The effect of social media content characteristics on online consumer engagement

(1)

Master Thesis

“The effect of social media content characteristics on

online consumer engagement.”

Supervisor: Kristopher Keller

Student: Vasiliki Altiparmaki (student number 11373024)

MSc in Business Administration

Digital Business Track

Final Version

Amsterdam

(2)

2 ACKNOWLEGMENTS

I would first like to express my deepest gratitude to my supervisor dhr. Kristopher Keller of the Faculty of Economics and Business at University of Amsterdam for the useful remarks during the entire process of this master thesis. Without his guidance this study would not have been completed and improved. Furthermore, I owe my warmest gratitude to my sister Ourania and my mother Angeliki for their support over the last year. Finally, I would like to dedicate this thesis to the memory of my father, Nikou.

Statement of Originality

This thesis is written by the student Vasiliki Altiparmaki who declares to take responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3 ABSTRACT

Social media platforms have been a new communication channel and most companies have already integrated them in their marketing strategy. The term of online engagement prevails on social media and brands target to create online communities with high consumer engagement. The purpose of this study is to explore how characteristics of firm generated content, like post media type and message appeal affect online consumer engagement. Furthermore, this study examines whether there is a difference regarding the relationship of content characteristics and online engagement between utilitarian and hedonic brands. The dataset of this study includes 481 posts from 30 Facebook brand pages collected over five weeks. The findings presented in this study suggest that videos provide higher online engagement while entertaining content has a positive impact only on shares. Another interesting finding is that informative content results to higher online engagement for brands with strong utilitarian attributes. As marketers try to face the new digital challenges, this study will provide considerable insights into social media content strategy and will have a significant empirical contribution to social media marketing literature.

Keywords: social media, online consumer engagement, firm-generated content, utilitarian brands, hedonic brands

(4)

4

Social networking sites or social media have been one of the disruptive phenomena of the 21st century. However, the impact of social media in business world was initially very small. Even though they have been an integral part of our society for over a decade, there has been a debate on the presence of brands in social media and companies were not willing to integrate these platforms into their marketing strategies (Lamberton & Stephen, 2016).

These doubts were eliminated and with 2,307 billion active social media users, nowadays, (Chaffey, 2016), companies identify that they can benefit in various ways from their social media brand pages. Even the simplest social media activities, such as posting for new products on Facebook, have a positive impact on brand awareness and brand image (Kim & Ko, 2012; Godey et al., 2016). Additionally, managers believe that companies can find new customers, acquire insights for consumers’ needs and most significantly strengthen the relationship with their customers (Tsimonis & Dimitriadis, 2014). Retailers, for example, who include customer reviews in their social media accounts and are in direct communication with consumers managing possible complains, can achieve higher customer satisfaction (Ramanathan, Subramanian & Parrot, 2017). Like Eric Qualman said “We do not have a choice on whether we do social media, the question is how well we do it?”

One concept, related to social media, is that of engagement. Although it is not a new concept in marketing, it emerged as a valuable topic in 2010 (Marketing Research Institute, 2010) and since then, there has been increasing interest about it from both the academic community and practitioners (Brodie, Hollebeek, Juric & Ilic, 2011). However, companies have not managed yet to understand and leverage consumer engagement into strong customer relationship.

A TrackMaven study (2016) showed that most brands use up to seven different social media platforms and content per brand per channel had increased by 35% by the end of 2015. However, engagement with brand social media posts fell 17% the same period. According to

(6)

6 Stelzner (2016), marketers continue not to know how to engage with their audience through social media. Furthermore, although brands might have multiple social media accounts, they do not know which social media message strategies are most effective concerning consumer engagement (Ashley & Tuten, 2015; Stelzner, 2016). A brand might have millions of fans on Facebook but this will not benefit the brand, unless these fans engage with the company’s posts.

People use social media for various reasons (Whiting & Williams, 2013) and therefore managers struggle to understand how to design social media messages to reach and engage more consumers. Many managers, for example, would like to know if Kohl’s video, which presents company’s initiative to provide presents to a family and had 750.000 likes and half a million shares (Hia, 2016), can be considered best practice for higher online engagement. Nevertheless, some strategies might be effective for one brand but not for another. The main reason is that brands have various characteristics, such as hedonic and utilitarian. Therefore, the research question that arises is:

“How do purpose and type of social media brand messages influence online consumer

engagement regarding utilitarian and hedonic brands?”

Despite the extensive literature on consumer engagement, less research has been conducted considering the factors that influence the consumer engagement on social media. Barger, Peltier and Schultz (2016) identify several antecedents of consumer engagement on social media including brand factors, product factors, consumer factors, content and social media factors, but they do not study how each factor influences consumer engagement. In their research, De Vries and Carlson (2014) examined the drivers of consumer engagement with brands but focused on users’ motivations. In a similar context, Phua, Jin and Kim (2017) researched on users’ attitude and gratifications across social media. However, they did not focus on specific brand or product category.

(7)

7 The literature has also paid little attention to firm-generated content. Although social media have empowered consumers providing them the opportunity to create and share their own experiences with brands, (Fournier & Avery, 2011) firm-generated content is still considered important as it has both offline and online consequences for brands. Firm-generated content, as part of social media brand strategies, affects the consumers’ attitudes toward brands (Schivinski & Dabrowski, 2016) and it can also trigger conversations between firms and consumers and among consumers on social media (Gensler, Volckner, Thompkins & Wiertz, 2013) leading to viral responses and eWOM (electronic word of mouth) (Schivinski & Dabrowski, 2016).

Most researches have explored the effects of firm-created content on customer behavior (Kumar, Bezawada, Rishika, Janakiraman & Kannan, 2016) without examining the characteristics of content such as the type of message. There are few studies that focused on message appeals (Ashley & Tuten, 2015) and components of popular brand posts (De Vries, Gensler & Leeflang, 2012) but the results are unclear or limited to one industry (Cvijikj & Michahelles, 2014). Furthermore, they provide general insights to managers, without taking into consideration that brands with different characteristics (hedonic and utilitarian) need different strategies. Overall, the literature has paid little attention to the effect of social media marketing communication on consumer engagement concerning hedonic and utilitarian products and my research focuses on covering this research gap.

This study will have theoretical implications regarding social media marketing. Particularly, my research will expand the current literature regarding factors that affect online consumer engagement. By exploring social media content characteristics and their impact on consumer engagement, my research will contribute to the existing knowledge of message type and appeals and provide insights for their effectiveness on social media. Moreover, this study will focus on the moderating effect of brand type, that is, brands with utilitarian and hedonic

(8)

8 attributes. Therefore, the results of my research will reveal whether these brand characteristics influence online consumer engagement and my findings can serve as the base for following studies.

Concerning managers, the results of this study will provide to them useful information about how to increase brands’ consumer engagement on Facebook. Firstly, my findings will indicate which media type is more effective regarding consumer engagement. Managers can also find insights on whether an informative or entertaining post is more attractive for Facebook users and which users choose to engage with. Lastly, my research will provide insights to marketers on which appeals and media characteristics to use to attract more social media users depending on the brand characteristics. Therefore, marketers will benefit because they will learn if the brand type affects and in which ways the online consumer engagement.

