A comparative study on the effect of positive and negative user generated videos on brand engagement

(1)

A comparative study on the effect of positive and

negative user generated videos on brand

engagement

MSc in Business Administration – Digital Business Track

University of Amsterdam

Name: Karampali Evgenia

Student number: 11374691

Supervisor: J. Y. Guyt

(2)

1

Statement of Originality

This document is written by Karampali Evgenia who declares to take fully responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no source other than those mentioned in the text and its references have been used in creating it.

The faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents

.

(3)

2

Abstract

The proliferation of social media platforms has contributed to the empowerment of customers who nowadays co-create content with brands. The aim of this study is to focus on the content that is generated by users in the form of videos and how their connotation can affect the engagement of customers with the brands as well as if this relationship is moderated by whether the brands offer goods or services. It was found that brand engagement changed after users watched either a positive or negative video about the brand, with the unfavorable videos having a greater impact on engagement than positive. It was also found that customers tend to be more engaged with brands that offer services instead of goods, although this relationship does not affect the impact of different-valenced videos on engagement. Nevertheless, the outcomes of the study provide valuable information for managers when they have to deal with negative user generated content. That is to say it is important not to ignore such content but rather respond to it to moderate its unfavorable impact on brand engagement.

Keywords: social media, YouTube, engagement, valence, goods, service, videos, user

(4)

3

1. Introduction

Social media are part of people’s everyday lives and even though the majority of them have an active account on Facebook, Twitter or Instagram they are oblivious of the fact that other social media platforms existed long before these ones. The very first social media site was launched in 1997, its name was Six Degrees (Conole, Galley and Culver, 2011) and combined users profiles, friends lists and the ability to invite external friends to join the site. Since then the world has faced a proliferation of social media and new ones come to light every now and then. LinkedIn and MySpace came out in the early 2000s and were followed by Facebook in 2004, YouTube in 2005, Twitter in 2006 and many others in the years that followed. At first social media were seen as a unique alternative of entertainment (Whiting & Williams, 2013) and a new way of communication among their users but as soon as brands recognized their influential and lucrative character, they seized the opportunity to use them as a new tool to reach potential customers. Hence, the ubiquitous nature of social networking sites and their unrestricted penetration in people’s everyday lives spurred researchers to study the relationship of social media users and brands. Their findings indicated that the creation of a social media community improves the relationship of companies with their customers (Kumar,Bezawada,Rishika,Janakiraman & Kannan, 2016).

Prior to the era of social media, the relationship of customers with brands was not bidirectional as people were passive and usually just consumed content without sharing or reflecting any thoughts or opinions. This behavior was inverted by the growth of Web 2.0 and the rise of social networking sites. Customers understood that they could alter, adjust or enhance the digital content that companies posted or even create their own (Labrecque, Vor dem Esche, Mathwick, Novak and Hofacker, 2013). Nowadays users comment, post, share and publicly express their feelings about brands while trying to either influence their attitudes or even discipline them for their irresponsible actions. The above forms of content that are

(6)

5

available via social media and generated by users themselves are characterized as user-generated content (UGC).

Although, UGC is conceptually broader than eWOM, the two meanings are often mutually used in cases that UGC is brand-related (Smith et al., 2012). These blur lines are clarified in the aspect that in UGC the content is created by the users themselves while in e-WOM the content is conveyed by them. An example of UGC in YouTube (which will be the social networking platform examined in this study) will help to clarify the two ambiguous meanings. A video in YouTube which is uploaded by a user is UGC, whereas a video that is sent to a friend is characterized as e-WOM. In the case that a shared video is also generated by users, then this content is defined as both e-WOM and UGC.

Previous literature has thoroughly attempted to address the impact of UGC on brands. Tellis and Tirunillai (2012) examined the connection of UGC and stock market performance whereas others like Stephen and Galak (2012) investigated the interaction of UGC with traditional media and the impact it could have on sales. Also, many researchers have focused on UGC on social media and the underlying motivations that impel users to contribute such content in social networking sites (Toubia &Stephen, 2013). Other studies have also tried to sort different taxonomies of UGC and so numerous different classifications have arisen through the years. One of the most important refers to the connotation of the content and categorizes it according to its valence as either positive, negative or neutral. The majority of the studies considered that neutral content had no impact and focused only on the effects of the positive and the negative and so the topic of neutral content was quite vague. Over the years researchers begun to reconsider the effect of the latter and thus many contradictory opinions arose. Godes and Mayzlin (2004) supported that it does not influence television ratings and Mudambi and Schuff (2010) emphasized its ineffective character especially for experienced products. Conversely, Sonnier, Mc Alister and Rutz (2011) claimed that there is a

(7)

6

positive effect between neutral UGC and sales. On the other hand Tang, Fang and Wang (2014) in order to moderate the contradiction of previous literature they distinguished neutral content in mixed and indifferent. They supported that mixed-neutral content amplifies the effects of negative or positive UGC, while indifferent content attenuates them. In this study UGC will be investigated solely in the aspect of its influence on engagement and since the efficacy of neutral content is still not elucidated it will be excluded from the study and only positive and negative valenced content will be taken into consideration.

Besides the various taxonomies of content regarding its connotation (positive or negative) many studies have also dealt with the impact of different valenced content. Specifically numerous controversial opinions have arisen regarding the impact of positive and negative while simultaneously different theories have supported that one type of content has greater effect than the other. The asymmetric evaluation theory by Kahneman is one theory that could clarify whether positive or negative valenced content has greater effect on brand engagement. The assumption of the theory is that the “impact of a difference on a dimension is greater when the difference is evaluated as a loss rather than when the same difference is evaluated as a gain” (Tversky and Kahneman, p.1040, 2017). That is to say “the response in a change is expected to be more intense when the change is unfavorable than when it is for the better” (Tversky & Kahneman, 2017, p.1055). Applying the above theory in the case of UGC would mean that if customers view a topic with negative connotation about a brand they like, it will have a greater impact on their relationship with the company than if they had viewed a positive content about their favorable brand. A contradictory opinion was cited in the article of Ullah, Amblee, Kim and Lee (2015) where the role of emotions in consumers’ responses was emphasized. The authors stated that extreme positive reviews have a greater emotional content than their negative counterparts. They declared that positive content in social media increases virality further compared to negative content. So their findings to some extend

(8)

7

contradict those of Tversky and Kahneman in terms of whether positive or negative content is more influential and which one has bigger impact on people’s perceptions and behaviors. Berger and Milkman (2012) researched further and found that the relationship between virality and valence depends on other factors such as arousal transmission. They claimed that content that evokes high arousal emotions such as amusement, interest or anger is more viral, regardless of whether these emotions are positive or negative. In other words whether people will share or engage with a topic does not depend solely on its valence, as the two aforementioned studies have demonstrated, but rather rely on the arousal it invokes.

