literature Review

2.1. Past and present of advertising

"Advertising is the ability to sense, interpret . . . to put the very heartthrobs of a business into type, paper and ink" (Leo Burnett).

Regardless of its extreme popularity during the last fifty years, advertising has existed long before. Its current, most basic description states that advertising is an activity, which brings a product or a service to the attention of current or potential customers by means of signs, brochures, commercials, direct mailings, e-mail messages or social network sites (Perner, 2014). In the constantly evolving time people live nowadays, advertising has become essential part of doing business and selling products and services.

The origins of advertising lie thousands of years in the past. One of the first known methods that come closer to what advertising is now was an outdoor display, usually an eye-catching sign painted on the wall of a building (Mogel, 1993). There are plenty of similar signs, discovered by archeologists, dating back from ancient Rome and Pompeii. Three thousand years ago Papyrus sheets were used in Thebes, Egypt for announcing the reward for return of runaway slaves. The first dated real advertisement was in the form of stenciled inscriptions, found on earthen bricks prepared by the Babylonians (Mogel, 1993).

Advertising was established in a market-driven system and developed in a capitalistic, free enterprise market economy. In the past, mass production and traditional advertising were essential tools of doing business. Urbanization, expansion, transportation, communication, and technology all facilitated and contributed to the use and growth of advertising. In 2014, advertising is seen in an evolutionary perspective and used in a modern industrialized society. It is firmly imbedded as a business function in our society with deeply rooted economic and cultural foundations.

In recent business terms, advertising is a part of the marketing product mix, specifically taking share of the promotion part, next to place, price and product. It allows companies to introduce, promote, test and evaluate goods and services through traditional mass media or new digital media channels in order to reach higher sales 7

volumes. All these activities have a purpose to persuade the receiver to take either immediate or future action in his buying or decision making behavior (Richards & Curran, 2002). Traditional advertising refers to the messages sent through newspapers, direct mail, radio, TV and other places, while online advertising suggests communication trough Internet, desktop or mobile devices. There are no specific rules or regulations guiding companies which type of advertising is better or more efficient. Rather, it is a very subjective substance, depending on the kind of business, size, industry, strategy, budget and a target that is set to reach (Bezjian-Avery, Calder, & Iacobucci, 1998).

2.2. Digital Advertising

In the last fifty years, world has witnessed a rapid growth in the number of methods that marketers employ to reach their customers. The rapidly changing and evolving technology has an additional push on the development of this business field. With the use of the Internet as an advertising medium in the 1990’s, web pages became the new medium for advertising efforts. Advantages are numerous, but most important reasons to use Internet advertising are the easiness of reaching huge amount of people with minimum efforts, the rather low budgets and very fast speed in comparison with traditional advertisement. Furthermore, Internet advertising augmented the traditional set of communication tools such as the television, radio, print and outdoor that marketers had used for the past 50 to 100 years (Lehmann, Winer, 1991).

The large expenditure on marketing activities has been generating high interest in recent years from small to large-scale enterprises. Large companies spend tens of thousands of dollars monthly, while small businesses have much limited spending. Regardless of this division the facts are obvious: In 2013 Internet Advertisement has reached $20.1 billion revenues – a historic high for half-year. It has grown with 18% for the same time compared to 2012 and continues to speed up with such an incredible rapidity (IDC, 2014).

Setting apart the intense popularization and usage of digital advertisement, consumers are becoming more and more careful and selective when deciding what they want to see. According to McCoy, Everard, Polak & Galletta (2007) Internet advertisements and their features usually interfere with retention and organization of site content. The

authors further conclude that pop-ups and pop-unders are considered to be more interfering to viewers than in-line ads. Users nowadays seem to prefer to focus their attention to the core of their activity and not divert it to take additional steps to close the pop-up or pop-under windows (McCoy, Everard, Polak & Galletta, 2007). For the purpose of this study the author assumes that users will have more negative attitudes towards a site having display advertisements as compared to a web site without digital advertisement displays.

2.3. Social Network Sites (SNS)

With the recent intense development and popularization of Internet technologies, new forms of Internet advertising have gained a lot of attention, both from marketers as well as academicians, namely Social Network Sites (SNS) advertising. The term ‘social networks’ existed before the Internet became commercially available, yet it was only the development of software programs such as Facebook (FB), Myspace and LinkedIn that made it more popular and commonly used (Hadija, Barnes & Hair, 2012).

SNS are platforms where individuals can construct a public profile within a bounded system (Boyd & Ellison, 2007). Social media content is created by participatory communication where users create and share various information about themselves. Users perform various actions through applications – they can send texts, images, audio and video messages. The majority of the operations performed do not require any specific knowledge, making it accessible for a large variety of people all over the world – from developed to third world countries (Akar & Topcu, 2011).

Due to the extremely far-reaching popularity and the large number of users, SNS have become a new medium for advertising and reaching customers directly (Kelley, Kerr & Drennan, 2010). Online social networks are able to offer free membership services largely due to advertising published on their sites, which is the most important income generator for any online social network (Hadija, Barnes & Hair, 2012). Moreover, SNS possess extensive and complex information about their users such as gender, education, interests, age etc., which can be used for targeting purposes. They can use information from the users’ profiles to correctly place the ad only on the pages of users, who fit that target demographics (McDonald & Cranor, 2010). Due to the large number of people that can be reached using only one advertising medium and the

relatively easy way to use SNS by both advertisers and consumers, advertising through social media is often quoted as one of the quickest and most inexpensive ways of reaching large number of potential clients. Nevertheless, firms using SNS to advertise their products and services need a careful marketing strategy planning and understanding of their customer’s behaviour and needs in this new domain (Curran, Graham and Temple, 2011).