The aim of this study is to explore communication strategies on social media and their effect on online consumer engagement. Therefore, the data for this research are primary data collected from Facebook. During my research, I collected information from 30 Facebook brand pages over a period of five weeks. The collected data included information about the post type, likes and shares. To classify the chosen brands, I conducted a survey, where participants assessed the chosen brands regarding utilitarian and hedonic characteristics.

(9)

9

2. Theory

2.1 Conceptual Framework

To understand and study the social media communication strategies I focused on two characteristics of social media posts, namely the purpose of the post and the post media type. Regarding the purpose of the post, Jahn and Kunz (2012) state that both informational and entertaining content is valuable and that brands should use interesting and entertaining content to attract more users. Therefore, I focused on these two appeals. Secondly, Cvijikj and Michahelles (2014) studied the vividness and interactivity of posts as key elements of post media type. In this study, for simplification I choose to study only vividness and particularly the impact of text-link and video on consumer engagement on Facebook brand pages.

The moderator of this study is brand type (hedonic and utilitarian value). Although most researches consider products either hedonic or utilitarian, there are some studies that support that there is not a one-dimensional scale but products can have both characteristics (Batra & Ahtola, 1991; Voss, Spangenberg & Grohnmann, 2003). This study adopts this two-dimensional conceptualization (high vs low utilitarian/hedonic value).

According to Hirschman and Holbrook (1982, p 94) utilitarian consumption relates to “…objects for which the consumer desires to maximize utility, where utility typically is measured as some function of the product’s tangible attributes.” Therefore, utilitarian attributes fulfill function-oriented and practical needs. On the other hand, hedonic dimension indicates “…those facets of consumer behavior that relate to the multisensory, fantasy and emotive aspects of one’s experience with products” (Hirschman & Holbrook, 1982, p 92). In other words, hedonic consumption refers to pleasure and affective experiences. Literature usually refers to products, but for this study there is brand segregation. Consumers purchase utilitarian and hedonic products to cover different needs. It will be interesting to explore

(10)

10 whether these different motivations affect their social media engagement and how consumers respond to similar content type used by brands with different characteristics.

Online consumer engagement (OCE) is a type of measurement of effectiveness of Facebook posts that reflects both quantitative and qualitative metrics (Haven, 2007). Although there are different definitions of consumer engagement, one that is used for consumer engagement on online brand communities is that of Brodie, Ilic, Juric and Hollebeek (2013, p.107): “Online consumer engagement is the concept that involves specific interactive experiences between consumers and the brand, and/or other members of the community.” OCE is a well-known type of measurement on Facebook and managers use it to understand consumer behavior on social media (Haven, 2007). In my research, OCE will be also used as measurement of effectiveness of the posts.

I present in figure 2.1.1 the conceptual framework of this study that addresses the above-mentioned research question. In the next section I elaborate the hypotheses according to prior literature. H3 H1 H4 H2

Figure 2.1.1 Overview of the conceptual framework and the hypotheses

Brand Type (hedonic value, utilitarian value) Online Consumer Engagement Post Media Type

(link/text, video) Purpose of Post (Informative, Entertaining)

(11)

11

2.2 Hypotheses

2.2.1 Purpose of Post

Many prior researches have used the uses and gratifications (U&G) theory, which is an audience-centered approach, to explore the consumers’ motivations to use social media platforms (Phua et al., 2017; Whiting & Williams, 2013). U&G theory could be also used to explain the motivations for online engagement.

Two main reasons for people to join online communities are information and entertainment. As the initial purposes of social media platforms were communication and entertainment, people used to enter social media for relaxation, and enjoyment. Today, the purpose of social media has extended and people also join online communities to seek out information for brands. They usually search information for offers, technical products’ specifications, events and businesses (Whiting & Williams, 2013). Brands use both rational (information) and emotional (entertainment) appeals to their Facebook page’s content to attract consumers (Ashley & Tuten, 2015).

Assuming that social media users who visit a Facebook brand page want to learn more about the brand and its products, they will be attracted more by informative messages. Even if consumers enjoy an entertaining post, they will be more interested in a post that includes information for prices, discounts and offers and they would want to share it with their social media friends. Moreover, brand Facebook pages with higher informational value are visited more frequently (De Vries & Carlson, 2014). This could lead to higher online engagement, as the presence of more visitors increases the possibilities of more likes and shares. Taking all above mentioned into consideration, the following hypothesis is proposed:

H1: Posts that contain informative content will have a greater effect on online consumer engagement than posts that contain entertaining content.

(12)

12

2.2.2 Post Media Type

Vividness represents the richness of the brand post and it has two dimensions, namely breadth and depth. The former refers to the ways that the message is presented (e.g. graphics) whereas the latter to the quality of the message (Fortin & Dholakia, 2005). Video increases the vividness of a post, as it includes both sound and image and amplify the breadth of the message. (De Vries et al., 2012). As social media platforms emerge new tools, brands do not post only text so often. Therefore, for this study, posts that include either text or link will reflect low level of vividness, while posts that contain videos reflect high level of vividness.

Consumers prefer visual content as they receive information more easily without consuming time on reading texts. One example is that the presence of animation on banner advertisements increases the click-through rate (Lothia, Donthu & Hershberger, 2003). Furthermore, most websites include images and videos instead of simple text to enrich their content and increase traffic. The same could apply for Facebook brand pages. A video will be more attractive than a link because social media users will find it easier to watch the video rather than visit a page that will probably have more text. Consequently, there are more possibilities to engage with a video than a link.

The fact that social media platforms have integrated video sharing tools, as Facebook did with the live video option, amplify the assumption that vividness of content is considered important and that video might attract more social media users. Therefore, the suggested hypothesis for the level of vividness is:

H2: Posts that include video will have a greater effect on online consumer engagement than posts that include text-link.

(13)

13

2.2.3 Brand Type as Moderator

Brands like products can have hedonic and utilitarian characteristics. Brands that trigger excitement and enjoyment to consumers are considered mostly as hedonic whereas brands that are perceived as helpful and necessary reflect mainly utilitarian value (Voss et. al, 2003). Therefore, utilitarian perceived brands are functional-oriented and hedonic perceived brands are pleasure-oriented. These characteristics could affect the way consumers engage on Facebook brand pages.

Purpose of post and Online Consumer Engagement

People take spontaneous decisions more frequently for hedonic products than utilitarian ones (Dhar & Wertenbroch, 2000). This indicates that consumers are interested about utilitarian products’ attributes and spend more time to search information and compare characteristics of these products. The same could apply for brands. Consumers who search for valuable information, will be persistent in their search and desire more and more information (Bateman, Gray & Butler, 2011). Therefore, Facebook users that visit a page of a brand with strong utilitarian characteristics expect to find information about the brand and products and will be attracted from informative posts. As consumers focus on informative posts, it is expected that they engage more with these posts.