Apart from the differences in the connotation of UGC another worth-mentioning classification refers to the distinct forms that the content can take such images, ratings, reviews, blogs’ and forums’ posts etc (Hautz, Fuller, Hutter & Thurridl, 2014). In this study UGC will be addressed in the form of user-generated videos (UGV) uploaded in social networking sites as YouTube and more specifically the study will focus on the different effects of different-valenced videos on engagement. To put it more simply, it will investigate the impact of positive and negative UGV on brand engagement and whether this relationship is moderated by the type of industry, meaning whether a brand offers goods or services. So,

does negative UGC affect the overall engagement more than positive does, in absolute values? Hence should marketers be worried about the effects of negative UGC or does it not influence brand engagement as much as positive UGC does? Additionally, is the valence of videos the only variable that affects engagement? Or the effects are moderated by whether the video concerns a goods’ brand or a services’ brand? A thorough study will provide insights and reveal the answers to the aforementioned questions that will shed light on the importance of UGC and the strategy that is best for marketers to follow when they have to cope with negative UGC.

(9)

8

In the following chapters, previous literature will be reviewed in order to provide a more in depth analysis of the theoretical concepts mentioned above. The scrutiny of existing literature will result to a conceptual framework that will help to visualize the relationships between engagement, valence and type of industry. Furthermore, for the aforementioned questions to be answered a survey was conducted and its findings will be discussed so as to help draw a conclusion about the significance of the connotation of UGC. Finally the implications of the findings for managers will be mentioned along with limitations and directions for future research.

2. Literature Review

New products and brands emerge every year and for companies to survive in this competitive environment, they are endeavoring to enhance their customers’ engagement by exploiting the influential character of their social networking sites. Nevertheless, social media have also empowered consumers who are able to participate and involve to the creation of social content in the form of e-WOM. As Goh, Heng and Lin (p.90, 2013) mentioned “this WOM buzz is typically defined as user-generated content (UGC) and is an output of consumer’s engagement on social media”. So it is clear that the concepts of social media, UGC and engagement are interdependent and affect the relationship of consumers with the brands.

2.1 User Generated Content

Although brands have been considered as the unique creators of marketing messages and brand-related content (Hautz, Fuller, Hutter & Thurridl, 2014), in recent years the emergence of social media has altered the roles. The interactive nature of social networking sites has

(10)

9

strengthened the power of consumers who are no longer passive recipients of the content that firms generate. Nowadays customers eagerly participate and create any form of content such as blogs, posts, tweets, images and videos and become representatives of products and brands by dissipating user-generated content (Hautz, Fuller, Hutter & Thurridl, 2014). Α question to be answered is how influential can the content that is produced by users for well-known brands be. And also, how reliable can a content, which is not generated by the brand itself, be? According to Goh, Heng and Lin (2013) compared to product information provided by firms, product information provided by consumers is regarded as more credible and trustworthy due to its relatively unbiased nature. Cheong and Morisson (p.45, 2008)explained this matter further and proved that “consumers look for product information or recommendations before purchase and they tend to voice more trust in product information created by other consumers”. So undoubtedly customers can produce content that can be even more influential than the content generated by brands themselves. Considering the above, UGC can be extremely beneficial for companies if the content that is created strengthens their image. On the contrary, users create content for reasons other than compliment the brand. In many cases they want to publicly declare their dissatisfaction and disapproval and expose the brand and so UGC can unfavorably backfire for the company. So it is apparent that UGC can become either advantageous or harmful and hence affect the relationship of customers with the brand.

Whether UGC will affect the above relationship is highly depended on the evaluation process of the content. In other words, how do people judge the content they view in social media and more specifically the content that is generated by other users? The underlying reason of the significance of the way people evaluate the content is that it is highly interdependent with the final evaluation of the brand So by identifying the process that people follow to judge content it can be revealed the way they judge the brand itself. Studies have

(11)

10

proven that there are two types of information that can be provoked in the mind of customers, namely declarative and experiential and according to these two types people judge based on either their reasoning or their feelings. When people rely on facts about a brand they use declarative knowledge whereas when they depend on their personal feelings to render evaluations they use experiential information. (Schwarz, 2004). The above theory is also validated from Brakus, Schitt and Zarantonello (2009) who further affirmed that customers use their feelings to render evaluative judgments about brands when they deal with the experiential view. In general both declarative and experiential information are used by customers to assess brands but a study from Esch et. al (p.81, 2012) has elucidated that “in certain contexts the experiential information alone can have primacy over declarative information”. The interpretation of this statement is that information that appeals to emotions, in some cases, is more important than facts when people evaluate brands. As the scope of this study is to investigate the effect of different-valenced videos, that will evoke contrary emotions about brands, on engagement, the outcome is based solely on the experiential way of evaluation.

2.2 User Generated Video

As mentioned before user generated content can take many forms as blogs, tweets and reviews along with others. One of these types that has gained momentum in the digital era is videos as they combine text, moving images and sound (Ertimur and Gilly, 2012). Demangeot and Broderick (2010) also advocated this opinion as they stated that whereas text was the symbol of analog era, video became the medium of the News Media age. As Ertimur and Gilly (p.116, 2012) mentioned in their research, “the tremendous growth of online social sites such as YouTube and Vimeo, has generated new media of expression in the form of

(12)

11

user-generated videos (UGV)”. Nevertheless, even though the ever increasing power of customers has lead to the growth in number of active spectators who produce and upload content, a study by van Dijck (2009) affirmed that this increase corresponds in about 13% of the total users of social media whereas the passive users or else “lurkers” that do not engage by creating content, correspond to 33%. Even if the percentage of passive customers is almost tripled compared to that of active users, O’ Donnell et al. (2008) stated that “lurkers” are more likely to have their opinion influenced by videos compared to active users. Hence, the facts that it is the passive users that are mostly influenced by UGV, the case that they represent a good percentage of the entire population of users in YouTube and considering that people use their emotions to evaluate brands, justify the importance of further analysis on the effects of different UGV’s in the relationship of customers and brands.

As mentioned above UGV will be examined under the scope of their valence and so it is important to elucidate its meaning. Valence refers to the “behavioral activation and the degree of positive (toward) or negative (away from) emotion for a stimulus” (Seo, Lee, Do Chung and Park, p.74, 2015). In other words as Teixeira, Wedel and Pieters (p.146, 2012) mentioned, “negative emotions prompt tendencies to avoid or reject an affective stimulus whilst positive ones arouse tendencies to approach or retain it”. Another study that emphasized the importance of the connotation of content is the one made by Cheng and Chiou (2003) who noted that negative online information is more influential than positive in shaping the attitude of customers towards brands. Murphy (2008) also recognized the importance of valence when brand related information is presented as she confirmed that connotation has the power to affect a brand’s perception either favorably or adversely.