2.4. SNS and Digital Advertising

Currently, social media activity is one of the most frequent while browsing the Internet (Meuter, McCabe & Curran, 2013). Users of SNS become more professional and advanced, create clear representations of their relationships with people they are connected with and give extended descriptions of what they do, like and prefer (Boyd and Ellison, 2007). These connections are often used as channels for exchange of information – either within group of friends or seen by third parties if the settings of the users allow it. They also establish certain relationships with other people, businesses, products or organizations in order to express their identities and self-expression (Sun, 2009). The most popular way of advertising on SNS is through banner ads, also known as display advertisements. They are paid and considered significantly efficient and effective due to various reasons: regardless of measured click-through rates, banner ads still create a favourable attitude toward the ad due to repeated exposure, consumers engage with banner ads more than TV or outdoor ads and banner ads have the ability to hyper-target by age, lifestyle, and geography better than most other mediums (Graydon & Clarc, 2003). Another way for marketers and companies to get in touch with customers on a SNS are interest groups or company profiles. Facebook, for instance, besides personal profiles, created Facebook Pages where organizations, communities, institutions, groups or individuals can create a page to build closer relationship with their audience or customers. Now, not only can you connect with your favourite artists and businesses, but you also can show your friends what you care about and recommend by adding Pages to your personal profile (Facebook.com/pages/create.php). Users do not have to be added as a friend, but simply have to click ‘follow’ or ‘like’ their page to become a ‘fan.’ Each time the advertiser posts an update, it will appear on all fans’ home pages. In the studies of Curran, Graham and Temple (2011) one of the biggest advantages of this method is that unless a person ‘unlike’ the business Facebook page or remove it from interest

lists, they will continue to receive the updates. In this manner, the businesses have even more opportunities to broaden their audience. Each time an individual adds the page to ‘interest list’, ‘likes’ or ‘becomes a friend’, it will appear on their profile where all their friends will see it.

2.4.1 Advantages of SNS advertisement

One of the most often spoken characteristics of effectiveness supporting digital advertising is that Internet develops capabilities that extend the function of advertising way more effectively than the traditional media is capable to accomplish. The growth in digital accessibility technology has had such influence on digital advertising and how people react towards it that the customers are not any longer passive receivers of the advertisement messages, but want to have more control over what they see and hear from the marketers (Hadija, Barnes,Hair, 2012). This interactivity became very important in the new medium. On one hand it facilitates great customer insights research opportunities yet customers have the option to watch TV programmes through digital interactive TV where they can skip the commercial break or they can choose to listen to the radio podcasts without having to hear the ads either. People can install special software, which minimises advert pop-ups while browsing the Internet or alter their email boxes so that the SPAM mails are automatically redirected to junk folder (Curran, Graham and Temple, 2001). Because of all these new feathers digital advertising offers over traditional advertising, current consumers are more cautious and careful about what they see and believe. In order to be more effective, marketers had to think about other, more innovative and attuned ways to reach the customer. The emergence of SNS filled those gaps and offered new ways to identify the potential target groups and to pass on the advertisement message.

As mentioned before SNS users share a lot of important information regarding their demographic characteristics, personal and professional interests, education and personal connections to other individuals. This allows targeting the ad specifically to their users’ preferences based on the information they expose about themselves on their pages. Targeting can make a significant difference regarding the amount of traffic on a specific SNS marketers try to promote. Moreover, viewers have a negative perception of the banner ads. They vary from annoying and irrelevant to more specific ads, personalized to individual’s desires. For example, sudiesstudies of Guha, Cheng and Francis (2010) on the online advertising systems conclude that demographics,

sexual preference, location and interests are the common aspects used in targeted advertising on Facebook. In addition, if an advertiser has a Facebook Page or application, they can use more advanced targeting features such as targeting ‘fans’ of the page, friends of ‘fans’ or people who are not yet ‘fan’ of the page or application. To sum up, ads on SNS, like Facebook, are considered to be effective for two main reasons. Firstly, they offer the ability to segment users and reach a specific target market. Secondly, they employ user’s own friends to enhance the impact of the ads. SNS allow advertisers to be cost-efficient through a choice of paying either by clicks or impressions. This enables a company to create advertisements which can best serve their promotion objectives (Boyd & Ellison, 2007). SNS advertisement offers to consumers a new way of looking at commercial than the already familiar ways of digital advertising. This research paper therefore focuses on SNS advertisement deeper and brings an additional view on the topic.

2.5. Word-of-Mouth (WOM)

Social Advertising is described as a marketing activity, which uses information about consumer’s social networks in order to target ads and provide personalized social signals (Bakshy et al, 2012). Marketers use Social Advertising in order to find the peers in the customer’ SNS that are affiliated with the brand, product or organization being advertised (Tucker, 2012). Moreover, Social Advertising displays social context about the ‘friends’ within the SNS who are affiliated with a particular advertising entity. They also employ Social Ads to spread attitudes and behaviours in the similar ways as word-of-mouth (WOM) and “viral marketing” (Bakshy, Eckles, Yan and Rosenn, 2012).

WOM is a very conventional way of communication between people where they meet physically and exchange information between each other orally. There are plenty research theories available on this ancient way of information transfer, but are they in line with the modern up-to-date type called eWOM and are they both equally influential? eWOM has been defined as “any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet” (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). eWOM of a SNS is combining the traditional and modern word of mouth in a way that is allowing users to electronically communicate

with friends and relatives, but at the same time have real physical contact at different occasions yet not only as it also applies to virtual connections. Therefore, companies nowadays should focus more and have a solid understanding of the new media opportunities and what makes them so attractive to consumers in order to catch their attention and affect their behaviour (Hennig-Thurau et al., 2010).

Meuter, McCabe & Curran (2013) found that interpersonal eWOM shared between friends is more influential than any of the various forms of eWOM. Furthermore, authors emphasize that the nature of eWOM channel is important and not the volume of recommendations received. As an example they found that positive recommendations from six Facebook friends wasn’t more significant that the one reviewed by three (Meuter, McCabe & Curran, 2013). These results imply the effectiveness of eWOM as an advertising medium to reach consumers and the impact an individual message or a so called “like” has on a decision-making process or purchasing behaviour.