Consumers also look up information for hedonic products but their orientation is the experience they have using these specific products (Anderson, Knight, Pookulangara & Josiam, 2014). Consequently, they do not focus so extensively on products’ characteristics but search for entertainment and enjoyment. In this case, affect-expressive appeals are more effective to perceived hedonic goods (Johar & Sirgy, 1991). Regarding Facebook posts of a brand with strong hedonic characteristics, entertaining content will be appreciated more than posts that include information for pricing and products.

(14)

14 H3a: When a Facebook post contains informative content, online consumer engagement will be higher for a brand with strong utilitarian characteristics compared to a brand with weak utilitarian characteristics.

H3b: When a Facebook post contains informative content, online consumer engagement will be higher for a brand with weak hedonic characteristics compared to a brand with strong hedonic characteristics.

Post media type and Online Consumer Engagement

According to Elaboration Likelihood Model (ELM) (Petty & Cacioppo, 1986), people tend to process stimulus of a message differently. When people use central route to evaluate information provided, they focus more on cognitive elements than image and sound that are not considered essential. On the other hand, with peripheral route processing people do not examine information thoroughly and are affected by affective stimulus, such as animation or music (Lothia et al., 2003).

Hedonic-oriented purchases are driven by affective involvement while utilitarian-oriented purchases are driven by cognitive involvement (Voss et al., 2003). This combined with the ELM means that utilitarian-focused consumers appreciate an online environment that provides more information. Links on social media posts lead to brand sites which can be considered as critical source of information. Regarding consumers that look for hedonic products, they choose enjoyment over information and they might not click on the link. As hedonic aspects of sites have a positive impact on consumers who seek hedonic products (Childers, Carr, Peck & Carson, 2001), the same concept can be applied to Facebook brand pages. Therefore, in a Facebook page of a highly hedonic perceived brand consumers are attracted by affective content.

As videos include affective components while links lead to more information, these hypotheses are assumed:

(15)

15 H4a: When a Facebook post contains text/link, online consumer engagement will be higher for a brand with strong utilitarian characteristics compared to a brand with weak utilitarian characteristics.

H4b: When a Facebook post contains text/link, online consumer engagement will be higher for a brand with weak hedonic characteristics compared to a brand with strong hedonic characteristics.

(16)

16

3. Methodology

The aim of this research is to study social media communication strategies used by brands and examine whether post media type and the purpose of the post have an impact on online consumer engagement. Moreover, this study tests the moderating effect of brand type (high vs low utilitarian/hedonic value). To this end, I conducted both social media research and survey.

3.1 Social Media Research

For this study, I primarily conducted social media research. As my focus was social media posts and their respective engagement, I used the data mining method to collect data from 30 Facebook brand pages. Using Facebook API and R Studio, I created a social media dataset that was considered sufficient to fulfill the purpose of this research. I chose Facebook because it is the most significant social media site with 1.79 billion monthly active users as of September 2016 (Facebook, 2017).

I was collecting the data over a period of five weeks, from 26March 2017 to 29 April 2017. I collected data for the Facebook posts daily until one week after the post creation. However, I noticed that there was no significant increase in the likes and shares after the fifth day. Therefore, I use the number of likes and shares of the fifth day for measuring the engagement. The total number of posts that I obtained was 486. After having collected the data, I did content analysis of the posts to distinguish them into informative and entertaining.

Regarding the brands for which I collected data, I used the site www.socialbakers.com as a source of information. This site includes a Facebook statistics directory with Facebook brand pages ranked according their fans. As the moderator of my research is the brand type, I focused on various industries to cover both hedonic and utilitarian context. Therefore, I chose seven industries, namely Fashion, Food, Electronics, Beauty Stores, Retailers, Auto and Home & Living. From each industry, I selected ten brands. The criteria for the selection of the brands were three: a) the Facebook fan page is created by the company, b) English is used as the

(17)

17 communication language, c) the number of page’s fans. The final choice was made using an online randomizer. In Appendix A, there is the list of selected brands and their Facebook pages.

3.2 Survey

To examine the brand type, it was necessary to classify the brands regarding their hedonic and utilitarian attributes. To achieve this segmentation and since brands usually have both characteristics (Okada, 2005), I conducted a survey.

The HED/UT scale presented by Voss et al. (2003) was selected to build the survey questionnaire. They developed a Semantic Differential scale with ten pairs of adjectives (Table 3.2.1), five for each category, that best reflect the hedonic and utilitarian dimensions of consumer attitude. While previous scales were used only for products, the HED/UT scale was used to measure the hedonic and utilitarian dimensions of both products and brands. As this study focuses on brands, this scale was considered the most appropriate. Moreover, I used a bipolar scale in the questionnaire to minimize the risk of bias that increases if only one of the two constructs is presented to respondents. An example of the questionnaire is in Appendix B. I used a non-probability sample for this complementary questionnaire. Particularly, the method of self-selection was used inviting people to participate voluntarily in the survey. After the introductory message, the participants were asked to assess five brands using the HED/UT scale. To ensure that the respondents had at least a little knowledge of the brand before they evaluate it, I also provided a link to brand’s website.

Table 3.2.1

Pairs of adjectives used in the HED/UT scale

Utilitarian Dimension Hedonic Dimension

Effective – Ineffective Fun – Not fun Helpful – Unhelpful Exciting – Dull Functional – Not functional Delightful – Not delightful

Necessary – Unnecessary Thrilling – Not thrilling Practical – Impractical Enjoyable – Unenjoyable

(18)

18

4. Data

4.1 Operationalization of variables

This research examines the purpose of post and post media type as independent variables. The collected data include information for the post media type (video or link/text) and I manually coded the purpose of post by identifying a post either as informative or entertaining. For the coding, I took into consideration the operationalization used by De Vries et al. (2012). Particularly, I considered a post as informative if it included the brand name or information for products related to the brand. On the other hand, I classified a post as entertaining when there was not any reference to the brand or related products. One example of informative post from the dataset is: “Through 4/11/17 get $5 flat rate shipping on our legendary cheesecakes from Harry & David with offer code: FIVECAKE.” (The Cheesecake Company-4.4.2017) Similarly, one example of entertaining post is: “Stuck on weekend plans?” (Topshop– 14.4.2017).

Additionally, I used the results of the survey to operationalize my moderator. In total, 78 respondents filled in the survey and 13 respondents assessed each brand. After processing the collected answers, I calculated the mean for the utilitarian and hedonic items (see Appendix C, Table 1). The following “perception map” (Figure 4.1.1) depicts how the selected brands were perceived. To create this map, I mean centered both the hedonic and utilitarian values (see Appendix C, Table 2). According to the results of the survey, the brands in the top left quadrant are those with mainly hedonic characteristics. Additionally, the brands in the bottom right can be considered more utilitarian. For the analysis, I use the utilitarian and hedonic means as values for the respective variables.