All the pre existing literature about UGV contributes to this study in two prominent ways. Firstly, it emphasizes that the connotation of content has great effect in shaping the attitudes and beliefs of people about brands. Further it indicates that one type of connotation,

(13)

12

namely either positive or negative carries more weight than the other, a theory that will be discussed later in this study.

2.3 Social Media Engagement

The definition of engagement has concerned many researchers through years. Many different conceptualizations have arisen and the majority of them converge in describing engagement as a multi-dimensional construct of three dimensions namely cognitive, affective and behavioral (Bhardwaj & Vohra, 2016; Hollebeek, Glynn & Brodie, 2014). In these three dimensions social was added, as the rise of social media and their ubiquitous nature have a great impact on customers’ engagement and cannot be overlooked. Table 1 below illustrates numerous definitions of engagement in the digital era as were formulated by several authors (Bhardwaj & Vohra, 2016).

Authors Terminology used Description

Cheung,Lee &Jin (2011)

Customer engagement in an online social platform

The level of customer’s physical, cognitive and emotional presence in connections with a particular online social platform

Cvijikj & Michahelles (2013) Online engagement

Measuring undertaken actions such as click through rates, page views etc

depending on the

possibilities offered by the platform

Mollen & Wilson (2010) Online engagement

Cognitive and affective commitment to an active relationship with the brand as personified by the website or other computer-mediated entities designed to communicate brand value

(14)

13

Wirtz et. al. (2013) Online brand community engagement

Consumer’s intrinsic motivation to interact and co-operate with community members;

Table 1

Definitions of engagement

As Bhardwaj and Vohra (p.361, 2016) mentioned in their article “Although interaction in social media platforms is described as one of the necessary pre-requisites to engagement, it is the content that the customers engage with’’. So when brands want to build engaging relationships with their customers in social media they should not only be interested in creating authentic interactions and monitoring the discussions but also concentrate their attention in the content that is being uploaded. This content as aforementioned when is generated by users can be either positive or negative. Customers that are dissatisfied with a brand “can take the route of social media to vent out their anger or negative feelings towards the brand. Such negative conversations can lead to negative engagement” (Bhardwaj & Vohra, p.362, 2016). Conversely if loyal customers grow feelings of intimacy and emotional attachment and become fans of the product or the brand, it results in customer engagement where people become advocates and co-create value with the brand. (Sashi, 2012).

All the foregoing definitions of engagement revolve around the concepts of sharing content and the involvement with the brand in either a cognitive or affective level. This study will focus only on the emotional dimension of engagement of users (affective level) and the degree that this initial engagement of customers to brands is influenced after people viewing UGV with positive or negative connotation. Therefore the extent to which the valence of UGV can affect user’s feelings towards the brands and hence their engagement. Although both positive and negative stimuli have an impact on engagement, there are many studies that

(15)

14

overemphasize the consequences of negative incentives. The phenomenon of ‘negativity bias’ describes the behavior of people to react differently in cases that they face either positive or negative stimuli. That is to say negative events evoke stronger emotional and behavioral responses than positive ones do (Baumeister et al. 2001; Rozin and Royzman, 2001). So it is hypothesized that:

H1: Negative UGV will have a greater effect on engagement than positive UGV.

2.4 Product and Service Marketing

Marketing is a conceptually broad term and so to be better understood and utilized many different classifications have arisen through the years. Some researchers made a distinction between hedonic and utilitarian goods (Wertenbroch and Dhar, 2000) whereas others assorted it to business-to-business and business-to-consumers strategies (Swani, Brown and Milne, 2014). One of the most important taxonomies that has been studied for years among researchers is the distinction between goods and services. As Vargo and Lusch (p.2, 2004) mentioned, “marketing has moved from a goods-dominant view, in which tangible output and discrete transaction were central, to a service-dominant view, in which exchange processes and relationships are central”. Τhe main differences between products and services are concluded in four different characteristics (Jackson, Neidell & Lunsford 1995). The first characteristic is the intangible nature of services which do not have a physical presence, followed by the inseparable character of production and consumption that defines them. The third feature that separates goods from services is heterogeneity which is the continuous alteration in the quality of services over time. Finally the last one is perishability which is defined as the inability of services to be stored for future use. Even though many studies refer

(16)

15

to intangibility as the major difference between services and products there are some researchers as Shostack (1977) that supported that intangibility should not be considered as a modifier but rather as a state and cannot possibly be compared with the experiences that services provide. This implies that, “a service is experienced. A service cannot be stored on a shelf, tasted or tried on for size” (Shostack, p.73, 1977). So as de Chernatony and Rilley (1999) illustrated, customers have an emotional need to be persuaded that all the intangible advantages that characterize services as for instance the experiences, will be delivered to them. All the above reveal that one of the major differences between goods and services is no other than the experiential character of the latter.

Concerning the evaluation of the two distinct industries by customers, Jackson, Neidell and Lunsford (1995) supported that the process of evaluating services is more frustrating than assessing goods. The reason is as aforesaid that services are mostly characterized by experience qualities meaning that their attributes can be understood only during or even after consumption. Contrariwise goods are characterized by search qualities which are attributes that customers can discern before purchase such as style, color, price, package etc (Zeithaml, 1981). Moreover another important element that separates the evaluation process of services to that of goods is that evaluating services is riskier than goods due to their experiential character (Murray and Schlacter, 1990) and so people rely heavily upon personal sources of information such as friends, family or other users when assessing services (Shostack, 1977). All the above indicate that the fact that people experience the consumption of services makes them to seek more information about them than they do about goods and so they are more involved with them (Bloemer & de Ruyter,2010), so it can be hypothesized that:

(17)

16

Finally, a third hypothesis could be formulated so as to identify whether the impact of different valenced videos on brand engagement would change depending on the two different kinds of industry. That is to say whether the effect on brand engagement caused by the different connotation of videos would be greater on either services brands or goods brands. This leads to the final hypothesis:

H3. The effect of the valence of videos on brand engagement will be greater for brands offering services than brands offering goods.

2.5 Conceptual Framework

UGC is a topic that has been investigated thoroughly through the years either from its positive or its negative perspective. Nonetheless the research that has been conducted on the effects of valence of user generated videos on consumers’ engagement comparatively is very limited. Simply put would positive UGV engage more customers than negative UGV would drive away? Does the type of industry and whether it deals with goods or services, matters in the effect on brand engagement? Considering the above the research question of this study is:

“Do different valenced UGV have different effect on brand engagement and is this effect moderated by the type of industry?”