SNS have transformed the ways consumers interact between each other by focusing on social exchange and engagement (Chu & Choi, 2011). By networking with their peers, they socialize with existing friends, make new ones, exchange information and experience about brands, obtain product-related information and make purchasing decisions. Since the contacts people have in SNS are embedded in their personal off-line network as well, consumers tend to perceive them as more credible and trustworthy than traditional advertising, online banners or unknown sources (Chu & Kim, 2011). eWOM trough SNS is therefore considered a powerful tool that consumers nowadays use acquire the information they need. It is worth mentioning here that eWOM can take place through various means and channels like blogs, review sites and emails. The freedom of the Internet source and its geographic and time constraints allows eWOM communication to spread incredibly quickly on a global sphere (Hennig-Thurau et al., 2004)

For the purpose of this Master’s, the author will focus on eWOM as a method of communication when looking for information about a product or service. This method of communication, used in a combination with modern SNS platforms offers large scale of possibilities to marketers that deserve attention and analysis. This way organization can improve the means by which they advertise their products and

services. SNS are being used to share ideas, recommendations and advices to help consumers strengthen and extend their connections with a wider range of people via Internet (Kumar et al., 2010; Trusov, Bucklin, & Pauwels, 2009)

When talking about eWOM and SNS, the author needs to highlight their relation to the human tie strength theory and its links to the matter of this research. Strong network ties generally have been found to exert greater influence than weak network ties (Brown & Reingen, 1987). On the other hand, weak- ties are critical as they allow information to flow rapidly and reach a much broader audience (Granovetter, 1973). Brown and Reingen (1987) examined the effect of social tie strength on the referral flows and found that weak- ties played a substantial connection in facilitating eWOM referral streams. Strong- ties on the other hand were perceived as influential in consumers’ decision making. The tie strength theory and its relation to this Master’s thesis will be presented in the following section of the research paper.

2.6. SNS and different types of social ties

A relationship between two individuals can be described in one of three ways: a ‘weak--tie’, a ‘strong- tie’ and a’ latent- tie’ and can be categorised based on the frequency of contact (Granovetter, 1973). According to Marsden & Campbell (1984), a combination of various factors distinguishing weak versus strong- ties. The authors further describe such factors as frequency of contact, duration of the association, intimacy of the tie, provision of reciprocal services etc. In later studies Granovetter (1983) changed the categorization of interpersonal ties based upon the subjective ‘type’ of social relationship and define friends, family and neighbours as a strong- tie, where friends of friends and acquaintances are the weak- ties. Another categorization was used in the research of Brown and Reingen (1987) where the social ties were based on the importance attached to the relationship. In the same studyies authors found that information seeking is more likely to be present via weak- ties source than strong ones.

Core ties have the potential to be highly influential in decision-making and exposure to ideas, issues, and opinions (Donath 2007). Previous research suggests that weak- ties such as with colleagues are used for information seeking behaviours while strong- ties related to family and friends, are used mostly in decision-making behaviour (Brown and Reingen, 1987; Granovetter, 1973). Moreover, the authors in

their studies extended this view in few important directions that are relevant to marketing. First, they found that active information seeking (e.g. seeking out product reviews on a website, initiating a conversation to obtain information) is more likely to occur via weak- tie sources than strong- tie sources. Second, advice received via strong- ties is more likely to be used in decision- making (e.g., the act of using information for a final purchase decision) than advice obtained via weak- ties. Third, when a consumer finds themselves endowed with both strong and weak- ties that are likely to contain information relevant to their decision, the consumer is more likely to refer to the strong- tie for consumption advice (Brown and Reingen 1987).

A conclusion can be drawn here that the various ties and their level of strengths can lead to diverse access of numerous resources. People’s daily lives, work, personal connections can be important decision-making influence pools, which shouldn’t be neglected by companies and marketers. With the existing new SNS platforms, the usefulness of strong and weak tiesweak-ties gives even more explanations on consumer behaviour. This new advertising medium can be very useful to look at the connections people have with close friends and relatives but also the huge opportunity it gives to have a contact with total strangers.

2.7. Expertise

Expertise refers to the consumer’s perception of the authority of an individual’s statements, where this individual is perceived to be an expert possessing essential or specialty skills, competency and knowledge regarding a product or service (Pamela, Lynn, 1990). The conceptualization of expertise on SNS can be seen as a high–low contrast based on their source characteristics: perceived credibility and perceived similarity (Wilson and Scherrell, 1993).

Ohanian (1990) explains that expertise, which may also be referred to as competence, expertness or qualification, is a key dimension of source credibility. He further explains that expertise is determined by experience, knowledge, qualification and skilfulness. In an SNS setting, users have the possibility to easily filter through large amount of information, find and use the most relevant and needed data for their purposes. Such a tool is important for users of SNS because it allows them to look and identify other users who might be useful to them in various situations (Ellison et al. 2007). Employment, education, activities and interests are common characteristics

that can be found on a SNS profile page and can provide cues of high or low expertise.

Constant et al. (1996) found that the perceived usefulness of an advice provided by a weak- tie source is predicted by the sources expertise and experience. When making a consumption decision, recipient will be more likely to be guided by an advice obtained by a weak- tie source with a combination of expertise (Trusov et al. 2009). It is therefore worth highlighting that if a person on an SNS platform has a level of experience or expertise that is visible to other users, individuals will most likely have intention to rely to his specialization and will trust the source for their future buying behaviors. This indirect eWOM is very interesting, offering a new way to organizations and marketers to custom make social advertising.

2.8. Types of goods and services

For the purpose of this research study, types of goods and services should be described and classified. The evaluation complexity of how goods and services are categorized differs: some of them are easy to choose, purchase, consume and evaluate, while others can be unclear and numinous (Obal et al 2011). Generally they can be located a continuum ranging from easy to difficult. Depending on the level of information needed and available to the consumers about specific good or service, they can be classified in three different groups: search, experience or credence (Darby&Karni 1973, Nelson 1970). Search goods are items that are easily evaluated before purchase and information about them is available and can be acquired from everywhere and by everyone. Examples of search goods are products such as clothing and furniture (Zeithami & Bitner, 2000). The second group of the classification – experience goods and services – rely on familiarity and practice characteristics, because their intangible nature prevents consumers from evaluating their quality until the moment of utilization. Examples of experience goods and services are vacations, telecommunication services and restaurants (Brush & Artz, 1999). Finally, the third category is called credence goods and services. They require specialized knowledge, which makes it difficult to the client to evaluate their quality even after purchase and consumption. Credence goods and services are financial investments such as insurances, paid educational services and health products such as medicines (Lovelock, 2001). As far as the first two categories are easy to measure, credence goods require more in-depth evaluations. The value of vitamin supplements for

instance is difficult to determine, even after they have been consumed, as one cannot easily define whether they would have grown sick had they not taken vitamins (Obal et al, 2011).