(19)

19

Figure 4.1.1 Perceptual Map of Brands

My dependent variable is online consumer engagement which I measured by shares, likes and comments of posts. However, the engagement through comments was too low, so I decided to exclude it from the analysis. I chose to evaluate engagement separately using likes and shares to avoid Facebook users overlapping and have a more extensive understanding of OCE. Each Facebook brand page has different number of fans and consequently, their posts have different reach. Therefore, the shares and likes cannot be used as complete indicators of Facebook engagement. The number of fans on Facebook brand pages at the day of each post is also considered valuable information for measuring Facebook engagement (Barger & Labrecque, 2013). Overall, the following ratio was used as measurement of OCE. I multiplied the ratio with 10,000 to have numbers that I can use in the analysis.

OCElikes = x10,000 (1) OCEshares = x10,000 (2)

Number of likes Number of fans

Number of shares Number of fans

(20)

20

4.2 Descriptive Statistics

In total, I collected 486 posts but after removing the outliers I kept 481 posts that I used in the analysis. Brands do not post on Facebook with the same frequency, so the number of posts differs for each brand (see Appendix C, Table 3).

Regarding the type of posts, posts containing informative content were the most frequent (308 posts, 64% of total) compared to posts containing entertaining content (173 posts, 36% of total). Tommy Hilfiger was the brand that focused on the informative content, while Topshop preferred mostly entertaining content (see Appendix D, Figures 1-3). I also noticed differences in terms of the type of the post. Links were the most frequently used across brands (242 posts, 50.3% of total) followed by videos (239 posts, 49.7% of total). Eight out of the thirty brands did not post any link during the five weeks. On the other hand, IKEA and Tommy Hilfiger posted more links than videos (see Appendix D, Figures 4-6).

To understand better, how social media users engage on Facebook, I analyzed the descriptive statistics for the two measurements of OCE. The data show that people like a post (mean 1027.63, std. 2166.628) more frequently than share it (mean 152.5, std. 628.994). Tables 4.2.1 and 4.2.2 include the descriptive statistics. Furthermore, I compared the average engagement in terms of likes and shares among the chosen brands. Figures 4.2.1 and 4.2.2 illustrate that BMW and Wendy’s had the higher average engagement regarding both likes and shares. Table 4.2.1 Frequency Table Frequency Percent Purpose of post Informative 308 64.0 Entertaining 173 36.0

(21)

21

481 100.0

Post Media Type

Video 239 49.7 Link_text 242 50.3 481 100.0 Table 4.2.2 Descriptive Statistics Likes Shares Minimum 4 0 Maximum 18,622 8,066 N 481 481 Mean 1027.63 152.54 Std. Deviation 2166.628 628.994

(22)

22

Figure 4.2.2 Average engagement (shares) per brand

4.3 Method

The dependent variables of this study are continuous (likes and shares) while the independent ones are categorical (purpose and type). For the analysis, I coded the purpose (link = 1, video = 0) and type (informative =1, entertaining = 0). Furthermore, I used two continuous variables (hedonic and utilitarian) as the moderator of this research. For understanding the effect of the independent variables on the dependent and confirming or not the hypotheses, I based the analysis of the data on multiple regression analysis. The regression included all the independent variables and the moderator and I performed it twice for each dependent variable. The equations (1) and (2) illustrate the model used for this study:

Ylikes= α0l +β1l Χ1 +β2l Χ2 +β3l Χ3 +β4l Χ4 +β5l (Χ1Χ3) +β6l (Χ1Χ4) +β7l (Χ2Χ3) +β8l (Χ2Χ4) +el (1)

Yshares=α0s+β1s Χ1+β2s Χ2+β3s Χ3 +β4s Χ4 +β5s (Χ1Χ3) +β6s (Χ1Χ4) +β7s (Χ2Χ3) +β8s (Χ2Χ4)+es (2)

where, Χ1 depicts the purpose of post, Χ2 depicts the type of post, Χ3 is the utilitarian value and

(23)

23

5. Results

In this chapter, I present the results of the analysis. For testing the hypotheses, I used SPSS 24.0. Before starting the analysis, I checked for outliers, that is, cases with extremely high or low value that could bias the statistics and affect the results. To identify possible outliers, I followed a procedure indicated by Seo (2006). I checked the standardized values of the dependent variables and I excluded five that had |z|>3.

Data should also not violate the five assumptions of linear regression, namely, linearity, non-multicollinearity, independence, homoscedasticity and normality (Osborne & Waters, 2002). To ensure the normal distribution of the residuals, I transformed the dependent variables using natural logarithms (log10). Through statistical tests and visual inspection of graphs, I confirmed that I could use the data for multiple regression as all the assumptions were met. I include the relevant statistics tables (skewness value for normality, tolerance index, VIF values and correlation table for multicollinearity, Durbin – Watson test for independence) and graphs (p-plots and histograms for normality, scatterplots for linearity and homoscedasticity) in Appendix C (Tables 4-7) and D (Figures 7-13 ), respectively.

Table 5.1

Results of multiple regression for Model A (likes) and Model B (shares) Unstandardized Coefficients Standardized Coefficients B SE β t Sig. (0.05) Model A (Constant) 2.165 .604 3.587 .000 Type -.274 .053 -.228 -5.186 .000 Purpose -.065 .055 -.052 -1.176 .240 Utilitarian -.496 .154 -.316 -3.221 .001

(24)

24 Note: significant at 0.05 level

Table 5.1 depicts the results of the analysis. Model A represents the regression for likes while model B refers to shares. Totally, eight predictors were entered in each multiple regression, namely, type, purpose, utilitarian, hedonic, the interaction term of purpose and utilitarian, the interaction term of purpose and hedonic, the interaction term of type and

Hedonic -.132 .095 -.129 -1.395 .164 Interaction Purpose *Utilitarian .133 .061 .182 2.174 .030 Interaction Purpose *Hedonic .086 .057 .112 1.513 .131 Interaction Type *Utilitarian -.076 .059 -.104 -1.288 .198 Interaction Type *Hedonic .029 .055 .037 .531 .596 R=.407, R2 = .165, F=11.693, p=.000 Model B (Constant) .419 .845 .495 .621 Type -.440 .074 -.268 -5.952 .000 Purpose -.253 .077 -.148 -3.295 .001 Utilitarian -.269 .216 -.125 -1.247 .213 Hedonic -.110 .133 -.078 -.827 .409 Interaction Purpose *Utilitarian .197 .085 .197 2.306 .022 Interaction Purpose *Hedonic .098 .080 .093 1.231 .219 Interaction Type *Utilitarian -.034 .082 -.034 -.414 .679 Interaction Type *Hedonic -.039 .077 -.036 -.503 .615 R=.355, R2 = .126, F=8.501, p=.000

(25)

25 utilitarian and the interaction term between type and hedonic. The results showed that model A was statistically significant F (8,472) = 11.693; p= .000 and explained 16.5% of variance in online consumer engagement in terms of likes.