(18)

17

H2 H3

Figure 1

Conceptual relationships between valence, type of industry and brand engagement

3. Research Design

3.1 Procedure

The empirical context of this study revolves around user generated videos that are uploaded in YouTube. So to comprehend the importance of UGC and the impact it could have on engagement firstly it needs to be understood the medium that this content is uploaded to. The proliferation of social media has provided customers the possibility to tell their stories about consumption of goods and services. “Consumption has a narrative nature, users become the storytellers” (Pace, p.213, 2008) and a sophisticated tool for depicting all these stories is YouTube. Pace (2008) perfectly illustrated the reason why YouTube is an optimal social media channel for the diffusion of content generated by customers but except the motivation for uploading such content it also needs to be elucidated the motivation for searching such

H1

Independent variable

Valence of videos’ content (positive or negative)

Dependent variable

Brand engagement

Moderator

Type of industry (goods or services)

(19)

18

content. Hence an important ambiguity that needs to be enlightened is the kind of information that people search for in social media and more specifically in YouTube. Many authors (Stafford, Stafford and Schkade, 2004; Shao, 2009; Khan, 2017) have tried to elaborate on the above vagueness based on User and Gratification Theory which studies the underlying motivations that impel people to use social media. One of these motives concerns the fact that people use social networking sited to search information about brands. After people seek information concerning a brand they can share it. A study that was conducted in 2012 was made by Smith, Fischer, and Yongjian (2012) found that between Facebook, YouTube and Twitter, social media users are more likely to share information about brands (either in the form of complaints or as objects of interest) on Twitter and YouTube rather than they do on Facebook. As the scope of this study is about UGVs, Twitter can be excluded from the above research as it does not support content in the form of videos and state that users tend to share UGVs with brand-related factual information more on YouTube than they do on Facebook. So the choice of YouTube as the medium of the empirical context was not random but rather was selected as previous literature has emphasized the preference of people in uploading in the particular social media channel their opinions and experiences about brands and also their preference in searching for brand related information in YouTube.

Concerning the structure of the experimental study, an exploratory, quantitative research was conducted and the research design was formulated as 3 x 2. More specifically a survey was created with UGV of two different brands. The brands were selected as to reflect the two different types of industries, namely a product brand (Pepsi) and a service brand (Shell). As the study measured the implications of the valence of videos on engagement, the survey included in total four different videos and more specifically two for each brand, a positive and a negative one. All the videos were uploaded to YouTube and were not created by the brands themselves but by actual customers. The objective of the study was to identify

(20)

19

any differences on the level of engagement before and after watching different-valenced videos and so both initial and final engagement were measured and benchmarked. When referring to ‘initial engagement’ it is meant the relationship that people had with the brands before watching any video and counterwise ‘final engagement’ describes the relationship of customers with the brands after they have viewed either a positive or a negative video. So as to estimate the initial engagement of people, respondents were asked to answer questions that revealed their preexisting relationship with the brands without them viewing any video. On the same way they were asked to respond in the same questions after watching either a positive or a negative video and so their final engagement was measured. The cascade of the questions was designed as to minimize question order biases deriving from the same people answering the same questions that measured initial and final engagement. A randomizer was used so each respondent answered questions measuring either their final engagement after watching a positive or a negative video, or their initial engagement for each brand respectively. Moreover the survey was designed to avoid demand characteristics biases (Orne, 1961) meaning the subconscious tension of participants to respond in such a way that they think will validate the hypotheses of the experiment. Hence after watching the videos, an irrelevant question followed (How many times do you use the product/service per week?) so that respondents would not understand immediately that the scope of the study is the analysis of engagement. The analysis of the aggregated results indicated whether there has been any alteration to the final engagement of users compared to the initial one after they have watched a UGV about each brand. Moreover except engagement, the awareness of each of the brands was also measured as people who are unaware of the companies could provide bias results given the fact that they would not have an engagement relationship with the brands. Each video was accompanied by a small text explaining what the respondents were about to watch

(21)

20

and also mentioning that it was content generated by customers and not by brands, as impartially as possible.

Before the final survey was distributed a pretest was conducted with a sample of 24 people where the respondents after answering questions measuring their awareness and engagement, they were also asked to characterize the video they have watched as either positive, negative or neutral. Τhe purpose of the pretest was to clarify whether the perception of the connotation of videos was objectively the same as that of the respondents. That is to say if the valence of each video was objectively considered as either positive or negative. The results indicated a disagreement about one of the videos of Pepsi as responses varied between positive and neutral and so this particular video was replaced with another. Moreover some ambiguities concerning the precision of the description above videos were clarified before the final survey was launched.

3.2 Measures

In order to measure the engagement of respondents, the definition that Hollebeek, Glynn and Brodie (2014) provided about it was taken into consideration. The authors noted that

engagement is characterized by three different dimensions namely the cognitive, the affective and the behavioral level. As this study examines the effect of positive and negative videos and how the emotions that they evoke impacts engagement, only the affective dimension of engagement was taken into consideration. The measures were taken from Hollebeek, Glynn and Brodie (2014) who in their turn consulted previous writers (Calder, Malthouse, and Schaedel, 2009; Sprott, Czellar, and Spangenberg 2009)and adjusted the measures in their study. Respondents were asked in a seven-point Likert scale from strongly disagree to strongly agree, to indicate their agreement with each item provided on the list. In total there

(22)

21

were four different items for the measurement of engagement in the affective level with Cronbach’s alpha being .839 (Table 2)

Affective level of engagement

o I feel very positive when I use the brand o Using [brand] makes me happy

o I feel good when I use [brand] o I am proud to use [brand] Τable 2: Scale items of engagement

In this study the level of engagement was measured along with awareness of people as the two concepts are strongly associated. Hoeffler and Keller (2002) stated that in the process of building brand equity there are six steps that brands have to build on. The first step is to raise awareness about the brand and the last one is achieving brand resonance or else engagement. In order for a brand to have engaged customers it has to have fulfilled all the previous five steps including brand awareness, so it is apparent that engagement and awareness are inseparably linked. The level to which respondents were aware about a brand or not was measured with a scale taken from Girard, Trapp, Pinar, Gulsoy, and Boyt, (2017). Respondents were asked in a seven-point Likert scale with Cronbach’s alpha .86, according to the items that can be seen in Table 3.

Awareness

o I recognize [brand] among other competing products o I am familiar with the products that [brand] offers

o I can quickly recognize the symbol or logo of [brand] products

Table 3

(23)

22

3.3 Sampling

The study was conducted using a non-probabilistic convenient sample as it was difficult to distribute it in a way that gives all individuals in the population the same chances of being selected. The majority of the respondents were Greek in a percentage of 78.54% and the rest of them were other nationalities (Table 10). For the dataset to be considered reliable 50 responses per condition was the reaching goal. In total 294 participants answered the survey and three of the responses were deleted as they were unfinished. Even though the range of ages was quite broad from 18 to 68 years old, the 57.4% of the sample was between 22-28 years old and the second biggest subset was in the ages of 47-53 that reached 10.7% of the total sample. Moreover the educational background of the respondents was mostly a bachelor’s or master’s degree with a percentage of 76.6%. Concerning the gender, women accounted it for approximately 63% of the total sample with the remaining being males.