Internet and its extreme popularization in recent years have also influenced the goods and services and their categorization along the continuum. Alexander & Colgate (1998) highlight that Internet serves as a new distribution channel for search goods, known as well standardized goods, and services. On the other side of the continuum, where information and expertise are essential factor, Internet serves as an information pool, rather than a distribution channel (Laing et al 2002).

For the purpose of this paper we will only focus on the two extremes of the continuum. Search and credence goods will be researched and compared in detail as to find out how social network sites and social ties influence our intention to rely on a product or service. Do people trust more their strong- ties when it comes to credence goods and their weak ties when it comes to search goods? Several scenarios will be investigated to evaluate how the nature of the good or service or the nature of the referral relationship impacts once decision-making process.

3. Theoretical framework 2.9 Theoretical Framework

This Master’s thesis literature review has focused so far on social advertising and the important platform it offers for companies and marketers to introduce,introduce,

promote and endorse products and services. Besides the numerous advantages the author has presented so far, many questions remain yet unanswered. Further research is needed to confirm the actual effectiveness and user’s perception of an advertisement on SNS. This research paper focuses on how users react on preferences by their social ties and if this affects their intention to trust a certain product or service. Search and credence products are used as reference products.

Dating back in time Granovetter (1973) sets up the argument of social ties claiming that one’s acquaintances, called weak ties, are less likely to be socially involved with one another. Controversially, we are more involved with our close circle of family and friends, which we call our strong- ties. We tend to build trust, exchange information easier and share experiences and interests rather with our strong, as opposed to our weak- ties. Existing research shows that consumers rely more on strong- ties when seeking for information about credence products (Obal et al 2011).

With the growing popularity of social network sites, recent research has focused on implementing social ties in it. Donath and Boyed (2004) suggest that social networks such as Facebook increasessocial networks such as Facebook increase the number of weak- ties. This phenomenon occurs mostly because of the maintenance easiness of such relations on network. It takes much less time, effort, and it is much cheaper than it would be otherwise. According to Haythorntwaite (2005) SNS provide technical means for activating weak- ties. SNS have made it possible to ‘get in touch’ with a friend from childhood, which normally would not be possible because of the certain barriers like distance, lack of time, money, etc. The lower barriers on SNS are explained to be the key reason for activating the absent tie and that the users of SNS will look to identify other users who might be useful in some capacity (Ellison, 2007). In a very basic and general categorization, weak- ties are used for information seeking

behaviours while strong- ties are used mostly in decision-making behaviour (Brown and Reingen, 1987; Granovetter, 1973).

When in need of an guidance or a recommendation about any type of good, the author believes consumers will talk easier to their strong- ties, if available, since one has already build a trust, faith and confidence with them. Therefore, the author believes that no matter the product type or service, people that are in our closest circle and we consider most sincere to us will give us the best possible and honest advice for any type of products or services. Therefore,

H1: Consumers will rely more on strong, rather than on a weak tie source

Depending on the type of product people are interested in, they can rely on different tie strength. Information about search goods is usually easy to find and consumers tend to form their purchasing decisions more comfortable and laid-back. Credence goods on the other hand, require more in-depth search, time intensity and specialty advice. Depending on the type of advertised product, weak-ties can turn out to be significantly influencing the purchase decision of the final consumer (Obal, Burtch and Kunz 2012). For example, information about search products is easily accessible and extensive, which leads consumers to be more tranquil and less intense in their information-gathering and decision-making process. On the other hand, credence products are more complex they require more information seeking and often -different opinions. The usefulness of the advice from weak-ties is judged on their expertise and experience (Constant, 1996). Moreover, the authors also found out that the advice from weak-ties on SNS based on experience and similarity are less relevant when considering experience products or credence products, because of they are much more difficult to evaluate. For that reason, consumers of search products seem to rely more on SNS weak tie advice (Obal, Burtch and Kunz, 2012).

H2: In an absence of strong-tie, a weak-tie social ad will have bigger influence for search products in comparison to credence products.

Expertise is a key dimension of source credibility (Ohanian, 1990). People naturally believe doctors, teachers and scientists only because it is understood that they have the expertise of a specific area of interests. Whether one knows the professional or not, when the expertise is acknowledged and proven, the trust is already existingdoes

already exist by default to a certain extend. Therefore the author expects that recipients will be more likely to use advice in a consumption decision if this advice is taken from a social tie, which has expertise in the assigned field. Thus the effect of expertise on intention to rely is not moderated by tie strength, giving us Hypothesis 2:

H2: There is a positive relation between perceived expertise and intention to rely.

The development and integration of Internet in people’s lives nowadays serves as an information pool tool (Lang et al 2002). However, how can a consumer recognize the right information among the incredible amount of data available? Do people trust everything they read or are they able to filter and recognize the trust in profiles? Previously mentioned, current technology has given access to extremely large amount of information to people. Consumers of search goods benefit greatly from it because of its easiness to acquire, low or no-cost and the diversity of products, reviews and opinions. But how it is with credence products and services, where consumers cannot evaluate their effect even after the purchase has taken place? The author believes that in situations where consumers need to form an opinion about credence good, their beliefs and intention to rely will strongly be influenced by “expertise”. Having the level of know-how is expected to add the special bit a consumer needs to believe in the qualities of credence goods. Therefore:

H3: The effect of expertise on intention to rely is moderated by type of product, in the sense that it is stronger for credence goods than for search goods.