In this model, purpose of post (β= -.052, p= .240) did not have any significant effect on the dependent variable. This result rejects H1 regarding likes. On the other hand, type of post was statistically significant (β= -.228, p= .000). It means that links have a negative impact on likes by .228 compared to videos, which confirms H2. The analysis also indicated if the relationship between each independent variable and OCE is moderated by brand type. For confirming this assumption, I checked the results for the four interaction terms. Only one interaction term out of four, namely purpose*utilitarian, was statistically significant (β= .182, p= .030). In other words, one brand with higher utilitarian value strengthens more the relationship between purpose of post and OCE compared to a brand with lower utilitarian value. The positive coefficients indicate that informative posts are more important for brands with strong utilitarian characteristics, supporting H3a in terms of likes. The results for the other interaction terms, that is, purpose* hedonic (β= .112, p= .131), type*utilitarian (β= -.104, p= .198) and type*hedonic (β= .037, p= .596), reject the H3b, H4a and H4b respectively.

Regarding the model B, it was also statistically significant F (8,472) = 8.501; p= .000) and explained 12.6% of variance in online consumer engagement in terms of shares. In the model B, both type of post (β= -.268, p= .000) and purpose of post (β= -.148, p= .001) were statistically significant. This shows that links cause less shares by .268 compared to videos and informative content has a negative effect over shares by .148 compared to entertaining content. Therefore, regarding shares, the results reject H1 while support H2. Like model A, model B had similar results for the four interaction terms. Purpose*utilitarian was statistically significant (β=.197, p=0.22), supporting H3a. On the other hand, purpose*hedonic (β=.093,

(26)

26 p=0.219), type*utilitarian (β= -.034, p=0.679) and type*hedonic (β=-.036, p=0.615) do not affect OCE.

To explore more the moderating effect of the utilitarian brand type, I plotted the interaction effect between utilitarian and purpose of post. Figure 5.1 illustrates the two interactions plots regarding likes and shares. The visual inspection of the plots confirms that for a brand with strong utilitarian characteristics, informative content can result to higher OCE compared to entertaining content. However, entertaining posts is more effective in terms of OCE for a brand with less utilitarian characteristics.

(27)

27

6. Discussion

The aim of this study was to study the characteristics of firm generated content on social media and how the purpose and type of post affect online consumer engagement in terms of likes and shares. Furthermore, this study explored how the way that social media users perceive a brand regarding the brand’s utilitarian and hedonic characteristics moderates the relationship between characteristics of post and online consumer engagement. The dataset that I used included 481 Facebook posts from 30 Facebook brand pages. Furthermore, the selected brands had different utilitarian and hedonic value as indicated by the results of the survey. Consequently, this research provides valuable insights regarding online consumer engagement. Table 6.1 summarizes the findings of this study.

Table 6.1

Summary of findings

Variable Results

Purpose of post (H1) Posts with entertaining content have a positive impact on shares but not likes. On the other hand, informative posts have a negative impact on shares.

Post media type (H2) Posts with video result on higher online consumer engagement (likes and shares) compared to posts with links.

Brand Type and purpose of post (H3a-b)

Utilitarian brand characteristics increase the impact of purpose of post on OCE. Hedonic brand characteristics do not affect this relationship.

Brand Type and post media type (H4a-b)

Neither utilitarian nor hedonic brand characteristics have an impact on the relationship between post media type and online consumer engagement.

Contrary to the first hypothesis, results indicated that informative posts have a negative effect on online consumer engagement compared to these that include entertaining content. Even if brands choose Facebook to create online brand communities and share information about products or services, the results of this study illustrate that Facebook users engage more with entertaining posts. One explanation could be the social media platform and its characteristics. Nowadays, people use Facebook to interact with each other and relax rather

(28)

28 than search for information. They probably focused on Facebook as a source of information a few years ago, but other social media platforms with more informative content have emerged. Therefore, people might prefer other social media platforms like Twitter rather than Facebook when they search for information.

The results confirmed the hypothesis that posts that contain video have higher levels of online consumer engagement. This indicates that social media users prefer visual content than simple text that can found browsing via links. Furthermore, when someone clicks a link, he visits another site, leaving behind the Facebook brand page. More social media users do not usually like or share a link before they click on it. Thus, if the website that they visit do not provide a way to like or share the content without visiting again the Facebook page, people might skip it. This might be another explanation of the links’ lower online engagement.

Since hedonic-oriented purchases are driven by affective involvement while utilitarian-oriented purchases are driven by cognitive involvement (Voss et al., 2003), I assumed that videos as more affective could be more engaging than links for highly hedonic brands. One second assumption was that links will have higher engagement for highly utilitarian brands. However, there was no moderating effect on the relationship between media post type and online consumer engagement. One explanation could be that people do not think as buyers when they visit Facebook brand pages. They want to engage with the brand but they do not consider Facebook as purchase channel.

Regarding the relationship between purpose of post and online consumer engagement, only utilitarian brands influence it. The results of the analysis supported the hypothesis that posts with informative content will be more engaging for brands with utilitarian characteristics. Therefore, for a brand with strong utilitarian characteristics, informative posts have higher level of engagement compared with one that has less utilitarian characteristics.

(29)

29

6.1 Theoretical implications

My study contributes to current literature for social media marketing which is one of the prevalent topics in academic journals (Lamberton & Stephen, 2016). This study focuses on two concepts, namely firm-generated content and online consumer engagement, to which the literature has paid little attention. On the other hand, there has been an increasing interest for online consumer engagement from marketers. Therefore, I hope that this study will be valuable in covering the existing gap between theory and marketers’ needs and its findings can trigger further research.

This study adds to prior researches by demonstrating that purpose of posts does not affect OCE regarding likes. Opposing to my hypothesis that informative content will result to higher engagement, entertaining content is more engaging in terms of shares. The findings indicate that although uses and gratifications theory has been used from previous researches to explain why people use social media, it cannot explain why people choose to engage.

Another key finding of this study is the results for the moderating effect of the brand type. To my knowledge, there have been no similar studies done on how either utilitarian or hedonic brand characteristics might influence online consumer engagement. This work has shown that only utilitarian characteristics affect the impact of purpose of post on OCE. These findings can improve our knowledge about different aspects of online consumer engagement on social media.

6.2 Managerial Implications

As stated in the introduction, the main objective of this study was to study firm-generated content published on Facebook brand pages and investigate how the media type and message appeals of Facebook posts influence online consumer engagement. The present findings can

(30)

30 constitute valuable insights for community managers and marketers who plan social media strategy. The following is a reference to the three main managerial implications.

Choosing post media type

There are currently five types of posts, namely, video, photo, link, event and text, among which Facebook page moderators can choose. Even if this study focuses only on link and video, the findings indicate that posts with high vividness are more attractive to Facebook users. This applies for both likes and shares. Therefore, managers should choose more vivid content, like a video or a photo to post on Facebook brand page. These posts could increase the overall online consumer engagement.

Choosing message appeal

Regarding message appeal, this study explored the effect of informative and entertaining content on online consumer engagement. Results presented in previous chapter showed that informative posts decrease the shares compared to entertaining posts. Marketers should take this finding into consideration when they are doing content planning. Apart from providing information for the brand or products, community managers should include enough entertaining content to their Facebook posts. This way, they can enhance the number of shares. Sharing is significant because it can increase online word of mouth which can potentially lead to more brand awareness or conversions.