4. Results

4.1 Construct reliability

In order to confirm that both scales of awareness and engagement consistently reflected the constructs they were measuring, a reliability analysis was conducted. The average Cronbach’s alpha for the engagement scale was 0.932 which indicates a high level of internal reliability. As all items were above 0.30 in the corrected item-total correlation it can be said that there is a good correlation of each item with the total scale. Moreover none of the items if was removed would result in an even higher Cronbach’s alpha so they were all kept (Table 8). The same reliability analysis was also run for awareness and the findings indicated a Cronbach’s alpha of 0.655 (Table 9). The reliability was lower than the average of 0.70 but as

(24)

23

de Vaus (2002) stated in his book there can be some random errors that do affect reliability but when these errors are small, the measure continues to remain reliable.

4.2 Measurement model assessment

H2

H3

H1

Figure 2 : Conceptual relationships between valence, type of industry and brand engagement

The three hypotheses as presented in the above framework were tested with a two-way Analysis of Covariance (Ancova). An Ancova analysis is used to examine the main and interaction effects of categorical variables on the dependent variable as in this case the effect of the valence of videos on engagement (H1), the effect of the type of industry on brand engagement (H2) and the moderating effect of the type of industry on the impact that valence has on engagement (H3). In an Ancova analysis all the aforementioned effects of the independent variables on the dependent are tested after the effects of covariates are controlled and in this study awareness was taken as the control variable.

Valence of videos’ content (positive or negative)

Brand engagement Type of industry

(25)

24

Concerning the first hypothesis and the assumption that negative valenced videos would have greater impact on brand engagement than positives would, the Ancova analysis (Table 4) revealed that there is a significant main effect of valence on engagement F(2, 284) = 19.12, p=0.000. The above result indicated that valence indeed affects final engagement but it does not reveal whether positive or negative valenced-videos have a greater impact. In order to validate H1, a Tukey post-hoc test was conducted (Table 11). The test revealed that both the differences between initial engagement and engagement after watching the positive video was statistically important, p=0.066 as also the difference between initial engagement and engagement after watching the negative video, p=0.000. As the point of the study was to determine which type of video affects more engagement in absolute values, a subtraction of means was made (Table 10). The difference of engagement of positive videos to initial engagement was 0,47 whereas the difference of engagement of negative videos to the initial engagement was -0,725. Considering the initial engagement as the zero point it can be drawn as a conclusion that the effect of negative videos on engagement in absolute values is almost doubled compared to the effect of positive user generated videos. So H1 is validated and negative UGV have greater impact on customers’ engagement than positive ones have.

The Ancova analysis further shed light on the relationship of the type of industry with brand engagement. More specifically the second hypothesis stated that brands offering services have greater effect on customers engagement than brands offering goods and the analysis revealed that indeed the type of industry has a significant main effect on engagement F(1,284) =7,917, p= 0,005 (Table 4). So as to classify whether the goods’ brands or the services’ brands have more engaged customers, the pairwise comparisons table was examined and the mean difference between them unveiled that brands which offer services have greater brand engagement. So H2 is also validated.

(26)

25

Hypothesis 3 referred to the interaction effect and whether the impact of different-valenced videos on engagement would be moderated by the type of industry. The analysis of the last hypothesis indicated that there is not a statistically significant effect with F (2,284) = .40 and p > 0.05 meaning that the type of industry is irrelevant to the effect on engagement caused by the differences in the connotation of the videos.

Furthermore the Ancova analysis revealed the impact of the moderator on the single dependent variable whereas simultaneously controlling for the effect of the control variable. More specifically it was hypothesized that the effect of the type of industry on engagement will be influenced by the degree of the level awareness of the brands. The results of the two-way Ancova analysis indicated that there is a significant effect of the type of industry on engagement after controlling for awareness F(1,284)=9,534, p=0,002 (Table 4). This finding supported that as stated in the theory, awareness of a brand plays an important role in the degree that customers develop an engaging relationship with it.

SS DF MS F Sig η2 Awareness 16,39 1 16,39 9,53 0,002 0,032 Type of industry 13,60 1 13,60 7,91 0,005 0,027 Valence 65,73 2 32,86 19,12 0,000 0,119 Type*Valence 1,39 2 0,695 0,40 0,668 0,003 Error 488,211 284 1,71 Total 4380,12 291

Significance at the p < .05 level

Table 4

(27)

26

Briefly the Ancova analysis revealed that two out of the three hypotheses were supported as can be seen from Table 5.

H1: Negative UGV will have a greater effect on engagement than positive

UGV Supported

H2: Brands that offer services have greater brand engagement than brands

that offer goods. Supported

H3: The effect of the valence of videos on brand engagement will be greater

for brands offering services than brands offering goods. Not Supported

Table 5

Overview of hypotheses acceptation

5. Discussion

People are no longer passive recipients of content created by companies but rather they like to post, share and generally co-create content along with the brands. Web 2.0 has empowered customers by “providing an outlet of self-expression” which is interpreted as UGC (Labrecque et al., p.261, 2013). This UGC can either be beneficial for companies that observe their customers praising the brand or adverse as dissatisfied customers can create negative content aiming to harm the image of the company. Many studies have addressed the issue of which content becomes ongoing WOM (Berger and Schwartz, 2011) and whether the venue (Berger and Iyengar, 2013) or the size of the audience (Berger and Barasch, 2014) affect the content people share. Nonetheless, little is known about the effect of content generated by users, on customers’ engagement. This study provides insights on the importance of different valenced videos and their impact on brand’s engagement. Moreover it sheds light on whether

(28)

27

goods’ or services’ brands have more engaged customers and if the differences in the valence of videos have a different impact on engagement based on the difference of the type of industry.

The results of the study validated two out of the three hypotheses. The first supposition concerning the connotation of videos was supported and demonstrated that negative-valenced videos have greater effect on brand engagement than the positive ones. More interestingly this impact is not just bigger in absolute values but almost twice as much as the effect of positive videos. That is to say the ratio of the influence of negative to positive videos on brand engagement is almost 2:1 meaning that UGV with negative connotation transform the engaging relationship of people with brands much more than positive videos do.

Moreover, another important finding of the study was the indication that brands offering services have greater brand engagement than the ones that offer goods. Although many researchers have highlighted the differences that distinguish goods from services, others have questioned them supporting that the boundaries between them are blurring (Levitt, 1972). This study by examining two typical examples of services’ and goods’ brands and taking into consideration the opinion of nearly 300 people of different ages and nationalities, demonstrated that there is a considerable difference among them at least in the level of engagement with their customers. Brands that offer services were found to have more engaged customers than the ones offering goods. This finding could be supported, as was aforementioned in theory by the fact that people experience the consumption of services more than that of products (Shostack, 1977) so they may feel more involved with them and thus more engaged.