Given the above-mentioned hypothesis, the author of this Master thesis provides the following conceptual model, representing all hypothesized statements in ffig. 1:

INTENTION TO RRELAY H2 H3 H1B H1A

Formatted: Font: Times New Roman, 12 pt

Formatted: Font: Times New Roman, 8 pt

Formatted: Font: Times New Roman, 8 pt

Formatted: Font: Times New Roman, 8 pt

Formatted: Font: Times New Roman Formatted: Font: 6 pt

Formatted: Centered

Formatted: Font: Times New Roman, 6 pt

Formatted: Centered

Formatted: Font: Times New Roman, 6 pt

Formatted: Centered

Formatted: Font: Times New Roman, 6 pt

Formatted: Centered

Formatted: Font: Times New Roman, 6 pt

4. Methodology

In the following section, the methodology used for this master’s thesis will be presented and described in detail. Justification of the chosen method and the reasons why specific instruments were selected over another will be offered.

4.1. Research Design and Sample

There are many different ways to execute a research study depending on the topic, audience, available resources and time restrictions. The very basic division research type is divided into three different categories, namely inductive, deductive and abductive approach. This research paper use a deductive approach starting from the facts that are already known, developing a hypothesis from the existing theories and later on testing the predicted spectacles in these proposed hypothesisthese proposed hypotheses (Malhotra & Birks, 2007). Moreover, experimental method of research is most suitable for this research topic, questions and objectives. An experiment is defined as manipulating an independent variable to see how it affects a dependent one, while also controlling the effect of additional extraneous variables (Burns & Bush 2002). Independent variables are described as those variables over which the researcher has control and wishes to manipulate, while dependent once are those that have little or no direct effect and control. Meyerson and Tryon (2003) confirm that data gathered from online experiments is reliable, valid and reasonably representative, highlighting additional advantages as cost effectiveness and efficiency. In addition, the environment, in which the research will take place, favors such way of conducting primary data best. Using online experiment as preferred research approach offers wide range of advantages in terms of cost, time and improved quality. Even more exciting is the fact that online experiment perfectly fits directly to the online environment of social network sites that is the main topic of this Master’s thesis. The study is quantitative in nature as data was collected through surveys, which involve a widely used research strategy within business and management research (Saunders, 2009). The survey was conducted through the use of questionnaires.

The survey was administered with the help of the online software Qualtrics. Such software allowed the author of this Master’s thesis to reach large amount of people, located in various countries and having different nationalities. Respondents were

anonymous and could also choose not to mention their nationality. There are few drawbacks of using such an electronic program to administrate a research questionnaire: there is a lack of control from the side of the administrator and no security that the answers are correct and honest. There can be a self-selection bias, meaning that some individuals can be more attracted to fill in the questionnaire than others (Thompson et al., 2003). Yet, this type of questionnaire and program to administer it was best suitable for the Master’s thesis research, allowing the author to reach large amount of respondents in a relatively quick time and requiring minimum financial investment.

Sample

The selection of site and sample are critical important decisions that affect the viability of the whole study (Marshall and Rossman, 1989). Facebook was chosen as a platform for empirical verification of this Master’s thesis research. The respondents of the survey for this research were randomly selected from the friends, relatives and friends of friends circle. A research with people and organizations that one already knows or is familiar with is very advisable and useful approach (Saunders et al, 2009). Personal connections and interactions are very important to select the participants, bring trust and convince them to take part of the experiment project. The author tried to give a better insight of the project where possible, convinced the participants beforehand that there wouldn’t be sensitive and/or offensive questions that might reduce trust and influence the behavior of the respondent and the rest of the questions, which needed to be answered. 245 people were presented the scenarios and the follow-up questionnaire through Facebook and email. Out of them there were 196 valid and usable questionnaires for the research.

4.2. Experimental manipulations

To test the hypothesis, the author executed 2(strong/weak- tie) x 2(high/low expertise) x 2(search/credence products) experiment in a context of social network site. At the beginning of the questionnaire, the respondents were presented with one of 8 different scenarios where tie strength, product type and expertise were manipulated in order to test the hypothesis. After the respondents had read the scenario and reviewed the presented Facebook profile page, a survey followed with questions related to the previously presented situation. The set of questions intended to evaluate the reference

to the person presented in Facebook profile page in terms of how likely they are to rely to this person’s advice and his expertise as well as previously mentioned product category. All respondents were asked to provide demographic information such as age, nationality, gender, education, usage frequency of the SNS and their attitude towards social advertising. Finally, there was a section with questions, asking the respondents how realistic the scenario was and if the participant would rate their own expertise in the specified product category.

4.3 Variables

Social tie – independent variable

“Weak- tie” and a “strong- tie” can be categorised based on the frequency of contact (Granovetter, 1973) or can be based upon the subjective ‘type’ of social relationship (Granovetter, 1983) where friends, family and neighbours as a strong tie, where friends of friends and acquaintances are the weak- ties. The tie strength was manipulated in the survey and provided in the introduction of the questionnaire. In case of “weak- tie” manipulation, the respondent was asked to imagine that person presented on a Facebook profile used in the survey is an acquaintance and the respondent has little contact in a physical world but only interacts time to time in a virtual environment of Facebook. The “strong- tie”, referred to a person presented as a close friend, and the respondent had not only a frequent connection on Facebook but additionally in the real life interactions.

Expertise – independent variable

Expertise is determined by experience, knowledge, qualification and skillfulness (Ohanian, 1999). Employment, education, interested and performed activities of the respondents were manipulated by the author on the profile page, presented in the scenario, in order to provide cues of high or low expertise. Specifically, profile pages with high expertise of the tie were related to the product in question, indicating that the presented Facebook profile of an individual shows that he is a pharmacist if the presented product is a cough medicine. For the search product – digital camera, the

profile would indicate that the person in this profile is interested in photography and have adequate education and experience. Low expertise profiles did not indicate

employment or interests that are related in any way to the product. To check if

perceived expertise is in the accordance with manipulated one, control questions were asked in the questionnaire about this variable.