Differences between utilitarian and hedonic brands

Even though the results did not highlight significant differences between utilitarian and hedonic brands, managers can still benefit from the findings. The fact that there was no interaction between brand type and post media type demonstrates that marketers do not need to consider the brand type characteristics when they choose post media type. On the other hand, managers should focus on informative content to achieve higher engagement, if the brand has strong utilitarian characteristics.

(31)

31

6.3 Limitations and further research

The amount of collected data was sufficient to perform the analysis and investigate the relationship between characteristics of firm generated content and online engagement. However, this research is subject to some limitations which may lead to further research.

Firstly, this study focused on Facebook and the online consumer engagement on this social media network. Nevertheless, social media platforms have different characteristics which can affect why and how people engage with content. For example, people use Twitter for informative purposes while hedonic posts are more popular on Instagram (Krallman, Pelltier & Adams, 2016). Future researches should investigate how media post type and purpose of post affect online consumer engagement to other social media platforms.

Secondly, this study was limited regarding measures of online consumer engagement and characteristics of posts. I chose only likes and shares as operationalization of online engagement. Even if comments are less compared to likes and shares, they still indicate engagement. One important characteristic of comments is sentiment, they can be negative or positive. Focusing on sentiment of comments and exploring how it affects online engagement, can also give useful insights for managers. Additionally, future research could focus on other significant characteristics of firm generated content like frequency of posts, exact day and time of posts. This information is also important for managers when they plan a social media posting calendar.

Lastly, I collected data for a limited time for specific industries. Some brands do not post so frequently and there was difference between the number of posts per brand. To confirm the results, future researches should expand the time for data collection. Furthermore, more industries could be used to investigate more the impact of brand type on online consumer engagement.

(32)

32 References

Anderson, K. C., Knight, D. K., Pookulangara, S., & Josiam, B. (2014). Influence of hedonic and utilitarian motivations on retailer loyalty and purchase intention: A facebook

perspective. Journal of Retailing and Consumer Services, 21(5), 773-779.

Ashley, C., & Tuten, T. (2015). Creative strategies in social media marketing: An exploratory study of branded social content and consumer engagement. Psychology &

Marketing, 32(1), 15-27.

Barger, V., Peltier, J. W., & Schultz, D. E. (2016). Social media and consumer engagement: A review and research agenda. Journal of Research in Interactive Marketing, 10(4), 268-287.

Barger, V. A., & Labrecque, L. (2013). An integrated marketing communications perspective on social media metrics.

Bateman, P. J., Gray, P. H., & Butler, B. S. (2011). Research note—the impact of community commitment on participation in online communities. Information Systems

Research, 22(4), 841-854.

Batra, R., & Ahtola, O. T. (1991). Measuring the hedonic and utilitarian sources of consumer attitudes. Marketing Letters, 2(2), 159-170.

Brodie, R. J., Hollebeek, L. D., Jurić, B., & Ilić, A. (2011). Customer engagement:

Conceptual domain, fundamental propositions, and implications for research. Journal of Service Research, 14(3), 252-271.

(33)

33 Brodie, R. J., Ilic, A., Juric, B., & Hollebeek, L. (2013). Consumer engagement in a virtual

brand community: An exploratory analysis. Journal of Business Research, 66(1), 105-114.

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

Chaffey, D. (2016). Global social media research summary 2016. Retrieved

from http://www.smartinsights.com/social-media-marketing/social-media-strategy/new-global-social-media-research/

Childers, T. L., Carr, C. L., Peck, J., & Carson, S. (2002). Hedonic and utilitarian motivations for online retail shopping behavior. Journal of Retailing, 77(4), 511-535.

De Vries, L., Gensler, S., & Leeflang, P. S. (2012). Popularity of brand posts on brand fan pages: An investigation of the effects of social media marketing. Journal of Interactive Marketing, 26(2), 83-91.

De Vries, N. J., & Carlson, J. (2014). Examining the drivers and brand performance implications of customer engagement with brands in the social media

environment. Journal of Brand Management, 21(6), 495-515.

Dhar, R., & Wertenbroch, K. (2000). Consumer choice between hedonic and utilitarian goods. Journal of Marketing Research, 37(1), 60-71.

Facebook, company info - stats . (2016). Retrieved from http://newsroom.fb.com/company-info/

(34)

34 Fortin, D. R., & Dholakia, R. R. (2005). Interactivity and vividness effects on social presence

and involvement with a web-based advertisement. Journal of Business Research, 58(3), 387-396.

Fournier, S., & Avery, J. (2011). The uninvited brand. Business Horizons, 54(3), 193-207.

Gensler, S., Völckner, F., Liu-Thompkins, Y., & Wiertz, C. (2013). Managing brands in the social media environment. Journal of Interactive Marketing, 27(4), 242-256.

Godey, B., Manthiou, A., Pederzoli, D., Rokka, J., Aiello, G., Donvito, R., & Singh, R. (2016). Social media marketing efforts of luxury brands: Influence on brand equity and consumer behavior. Journal of Business Research, 69(12), 5833-5841.

Haven, B. (2007). Marketing’s new key metric: Engagement. Marketing,

Hia, R. (22 December 2016). Brands tap into emotion for the 5 most engaging facebook posts of 2016. Retrieved from http://adage.com/article/digital/brands-engaging-facebook-posts-2016/307274/

Hirschman, E. C., & Holbrook, M. B. (1982). Hedonic consumption: Emerging concepts, methods and propositions. The Journal of Marketing, 92-101.

Jahn, B., & Kunz, W. (2012). How to transform consumers into fans of your brand. Journal of Service Management, 23(3), 344-361.

Johar, J. S., & Sirgy, M. J. (1991). Value-expressive versus utilitarian advertising appeals: When and why to use which appeal. Journal of Advertising, 20(3), 23-33.

(35)

35 Kim, A. J., & Ko, E. (2012). Do social media marketing activities enhance customer equity?

an empirical study of luxury fashion brand. Journal of Business Research, 65(10), 1480-1486.

Krallman, A., Pelletier, M. J., & Adams, F. G. (2016). @Size vs. #Impact: Social media engagement differences amongst facebook, twitter, and instagram. In K. K. Kim (Ed.), Celebrating Americaâ€™s pastimes: Baseball, hot dogs, apple pie and marketing? proceedings of the 2015 academy of marketing science (AMS) annual conference (pp. 557-561). Cham: Springer International Publishing. doi:10.1007/978-3-319-26647-3_112

Kumar, A., Bezawada, R., Rishika, R., Janakiraman, R., & Kannan, P. (2016). From social to sale: The effects of firm-generated content in social media on customer

behavior. Journal of Marketing, 80(1), 7-25.

Lamberton, C., & Stephen, A. T. (2016). A thematic exploration of digital, social media, and mobile marketing: Research evolution from 2000 to 2015 and an agenda for future inquiry. Journal of Marketing, 80(6), 146-172.

Lothia, R., Donthu, N., & Hershberger, E. K. (2003). The impact of content and design elements on banner advertising click-through rates. Journal of Advertising

Research, 43(4), 410-418.