The third hypothesis which concerned the degree to which the differences in the types of industry would affect the impact of the different-valenced videos on engagement, was rejected. The findings suggested that negative valenced videos influence equally customers’

(29)

28

engagement with a brand regardless of whether it offers goods or services. The rejection of the hypothesis could be justified by the fact that the study only examined one brand of each industry and so the results might not reflect the reality due to the limited sample. Future research could shed more light on whether the findings of the study are representative or not by containing a more diverse sample of brands.

6. Managerial Implications

It is common practice for companies to promote contests and campaigns that support and encourage their fans to create content and upload it in their social media. Nevertheless the findings of the study demonstrated that is important for brands not only to prompt their customers to create positive content, but also to be cautious of the negative content that people upload. This study revealed that negative-valenced UGC can have twice the influence that positive UGC has on brand engagement and so it becomes essential for companies not to overlook such content but rather respond to it. Previous literature has further supported the ramifications of ignoring negative content which can lead to a negativity spiral where negative messages generate more volume that is followed by even more negative-valenced messages (Hewett, Rand, Rust and Heerde, 2016). Hence in the case that a brand has to cope with unfavorable content it would be better not to disregard it but rather respond to it. For instance the response could either take place in the brand’s website in the form of a statement or even in its social media platforms so as to reach a broader audience.

Along these lines the significance of not neglecting negative content is also evident as potential customers that might not know the brand and see one unfavorable video about it can form an overall negative attitude towards it. This implication is corroborated by the prospect theory of Kahneman (2017) who supported that when people make evaluations intuitively as

(30)

29

it is the case of evaluating a brand after watching a video, this decision is not made in absolute values but rather is reference-dependent, meaning that the evaluation will depend on the context that it is framed. That is to say if the reference-dependent context is negative as in the case of a negative video, it is almost certain as stated by prospect theory, that people will assess negatively the brand and will not create an engaging relationship with it. Therefore companies should pay attention on the context that their brand is framed on social media and if it is unfavorable they should try to reverse it.

Another important implication for brands concerns the distinction between companies that offer services and these that offer goods and the relationship they have with their customers. The findings revealed that even if brands that offer services have more engaged customers than their counterparts, negative valenced videos impact both of the types of industry equally. Hence it would be good for both brands that offer services and brands that offer goods, to be cautious regarding unfavorable content that is uploaded by their consumers.

7. Limitations and direction for future

studies

One limitation of the study regards the use of convenient sampling, as the 78.54% of respondents were Greek and the remaining percentage was split in Dutch, German and other nationalities. Future studies could expand to a more diverse sample of respondents as different nationalities represent different cultures that could have distinct relationships and perceptions about the same brands.

Another important limitation is the fact that the study focused only in one social networking site, namely YouTube. A study by Schweidel and Moe (2014) indicated that when valence is measured it is important for the research to be done in multiple venues and not just

(31)

30

one as this could lead to a misestimate of the sentiment. Hence future studies could expand the research scope and either encounter for differences on the connotation of videos among different venues as YouTube, Facebook, Vimeo etc or present aggregated results for the effects of different-valenced videos across all social media and not just one.

One more worth-mentioning limitation stems from the fact that the findings resulted from the examination of two brands, namely Pepsi, that represented brands offering goods and Shell that portrayed brands offering services. The rejection of the third hypothesis, could be due to the limited number of brands used in the study. Future research could examine further the role of the different types of industries by including more brands from each industry.

An additional direction for future research could revolve around the different topics that, when are brought up by users, brands should initiate activity. Chen and Berger (2013) found that issues that raise controversy are more likely to be discussed among people up to a point that they will reach a certain point of discomfort. After this point people avoid talking about topics that raise disputable opinions. So future research could identify which are these negative issues that when brought up by users, they could initiate ongoing WOM and which ones would not due to the uneasiness they raise. The topics that could activate a continuing negative discussion about the brand on social media should be the ones that brands should take action on and try to moderate them.

Finally, the study measured only the impact of the differences in valence of one type of UGC, meaning videos. Deeper insights on the effect that different-valenced images or reviews or tweets etc, could have on brand engagement may help brands to better understand how to deal with unfavorable content of different forms.

(32)

31

8. References

 Aksoy, L., van Riel, A., Kandampully, J., Wirtz, J., den Ambtman, A., Bloemer, J, Kandampully, J. (2013). Managing brands and customer engagement in online brand communities. Journal of Service Management, 24(3), 223–244.

https://doi.org/10.1108/09564231311326978

 Barasch, A., & Berger, J. (2014). Broadcasting and Narrowcasting: How Audience Size Affects What People Share. Journal of Marketing Research, 51(3), 286–299. https://doi.org/10.1509/jmr.13.0238

 Baumeister, R., Bratslavsky, E., Vohs, K., & Finkenauer, C. (2001). Bad is stronger than good. Psychologist.

 Berger, J., & Iyengar, R. (2013). Communication Channels and Word of Mouth: How the Medium Shapes the Message. Journal of Consumer Research, 40(3), 567–579. https://doi.org/10.1086/671345

 Berger, J., & Milkman, K. L. (2012). What Makes Online Content Viral? Journal of

Marketing Research, 49(2), 192–205. https://doi.org/10.1509/jmr.10.0353

 Berger, J., & Schwartz, E. M. (2011). What Drives Immediate and Ongoing Word of Mouth? Journal of Marketing Research, 48(5), 869–880.

https://doi.org/10.1509/jmkr.48.5.869

 Bhardwaj, N., & Vohra A.. (2016). A Conceptual Presentation of Customer Engagement in the context of Social Media – An Emerging Market Perspective.

International Journal in Management and Social Science, 4(1), 351–366.

 Bloemer, J., & de Ruyter, K. (2010). Customer Loyalty in High and Low Involvement Service Settings: The Moderating Impact of Positive Emotions. Journal of Marketing

(33)

32

 Brakus, J. J., Schmitt, B. H., & Zarantonello, L. (2009). Brand Experience: What Is It? How Is It Measures? Does It Affect Loyalty? Journal of Marketing, 73(May), 52– 68. https://doi.org/10.1509/jmkg.73.3.52

 Calder, B. J., Malthouse, E. C., & Schaedel, U. (2009). An Experimental Study of the Relationship between Online Engagement and Advertising Effectiveness. Journal of

Interactive Marketing, 23(4), 321–331. https://doi.org/10.1016/j.intmar.2009.07.002

 Chen, Z., & Berger, J. (2013). When, Why, and How Controversy Causes Conversation. Journal of Consumer Research, 40(3), 580–593.

https://doi.org/10.1086/671465

 Cheong, H. J., & Morrison, M. A. (2008). Consumers’ reliance on product

information and recommendations found in UGC. Journal of Interactive Advertising,

8(2), 38–49. https://doi.org/10.1080/15252019.2008.10722141

 Cheung, C. M. K., & Lee, M. K. O. (2011). Customer E Ngagement in an O Nline S Ocial P Latform : a C Onceptual M Odel and S Cale D Evelopment. Thirty Second

International Conference on Information Systems, 1–8.