Product type – independent variable

In order to choose products for evaluation in the research, the author used previous existing research literature that supported classification of the three product/service categories: search, experience and credence goods (Nelson 1970, Darby and Karni 1973, Hsieh et al 2005). Two products were chosen to represent the two types of extremes when describing goods and services. Digital camera was an example of a search good, while cough medicine was an example of credence good. In order to simplify the research process the middle category, experience goods, was excluded. These products were chosen based on previous study that provisions the above classification (Obal, Burtch and Kunz, 2012) and with accordance to studies of Hseih (2005) where the author intended to determine what products consumers would see as search, experience and credence products.

Intention to rely – dependent variable

In the questionnaire dependent variable was measured with accordance to studies of Obal, Burtch and Kunz (2012) and consisted of 5 questions which were all measured on a 7-point Likert scale (1-not at all…, 7- totally agree) :

I would rely on the reference to this person to make a purchase decision For the purchase decision, I would ignore the reference to this person I would definitely click on the ad because of the reference to this person I would consider the reference to this person as very helpful for my purchase

This information would be important for my purchase decision.

4.4. Data collection and data analysis

The survey was distributed for 8 weeks in the autumn of 2013. The respondents were approached in several ways. One of them was posting on the author’s Facebook Page a request to participate in the research and fill in the questionnaire or to use his personal network of friends and family. The same request was sent through private messages or posts on walls of Facebook friends. They were also asked to forward the survey to their friends and family either via Facebook, Twitter or E-emails. Facebook Interest Groups were also used to post the survey link if the posting was possible. The

electronic mailing system of University of Amsterdam was also utilized to recruit respondents. After one week of first distribution a Facebook message or an email was sent to people to remind them about it.

The data was analyzed using SPSS 18. Various types of statistical analysis were performed to obtain results from the survey.

5. Results and Analysis

In the coming chapter of this Master's thesis, the results of all conducted surveys in the experiment will be summarized and presented to the reader. The findings are going to be grouped and presented in the same sequence as the proposed hypothesis in the Literature review section. By analyzing the collected empirical data, the initially designed hypothesis will be justified or rejected.

245 questionnaires were sent out to participants in the research. Out of them 196 were valid or fully finished and usable for data analysis (N=196). Complete information in tables about the participants and their descriptive statistics can be found in Appendix II. In summary, out of the 196 respondents, there were 101 female and 95 male. As previously said the selection of the sample was random, however the author used his personal connection, friends and friends of friends to get the sample size growing. The age of the respondents ranged from 20 to 56 years old. The most represented group was aged between 20 and 35 years old. Among the participants of the questionnaire, there were many different nationalities, namely 24. The largest three countries represented were The Netherlands (87), Poland (45) and Germany (21). The educational background of the sample was also very diverse, presenting all educational possibilities. The highest share of the participants had completed Bachelor education (77), followed by a Master degree (66), Community College (27) and the least number of respondents had only high school education (21).

As previously mentioned, two products were used to represent the categories of search and credence goods. The selection was based on the previously done research by Hsieh et al. (2005) that states a digital camera to represent an example of a search good and a cough medicine to represent an example of credence good. All acquired valid and usable data was put into SPSS statistical software and analyzed in detail

troughthrough various methods in order to justify or rejected the initially proposed hypothesis.

To test the variables used in the questionnaire, the author of this Master’s thesis conducted correlation analysis, which are presented in the table 1 below:

Table 1

Correlations, means and standard deviation of model variables

mean sd 1 2 3

1. Tie ,49 0,50

2. Product ,51 0,50 -,061

3. ExperiseExpertise ,47 0,50 -,070 ,079

4. Intension to relay 4,1367 1,12 ,320** -,055 ,349** Note.**. Correlation is significant at the 0.01 level (2-tailed).

The above table of correlations shows us the independent variables – social tie, type of produce and expertise - in relation to the intention to rely on social ad. Based on Pearson correlation coefficient r, it provides information on the strength and direction of the liner relationship between a pair of variable and can take any value between -1 and +1. A coefficient 0 suggests that there is no relation and a coefficient of -1 or +1 assumes perfect correlation (Bryman and Cramer, 2009).

To test for hypothesis 1, 2 and 3, 3-way factorial ANOVA analysis was conducted to the primary collected data. This was the most suitable method for this purpose since there was more thenthan one existing independent variable and each was measuring using different participants between groups (Field 2009).

Table 2

Levene's Test of Equality of Error Variancesa for dependent variable Intention to rely

F df1 df2 Sig.

0,463 7 188 0,86

Note. Tests the null hypothesis that the error variance of the dependent variable is equal across groups.

Table 3

Factorial analysis of variance for dependent variable: Intension to relay

Source df Mean Square F Sig.

(A) Tie 1 36,184 39,296 0,00* (B) Experise 1 43,566 47,313 0,00* (C) Product 1 0,166 0,181 0,671 A * B 1 2,469 2,681 0,103 A * C 1 0,168 0,182 0,67 B * C 1 0,493 0,536 0,465 A* B* C 1 2,475 2,688 0,103

Error (within groups) 188 0,921

Note. * p<0.01

Leven’s test (Table 2) of equality tells us that there are significant differences between the group variances and therefore there is a non-significant result. More specifically, the social ties, types of product and level of expertise are significantly different.

Looking at the F-ratio from the above results, we can see that only “social ties” F (1, 188) = 39,296, p<.001 and “expertise” F(1, 188) = 47,313, p<.001have a significant effect on the dependent variable “intention to rely”. “Type of product” has low F-ratio (1, 188) =, 181, p=.671>.001 and no effect on “intention to rely”. In other words, if we ignore the strength of the tie and level of expertise, the product category did not influence the intention to rely.

Looking at the Mean measurements of the variables, the author can conclude that the strong tie has bigger influence on the intention to rely

Additionally, all possible combinations of the variables were tested to see if there is a significance level of interaction between either of them, however there was no valuable result found.