Marketing Science Institute. (2010). 2010-2012 research priorities Boston.

Okada, E. M. (2005). Justification effects on consumer choice of hedonic and utilitarian goods. Journal of Marketing Research, 42(1), 43-53.

(36)

36 Osborne, J. W., & Waters, E. (2002). 1 four assumptions of multiple regression that

researchers should always test.

Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123-205.

Phua, J., Jin, S. V., & Kim, J. (. (2017). Gratifications of using facebook, twitter, instagram, or snapchat to follow brands: The moderating effect of social comparison, trust, tie strength, and network homophily on brand identification, brand engagement, brand commitment, and membership intention. Telematics and Informatics, 34(1), 412-424.

Ramanathan, U., Subramanian, N., & Parrott, G. (2017). Role of social media in retail network operations and marketing to enhance customer satisfaction. International Journal of Operations & Production Management, 37(1), 105-123.

Schivinski, B., & Dabrowski, D. (2016). The effect of social media communication on

consumer perceptions of brands. Journal of Marketing Communications, 22(2), 189-214.

Seo, S. (2006). A Review and Comparison of Methods for Detecting Outliers in Univariate Data Sets.

Stelzner, M. A. (May 2016). 2016 social media marketing industry report. how marketeres are using social media to grow their businesses..Social Media Examiner.

TrackMaven (2016). The content marketing paradox revisited: Time for a reboot? Washington DC

Tsimonis, G., & Dimitriadis, S. (2014). Brand strategies in social media. Marketing Intelligence & Planning, 32(3), 328-344.

(37)

37 Voss, K. E., Spangenberg, E. R., & Grohmann, B. (2003). Measuring the hedonic and

utilitarian dimensions of consumer attitude. Journal of Marketing Research, 40(3), 310-320.

Whiting, A., & Williams, D. (2013). Why people use social media: A uses and gratifications approach. Qualitative Market Research: An International Journal, 16(4), 362-369.

(38)

38

Appendix A: Brands 1. Primark – https://www.facebook.com/Primark/

2. Topshop – https://www.facebook.com/Topshop/

3. Dunkin Donuts – https://www.facebook.com/DunkinDonutsUS/

4. The cheesecake factory – https://www.facebook.com/thecheesecakefactory/

5. Staples – https://www.facebook.com/staples/

6. The body shop – https://www.facebook.com/TheBodyShop/

7. Whole Foods Market – https://www.facebook.com/WholeFoods/

8. Walgreens – https://www.facebook.com/Walgreens/

9. Lego – https://www.facebook.com/LEGO/

10. Craftsman– https://www.facebook.com/craftsman/

11. Reliance Digital - https://www.facebook.com/reliancedigital/

12. BMW – https://www.facebook.com/BMW/

13. MAC Cosmetics – https://www.facebook.com/MACcosmetics/

14. STIHL – https://www.facebook.com/STIHL/

15. Nordstrom – https://www.facebook.com/Nordstrom/

16. Kohl’s – https://www.facebook.com/kohls/

17. United Colors of Benetton – https://www.facebook.com/Benetton/

18. IKEA – https://www.facebook.com/IKEAUSA/

19. Subway – https://www.facebook.com/subway/

20. Wendy’s – https://www.facebook.com/wendys/

21. Tommy Hilfiger - https://www.facebook.com/tommyhilfiger/

22. Target – https://www.facebook.com/target/

23. Sephora – https://www.facebook.com/sephora/

24. Foot Locker – https://www.facebook.com/footlocker/

25. Toys”R”Us – https://www.facebook.com/toysrus/

26. Best Buy - https://www.facebook.com/bestbuy/

27. Lowe’s Home Improvement – https://www.facebook.com/lowes/

28. Old Navy – https://www.facebook.com/oldnavy/

29. DICK’s – https://www.facebook.com/dickssportinggoods/

(39)

39 Appendix B: Example of the questionnaire

Hello and welcome to my master thesis survey!

Thank you for agreeing to take part in this survey. I would like to inform you that the participation in this study is voluntary and you can stop at any moment and cancel your session. The questionnaire contains questions about five brands and it will take about 10 minutes to complete. Your responses are a vital part of my research and the data collected

will be treated confidentially and used only for the research purposes of my Master Thesis in

Business Administration (University of Amsterdam). Only aggregated results will be

presented, not individual responses. Therefore I would like to ask you to answer the questions honestly. After I have collected the necessary responses, I will raffle two gifts cards for two respondents. If you have any question for the survey, please contact me at

vasia.altiparmaki@student.uva.nl. Thanks in advance for your participation. Vasiliki Altiparmaki

Instructions: The following questions are about five brands. You should assess each brand regarding ten different adjective pairs. For each pair, select the adjective that reflects best your opinion about the brand.

The next question is about the brand "Dunkin Donuts". If you do not know the brand you can

visit their site here. Please choose one option from the rating scale for each row. "I find Dunkin

Donuts..."

1 2 3 4 5

Ineffective ○ ○ ○ ○ ○ Effective Unhelpful ○ ○ ○ ○ ○ Helpful

(40)

40 Not functional ○ ○ ○ ○ ○ Functional

Unnecessary ○ ○ ○ ○ ○ Necessary Impractical ○ ○ ○ ○ ○ Practical Not fun ○ ○ ○ ○ ○ Fun Dull ○ ○ ○ ○ ○ Exciting Not delightful ○ ○ ○ ○ ○ Delightful Not thrilling ○ ○ ○ ○ ○ Thrilling Unenjoyable ○ ○ ○ ○ ○ Enjoyable

The next question is about the brand "Staples". If you do not know the brand you can visit

their site here. Please choose one option from the rating scale for each row. "I find Staples..."

1 2 3 4 5

Ineffective ○ ○ ○ ○ ○ Effective Unhelpful ○ ○ ○ ○ ○ Helpful Not functional ○ ○ ○ ○ ○ Functional Unnecessary ○ ○ ○ ○ ○ Necessary Impractical ○ ○ ○ ○ ○ Practical Not fun ○ ○ ○ ○ ○ Fun Dull ○ ○ ○ ○ ○ Exciting Not delightful ○ ○ ○ ○ ○ Delightful Not thrilling ○ ○ ○ ○ ○ Thrilling Unenjoyable ○ ○ ○ ○ ○ Enjoyable

The next question is about the brand "BMW". If you do not know the brand you can visit their

(41)

41 1 2 3 4 5

The next question is about the brand "United Colors of Benetton". If you do not know the

brand you can visit their site here. Please choose one option from the rating scale for each row.

"I find United Colors of Benetton..."

1 2 3 4 5

Ineffective ○ ○ ○ ○ ○ Effective Unhelpful ○ ○ ○ ○ ○ Helpful Not functional ○ ○ ○ ○ ○ Functional Unnecessary ○ ○ ○ ○ ○ Necessary Impractical ○ ○ ○ ○ ○ Practical Not fun ○ ○ ○ ○ ○ Fun Dull ○ ○ ○ ○ ○ Exciting Not delightful ○ ○ ○ ○ ○ Delightful

(42)

42 Not thrilling ○ ○ ○ ○ ○ Thrilling

Unenjoyable ○ ○ ○ ○ ○ Enjoyable

The next question is about the brand "Toys "R" US". If you do not know the brand you can

visit their site here. Please choose one option from the rating scale for each row. "I find Toys

"R" US..."