 Chiou, J. S., & Cheng, C. (2003). Should a company have message boards on its web sites? Journal of Interactive Marketing, 17(3), 50–61.

https://doi.org/10.1002/dir.10059

 Conole, G., Galley, R., & Culver, J. (2011). Frameworks for understanding the nature of interactions, networking, and community in a social networking site for academic practice. International Review of Research in Open and Distance Learning, 12(3), 119–138. https://doi.org/10.1111/j.1083-6101.2007.00393.x

 Cvijikj,P., & Michahelles, F. (2013). Online engagement factors on Facebook brand pages. Social Network Analysis and Mining, 3(4), 843–861.

(34)

33

 de Chernatony, L., & Riley, F. (1999). Expertsʼ view about defi ning service brands and the principles of service branding. Journal of Business Research, 2963(46), 181– 192.

 de Vaus. (2002). Surveys in social research (5th ed.). Allen & Unwin.

 Demangeot, C., & Broderick, A. J. (2010). Consumer Perceptions of Online Shopping Environments. Psychology & Marketing, 30(6), 461–469. https://doi.org/10.1002/mar  Dhar, R., & Wertenbroch, K. (2000). Consumer choice between hedonic and

utilitarian goods. Journal of Marketing Research, 37(February), 60–71. https://doi.org/10.1509/jmkr.37.1.60.18718

 Ertimur, B., & Gilly, M. C. (2012). So Whaddya Think? Consumers Create Ads and Other Consumers Critique Them. Journal of Interactive Marketing, 26(3), 115–130. https://doi.org/10.1016/j.intmar.2011.10.002

 Esch, F. R., Möll, T., Schmitt, B., Elger, C. E., Neuhaus, C., & Weber, B. (2012). Brands on the brain: Do consumers use declarative information or experienced emotions to evaluate brands? Journal of Consumer Psychology, 22(1), 75–85. https://doi.org/10.1016/j.jcps.2010.08.004

 Girard, T., Trapp, P., Pinar, M., Gulsoy, T., & Boyt, T. E. (2017). Consumer-Based Brand Equity of a Private-Label Brand: Measuring and Examining Determinants.

Journal of Marketing Theory and Practice, 25(1), 39–56.

https://doi.org/10.1080/10696679.2016.1236662

 Godes, D., & Mayzlin, D. (2004). Using Online Conversations to Study Word-of-Mouth Communication. Marketing Science, 23(4), 545–560.

https://doi.org/10.1287/mksc.1040.0071

 Goh, K., Heng, C.-S., & Lin, Z. (2013). Social Media Brand Community and Consumer Behavior: Quantifying the Relative Impact of User- and Marketer-

(35)

34

Generated Content. Information Systems Research, 24(August 2014), 88–107. http://doi.org/10.1287/isre.1120.0469

 Hautz, J., Fuller, J., Hutter, K., & Thurridl, C. (2014). Let users generate your video ads? The impact of video source and quality on consumers’ perceptions and intended behaviors. Journal of Interactive Marketing, 28(1), 1–15.

http://doi.org/10.1016/j.intmar.2013.06.003

 Hewett, K., Rand, W., Rust, R. T., & van Heerde, H. J. (2016). Brand Buzz in the Echoverse. Journal of Marketing, 80(3), 1–24. https://doi.org/10.1509/jm.15.0033  Hoeffler, S., & Keller, K. L. (2002). Building Brand Equity Through Corporate

Societal Marketing. Journal of Public Policy & Marketing, 21(1), 78–89. https://doi.org/10.1509/jppm.21.1.78.17600

 Hollebeek, L. D., Glynn, M. S., & Brodie, R. J. (2014). Consumer brand engagement in social media: Conceptualization, scale development and validation. Journal of

Interactive Marketing, 28(2), 149–165. https://doi.org/10.1016/j.intmar.2013.12.002

 Jackson, R. W., Neidell, L. A., & Lunsford, D. A. (1995). An empirical investigation of the differences in goods and services as perceived by organizational buyers.

Industrial Marketing Management, 24(2), 99–108.

https://doi.org/10.1016/0019-8501(94)00037-W

 Khan, M. L. (2017). Social media engagement: What motivates user participation and consumption on YouTube? Computers in Human Behavior, 66, 236–247.

https://doi.org/10.1016/j.chb.2016.09.024

 Kumar, A., Bezawada, R., Rishika, R., Janakiraman, R., & Kannan, P. K. (2016). From Social to Sale: The Effects of Firm Generated Content in Social Media on Customer Behavior. Journal of Marketing, 80(1), 7–25.

(36)

35

 Labrecque, L. I., vor dem Esche, J., Mathwick, C., Novak, T. P., & Hofacker, C. F. (2013). Consumer power: Evolution in the digital age. Journal of Interactive

Marketing, 27(4), 257–269. https://doi.org/10.1016/j.intmar.2013.09.002

 Levitt T., (1972). Production-Line approach to service. Journal of the Academy of

Marketing Science, 28(1), 24–30.

 Mollen, A., & Wilson, H. (2010). Engagement, telepresence and interactivity in online consumer experience: Reconciling scholastic and managerial perspectives.

Journal of Business Research, 63(9–10), 919–925.

https://doi.org/10.1016/j.jbusres.2009.05.014

 Mudambi, S., & Schuff, D. (2010). Management Information Systems Research Center, University of Minnesota. MIS Quarterly, 26(1), 1–14.

https://doi.org/10.2307/4132321

 Murphy, E. R. (2008). Neuroethics of neuromarketing. Journal of Consumer

Behaviour, 12(4), 253–266. https://doi.org/10.1002/cb

 Murray, K. B., & Schlacter, J. L. (1990). The impact of services versus goods on consumers’ assessment of perceived risk and variability. Journal of the Academy of

Marketing Science, 18(1), 51–65. https://doi.org/10.1007/BF02729762

 O’Donnell, S., Gibson, K., Milliken, M., & Singer, J. (2008). Reacting to YouTube Videos: Exploring Differences Among User Groups. International Communication

Association Annual Conference, (May).

 Orne, M. T. (1961). on the Social Psychology of the Phycological Experiment. In

American Psychological Association Convention (pp. 776–783).