Post hoc and contrast analysis could have been useful to find significant results for some of the hypothesis, however they are performed when there are more than two indicators of specific variable. In the case of this Master’s thesis the strong and weak tie could give the author a better insight of the behavior of the respondents. However the limitation to only these two categories made the analysis unable to perform. The above results show the following outcomes regarding the hypothesis:

Hypothesis H1 A: Results show that there is no significant difference to consumers regarding the type of product. This means that no matter the product category – search or credence – users will prefer to rely on a strong-tie reference in comparison to a weak-tie reference when looking at social ads. More specifically, regardless if the product is a digital camera or a cough medicine, users on SNS would more rely of reference from their close friends or family, rather thenthan someone they do not consider part of their close circle of network.

Based on the above, Hypothesis H1 A is accepted.

Hypothesis H1 B: Similar to the above statement, the results show that product type has no influence for the weak tie reference to social ad. The F-ratio of relation analysis between Social Tie and Type of Product on the intention to rely on social ad has no significant effect F (1, 188) = .182, p= .670 < .001.

Based on the above said, Hypothesis H1 B is rejected.

Hypothesis H2: The acquired results show that the intention to rely on a social ad, regardless of the tie is positively influenced by the perceived expertise of the tie. In other words, when expertise is present, the strength of the tie has no additional influence on the intention to rely. The F-ratio of relation analysis between Social Tie and Expertise on the intention to rely on social ad has no significant effect F (1, 188) = 2.681, p= .103 > .001

Based on the above said Hypothesis H2 is supported.

H3: The results show that the relation between type of product, level of expertise and tie strength has no significant effect on the intention to rely on social ad with F (1, 188) = 2.688, p= .103 > .001. Specifically, no matter search or credence good, the

expertise will remain the same on intention to rely on social ad from a weak tie reference.

Based on the above said H3 is rejected.

6. Conclusion and Discussion

The main purpose of this master’s thesis is to analyze social ties on Facebook and how they influence the perception of social advertising when using friends as reference. At the initial stage of the research, the author created 8 different scenarios, which were presented at the beginning of the study, followed by set of questions in a

survey. In the following section of this master’s thesis, important findings, supported

propositions and identified patterns will be summarized and concluded. Summing-up the final words of this research project, the author will add his interpretations from observing the sample and its behavior towards social advertising.

This Master’s thesis hypothesized and found empirical support that individuals rely on their strong-tie connections for reference regardless of the type of product they are considering or looking at. Furthermore the results show that product type has no influence either for the weak-tie reference to social ad and if strong-tie is existing and available, the respondents will prefer it. In particular, the author found evidence that the intention to rely on a social ad, regardless of the tie is positively influenced by the perceived expertise of the tie. This implies that when expertise is present and constant at someone’s profile characteristic, the strength of an individual with that tie has no additional influence on the intention to rely. Because of this, consumers will prefer to rely on an advice from an expert, rather than any of their close family and friends. Last, but not least, the results do not support the hypothesis and reject the assumption that the strength of the tie has an additional influence on the intention to rely when considering an individual with expertise.

According to Obal et al,. 2012 customers of search products tend to rely on advice from a weak-tie sources rather than consumers of experience or credence products. Based on the collected data, this Master’s thesis contradicts the above literature, showing a different view of reference in context of social advertising. Research findings of this Master’s thesis emphasize that strong-ties on social network profiles have bigger influence on the intention to rely as a reference for products and services. The type of product, whether search or credence good, is not showing significant difference to the behavior and decision-making of the user. This conclusion can be used as guidance to the marketers and companies, helping them to focus more on

strong, rather thenthan weak-ties, when advertising their products or services on SNS. Organization shouldn’t rely only on weak-ties when getting in touch with the customers. Rather, individuals nowadays prefer their strong-ties and will trust them better when forming their purchase decisions. The massive invasion of SNS platforms and the enormous popularity among the world’s population has made consumers to be more careful and alert, filtering troughthrough their connections and valuing their most sincere once, even if they are considering to buy a search good.

Expertise is a key dimension of source credibility (Ohanian, 1990). According to Constant et al (1996), the usefulness of advice provided by weak-ties source is predicted by their expertise and experience. This Master’s thesis research concludes its results and confirms the existing literature that expertise and experience are important qualities, regardless of the strength of the tie. If experience is obvious in a user’s profile, the viewer will pay attention to their references, not matter if this user is someone from his close friends or a person he really have contact with. The support of this hypothesis suggests that businesses should focus on finding a way to advertise through people that are experts and have knowledge in the field of the advertised product or service on SNS.

Current literature supports the theory that SNS allows consumers to determine credibility based on surrogate information like expertise and similarity, which they can obtain through user profiles (Ohanian 1990). In this case individuals will be less of need to refer to strong-tie sources (Ohanian 1990). Furthermore, it is scientifically proven that experience and credence goods are more ambiguous and difficult to evaluate therefore individuals rely more on strong-tie advice when making a consumption decision for these goods in an online context (Brown and Reingen 1987). This master's thesis study did not support the literature that expertise is more influential for credence as opposed to search goods. Therefore, in the practical side, not only companies producing search goods should focus on communicating with their customers through SNS. Providers of credence products and services should also be heavily invested in the most popular social networking sites, such as Facebook and Twitter, allowing them to reach mass quantity of consumers. However, marketers at the same time are to think of creative ways where they can combine expertise and experience in order to gather the trust of the consumers and form their intention to rely.

7. Limitations and Directions for Future Research

As any other research paper, this Master’s thesis has few limitations that need to be listed. The sample size of the study was quite limited with the number of N=196 valid questionnaires. While the number is not necessarily small, when divided into the eight different scenarios, the data per setting becomes smaller and less reliable. While the total sample size can theoretically lead to significant results, the more participants in the various scenarios can allow for better analysis of the situation. Furthermore, participants were selected trough the close network of the author. This has a potential to lead to similar characteristics in interests, behaviour, education etc. In order to reduce this limitation, future research can focus on a broader and more diverse sample. Regarding the product categorization, this Master’s thesis used only two product categories – search and credence goods. They were based on existing literature and were not checked and categorized by the respondents. This situation can lead to subjective results, because one type of product can mean and be categorized very differently by separate respondents.