1 2 3 4 5

Please answer the following three general questions.

What is your gender?

○ Female ○ Male

What is your age?

○ <16 ○ 16-25 ○ 26-35 ○ 36-45 ○ 46-55 ○ >55

(43)

43 Appendix C: Tables

Table 1

Hedonic and utilitarian value per brand

Brands N Utilitarian Mean Hedonic Mean

Burger King 13 2.89 2.72

Primark 13 3.00 2.72

Wendy’s 13 3.17 3.12

United Colors of Benetton 13 3.29 3.14

Kohl’s 13 3.40 3.38 Old navy 13 3.48 3.38 Sephora 13 3.60 3.54 Foot Locker 13 3.60 3.05 Best Buy 13 3.62 2.80 Walgreens 13 3.63 2.86 DICK’s 13 3.65 3.31

Lowe’s Home Improvement 13 3.83 3.02

Target 13 3.94 3.45 Reliance Digital 13 3.97 2.68 Staples 13 3.98 2.62 Craftman 13 4.20 2.91 STIHL 13 4.22 2.69 IKEA 13 4.37 3.48 Tommy Hilfiger 13 2.95 3.28 Subway 13 3.23 3.31

(44)

44

Nordstrom 13 3.31 3.40

MAC Cosmetics 13 3.20 3.65

The cheesecake factory 13 2.94 3.66

Toys”R”Us 13 3.58 3.69

Whole Foods Market 13 3.74 3.83

Dunkin Donuts 13 2.98 3.89

BMW 13 3.78 3.97

The body shop 13 3.58 3.98

Lego 13 3.52 4.20

Topshop 13 4.11 4.45

Table 2

Mean Center Values per Brand

Brands Utilitarian Mean Hedonic Mean

Primark -0,56 -0,62

Topshop 0,55 1,11

Dunkin Donuts -0,57 0,55

The Cheesecake Factory -0,62 0,32

Staples 0,43 -0,72

The Body Shop 0,03 0,65

Whole Foods Market 0,18 0,49

Walgreens 0,07 -0,48

Lego -0,04 0,86

(45)

45 Reliance Digital 0,41 -0,66 BMW 0,23 0,63 MAC Cosmetics -0,36 0,31 STIHL 0,66 -0,65 Nordstrom -0,25 0,06 Kohl's -0,16 0,05 Colors of Benetton -0,27 -0,20 IKEA 0,81 0,14 Subway -0,33 -0,03 Wendy's -0,39 -0,22 Tommy Hilfiger -0,61 -0,06 Target 0,38 0,11 Sephora 0,04 0,20 Foot Locker 0,04 -0,29 Toys"R"Us 0,03 0,35 Best Buy 0,06 -0,54

Lowe's Home Improvement 0,27 -0,32

Old Navy -0,08 0,05

DICK's 0,09 -0,03

Burger King -0,67 -0,62

Table 3

Number of posts per brand

(46)

46

Topshop 59

Lego 49

Whole Foods Market 47

Tommy Hilfiger 34 Toys”R”US 31 Walgreens 29 Primark 29 Best Buy 20 IKEA 18

The body sjop 17

Reliance Digital 16 Staples 16 Old navy 13 Foot Locker 11 STIHL 11 Kohl's 9 Reliance Digital 9

Dick’s Sporting Goods 8 United Colors of Benetton 8 Lowe’s Home Improvement 7

MAC Cosmetics 6

BMW 6

Craftman 5

(47)

47

Table 4

Check for normality: Skewness and Kurtosis Statistics

Skewness Kurtosis

Statistic Std. Error Statistic Std. Error

Hedonic .190 .111 -.520 .222

Utilitarian .332 .111 -.493 .222

Likes_ratio -.124 .111 -.525 .222

Shares_ratio .068 .111 -1.193 .222

Table 5

Check for independence of variables and multicollinearity Collinearity Statistics (likes) Durbin -Watson (likes) Collinearity Statistics (shares) Durbin -Watson (shares)

Tolerance VIF Tolerance VIF

Purpose .915 1.093 1.303 .915 1.093 1.618

Type .913 1.096 .913 1.096

Utilitarian .184 5.441 .184 5.441

Hedonic .206 4.853 .206 4.853

The Cheesecake factory 5

Subway 5

Dunkin Donuts 3

Target 3

Wendy's 3

(48)

48 Interception _Purpose _Utilitarian .253 3.954 .253 3.954 Interception _Purpose _Hedonic .323 3.098 .323 3.098 Interception _Type _Utilitarian .273 3.659 .273 3.659 Interception _Type _Utilitarian .358 2.794 .358 Table 6

Means, Standard Deviations Correlations (likes_ratio)

Variables M SD 1 2 3

1. Likes_ratio -.3026 .59998 -

2. Hedonic 3.5228 .58517 -.098* -

3. Utilitarian 3.6251 .38211 -.282** ,282** -

** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed)

Table 7

Means, Standard Deviations Correlations (shares_ratio)

Variables M SD 1 2 3

1. Shares_ratio -1.3538 .82109 -

2. Hedonic 3.5228 .58517 -.022* -

3. Utilitarian 3.6251 .38211 -.027 ,282** -

(49)

49 Appendix D: Figures

Figure 1 Entertaining and Informative posts per brand

(Craftman, Dunkin Donuts, Lego, Primark, Staples, The Body Shop, The Cheesecake Factory, Topshop, Walgreens, Whole Foods Market)

(BMW, IKEA, Kohl’s, MAC Cosmetics, Nordstrom, Reliance Digital, STIHL, Subway, United Colors of Benetton, Wendy’s)

(50)

50

(Best Buy, Burger King, Dick’s Sporting Goods, Foot Locker, Lowe’s Home Improvement, Old Navy, Sephora, Target, Tommy Hilfiger, Toys”R”US)

Figure 4 Type of post per brand

(Craftman, Dunkin Donuts, Lego, Primark, Staples, The Body Shop, The Cheesecake Factory, Topshop, Walgreens, Whole Foods Market)

(51)

51

Figure 5 Type of posts per brand

(BMW, IKEA, Kohl’s, MAC Cosmetics, Nordstrom, Reliance Digital, STIHL, Subway, United Colors of Benetton, Wendy’s)

Figure 6 Type of posts per brand

(Best Buy, Burger King, Dick’s Sporting Goods, Foot Locker, Lowe’s Home Improvement, Old Navy, Sephora, Target, Tommy Hilfiger, Toys”R”US)

(52)

52

Figure 7 Histogram: before and after normalization of likes

(53)

53

Figure 9 P-Plots of utilitarian and hedonic values

The effect of social media content characteristics on online consumer engagement

Master Thesis