(37)

36

 Pace, S. (2008). YouTube: an opportunity for consumer narrative analysis?

Qualitative Market Research: An International Journal, 11(2), 213–226.

https://doi.org/10.1108/13522750810864459

 Rozin, P., & Royzman, E. B. (2001). Negativity Bias, Negativity Dominance, and Contagion Paul. Personality and Social Psychology Review, 5(4), 296–320. https://doi.org/10.1207/S15327957PSPR0504

 Sashi, C. M. (2012). Customer engagement, buyer-seller relationships, and social media. Management Decision, 50(2). https://doi.org/10.1108/00251741211203551  Schwarz, N. (2004). Metacognitive Experiences in Consumer Judgment and Decision

Making. Journal of Consumer Psychology, 14(4), 332–348. https://doi.org/10.1207/s15327663jcp1404_2

 Schweidel, D. A., & Moe, W. W. (2014). Listening In on Social Media: A Joint Model of Sentiment and Venue Format Choice. Journal of Marketing Research,

51(4), 387–402. https://doi.org/10.1509/jmr.12.0424

 Seo, K.-K., Lee, S., Chung, B. Do, & Park, C. (2015). Users Emotional Valence , Arousal , and Engagement Based on Perceived Usability and Aesthetics for Web Sites. Lnt. Journal of Human-Computer Interaction, 31(April 2015), 72–87. https://doi.org/10.1080/10447318.2014.959103

 Shao, G. (2009). Understanding the appeal of user‐generated media: a uses and gratification perspective. Internet Research, 19(1), 7–25.

https://doi.org/10.1108/10662240910927795

 Shostack, G. L. (1977). Breaking Free from Product Marketing. Journal of Marketing,

(38)

37

 Smith, A. N., Fischer, E., & Yongjian, C. (2012). How Does Brand-related User-generated Content Differ across YouTube, Facebook, and Twitter? Journal of

Interactive Marketing, 26(2), 102–113. http://doi.org/10.1016/j.intmar.2012.01.002

 Sonnier, G. P., McAlister, L., & Rutz, O. J. (2011). A Dynamic Model of the Effect of Online Communications on Firm Sales. Marketing Science, 30(4), 702–716.

https://doi.org/10.1287/mksc.1110.0642

 Sprott, D., Czellar, S., & Spangenberg, E. (2009). The Importance of a General Measure of Brand Engagement on Market Behavior: Development and Validation of a Scale. Journal of Marketing Research, 46(1), 92–104.

https://doi.org/10.1509/jmkr.46.1.92

 Stafford, T. F., Stafford, M. R., & Schkade, L. L. (2004). Determining uses and gratifications for the internet. Decision Sciences, 35(2), 259–288.

https://doi.org/10.1111/j.00117315.2004.02524.x

 Stephen, A. T., & Galak, J. (2012). The Effects of Traditional and Social Earned Media on Sales: A Study of a Microlending Marketplace. Journal of Marketing

Research, 49(5), 624–639. https://doi.org/10.1509/jmr.09.0401

 Swani, K., Brown, B. P., & Milne, G. R. (2014). Should tweets differ for B2B and B2C? An analysis of Fortune 500 companies’ Twitter communications. Industrial

Marketing Management, 43(5), 873–881.

https://doi.org/10.1016/j.indmarman.2014.04.012

 Tang, T. (Ya), Fang, E. (Er), & Wang, F. (2014). Is neutral really neutral? The effects of neutral user-generated content on product sales. Journal of Marketing, 78(4), 41– 58. https://doi.org/10.1509/jm.13.0301

(39)

38

 Teixeira, T., Wedel, M., & Pieters, R. (2012). Emotion-Induced Engagement in Internet Video Advertisements. Journal of Marketing Research, 49(2), 144–159. https://doi.org/10.1509/jmr.10.0207

 Tirunillai, S., & Tellis, G. J. (2012). Does Chatter Really Matter? Dynamics of User-Generated Content and Stock Performance. Marketing Science, 31(2), 198–215. https://doi.org/10.1287/mksc.1110.0682

 Toubia, O., & Stephen, A. T. (2013). Intrinsic vs. Image-Related Utility in Social Media: Why Do People Contribute Content to Twitter? Marketing Science, 32(3), 368–392. https://doi.org/10.1287/mksc.2013.0773

 Tversky, A., & Kahneman, D. (2017). Loss Aversion in Riskless Choice : A Reference-Dependent Model Author ( s ): Amos Tversky and Daniel Kahneman Published by : Oxford University Press Stable URL :

http://www.jstor.org/stable/2937956 REFERENCES Linked references are available on JSTOR for t, 106(4), 1039–1061.

 Ullah, R., Amblee, N., Kim, W., & Lee, H. (2015). From valence to emotions: Exploring the distribution of emotions in online product reviews. Decision Support

Systems, 81, 41–53. https://doi.org/10.1016/j.dss.2015.10.007

 van Dijck, J. (2009). Users like you? Theorizing agency in user-generated content.

Media, Culture & Society, 31(1), 41–58. https://doi.org/10.1177/0163443708098245

 Vargo, S. L., & Lusch, R. F. (2004). Evolving to a New Dominant Logic for Marketing. Journal of Marketing, 68(1), 1–17.

https://doi.org/10.1509/jmkg.68.1.1.24036

 Williams, D., & Whiting, A. (2013). Why people use social media: a uses and gratifications approach. An International Journal Aslib Proceedings Georgios

(40)

39

Research in Interactive Marketing, 16(4), 362–369.

https://doi.org/10.1108/QMR-06-2013-0041

 Zeithaml, V. A. (2016). How Consumer Evaluation Processes Differ for Products and Services, (SEPTEMBER 1981).

(41)

40

9. Appendices

Pre-test

Items _{Cronbach’s Aplha} Cronbach’s Aplha if Item

deleted

I feel very positive when I use the brand

.804 .952

Using the brand makes me happy

.900 .933

I feel good when I use the brand

.957 .905

I am proud to use the brand .900 .932

Overall (4 items) .948

Table 6

Reliability analysis for engagement

Items

Cronbach’s Aplha Cronbach’s Aplha if Item deleted

I recognize the brand among other competing brands

.453 .868

I am familiar with the products/services the brand offers

.733 .577

I can quickly recognize the logo or symbol of the brand

.709 .607

Table 7

(42)

41

Survey

Items

I feel very positive when I use the brand

.815 .920

Using the brand makes me happy

.877 .900

I feel good when I use the brand

.883 .898

I am proud to use the brand .791 .928

Table 8

Reliability analysis for engagement

Items

I recognize the brand among other competing brands

.492 .532

I am familiar with the products/services the brand offers

.437* .666

I can quickly recognize the logo or symbol of the brand

.534 .510

Table 9

A comparative study on the effect of positive and negative user generated videos on brand engagement