The platform used for the study was only one – Facebook. Future research can focus on many different aspects in this direction. First, other SNS platform, different than Facebook, can lead to significantly different results. Facebook represents great diversity, but more specialized SNS can lead to more precise and interesting results. For example, LinkedinLinkedIn is an SNS platform for higher educated professional, assuming there will be a lot of people who already have expertise in various disciplines. Would they have an intention to rely on other experts? Furthermore, information about social advertising and SNS, considered in this Master’s thesis, is in a context when Facebook was mainly used on personal computer devices. With the rapid grow of technology, nowadays more and more users log into SNS through their mobile devices. Therefore, future research should also focus on examining what is the behaviour on SNS when access troughthrough a mobile portable device or other up-to-date technology, which can effectaffect the perceptions of social advertising.

SNS are incredible source of information gathering and exchange - people are able to communicate, exchange information, judge and educate themselves based on their

friend’s preferences from everywhere around the world and at minimum costs. The feature is allowing nearly everyone around the globe to use this information source at no cost and minimum effort to form opinion simply by looking at the profiles of their friends online. In order to use this platform effectively, companies should realize the strength of ties people form on SNS and address custom-made solutions to market their products and services accordingly.

References

Alexa (2011), "About Facebook (Facebook.Com): A Social Utility That Connects People, to Keep up with Friends, Upload Photos, Share Links and Videos," Retrieved April 13, 2011, from http://www.alexa.com/siteinfo/Facebook.com. Alexander, Nicholas, & Colgate, Mark. (1998). The evolution of retailer,banker and

customer relationships: A conceptual framework. International Journal of Retail and Distribution Management, 26(6), 225–236.

Allen, T. (1984). Managing the flow of technology: Technology transfer and the dissemination of technological information within the R\&D

organization. MIT Press Books, 1.

Akar, E., & Topçu, B. (2011). An examination of the factors influencing consumers' attitudes toward social media marketing. Journal of Internet Commerce, 10(1), 35-67.

Andrews, D. (2002), "Audience-Specific Online Community Design," Communications of the ACM, 45,4, 64-68.

Bakshy, E., Eckles, D., Yan, R., & Rosenn, I. (2012). Social influence in social advertising: evidence from field experiments, 146--161.

Bezjian-Avery, A., Calder, B., & Iacobucci, D. (1998). New media interactive advertising vs. traditional advertising. Journal of advertising research, 38, 23-32.

Boyd, D. M. and Ellison, N. B. 2007. Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication 13, 1, 210–230. Brown, J. and P. Reingen (1987), “Social Ties and Word-of-Mouth Referral

Behavior,” Journal of Consumer Research, 14 (3), 350–62.

Brush, T. H., & Artz, K. W. (1999). Toward a contingent resource-based theory: The impact of information asymmetry on the value of capabilities in veterinary medicine. Strategic Management Journal, 20(3), 223-250.

Bryman, A., Cramer, D. (1999). Quantitative Data Analysis with SPSS Release 8 for Windows. A Guide for Social Scientists. London: Routledge.

Bryman, A., Cramer, D. (2005). Quantitative Data Analysis with SPSS12 and 13. A Guide for Social Scientists. East Sussex Routledge.

Burns, A. C., & Bush, R. F. (2002). Marketing research: Online research applications. New Jersey: Pearson Education, Inc.

Chu, S. & Kim, Y. (2011). Determinants of consumer engagement in electronic word-of- mouth (eWOM) in social networking sites. International Journal of Advertising, 30(1), 47–75.

Chu, S. C., & Choi, S. M. (2011). Electronic word-of-mouth in social networking sites: a cross-cultural study of the United States and China. Journal of Global Marketing, 24(3), 263-281.

Constant, D., L. Sproull, and S. Kiesler (1996), “The Kindness of Strangers: The Usefulness of Electronic Weak Ties for Technical Advice,” Organization Science, 7 (2), 119–35.

Curran, K., Graham, S., & Temple, C. (2011). Advertising on Facebook.International Journal of E-Business Development, 1(1), 26-33.

Darby, M., and Karni, E. (1973), "Free Competition and the Optimal Amount of Fraud," Journal of Law and Economics, 16, 1, 67-88.

Donath J. (2007). 'Signals in social supernets', Journal of Computer-Mediated Communication, vol 13.

Donath, J. and D. Boyd (2004), “Public Displays of Connection,” BT Technology Journal, 22 (4), 71–82.

Ellison, N., C. Steinfield, and C. Lampe (2007), “The Benefits of Facebook ‘Friends:’ Social Capital and College Students’ Use of Online Social Network Sites,” Journal of Computer Mediated Communication, 12 (4), 1143–68.

Gefter A 2007, 'The difference between "real" and online is no longer clear-cut', in S Engdahl (ed.) Online Social Networking. Greenhaven Press, Detroit

Goel, S., Hofman, J. M., and Sirer, M. I. 2012. Who does what on the web: Studying web browsing behavior at scale. In Proc. 6th International Conference on Weblogs and Social Media.

Granovetter, M. (1973), “The Strength of Weak Ties,”American Journal of Sociology, 78 (6), 1360–80.

Granovetter, M. (1983), “The Strength of Weak Ties: A Network Theory Revisited,” Sociological Theory, 1, 201–33.

Graydon, S., & Clark, W. (2003). Made you look: how advertising works and why you should know. 1st edition. Toronto: Annick Press.

Guha, S., Cheng, B. and Francis, P. (2010) Challenges in Measuring Online Advertising Systems, ACM IMC 2010 Proceedings, November 1-3, 2010, Melbourne, Australia.

Homer, Pamela M., and Lynn R. Kahle (1990), "Source Expertise, Time of Source identification, and Involvement in Persuasion: An Elaborative Processing Perspective," Journal of Advertising, 19 (1), 30-39.

Field, A. P. (2009). Discovering statistics using SPSS: and sex and drugs and rock 'n' roll (3rd edition). London: Sage.