Pictures, Facebook and Generation Z : an exploratory study of Generation Z and its online content preferences

(1)

Master Thesis Marketing

Pictures, Facebook and Generation Z:

An exploratory study of Generation Z and its online

content preferences.

Author: Paulina Daria Kowalska (10828052)

MSc. in Business Administration, Marketing Track

Amsterdam Business School, University of Amsterdam

Faculty of Economics and Business

Supervisor: Adriana Krawczyk, PhD

(2)

1

Acknowledgements

I would like to address my sincere gratitude to my Thesis Supervisor, Adriana Krawczyk, for the great patience, understanding and support within the thesis writing process. I deeply appreciate the obtained feedback and willingness to help me with creating a representative research paper.

I would like to thank the anonymous observers, who agreed on participation in the study and contributed to elicitation of reliable findings.

Special thankfulness is addressed to my fiancé, Lukas Maik, who provided me with both mental and physical support during thesis writing process. Last but not least, I would like to thank Steven Hooker, for the strong motivation to wake up early in the morning and go to bed late at night for the sake of working on my Master Thesis.

With my best regards,

Paulina Kowalska

Statement of originality

This document is written by Student Paulina Kowalska who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

2

Abstract

The subject of this Master Thesis is Generation Z (Gen Z) and its online content preferences with regard to pictorial material.

The literature concerning Generation Z has been sparse, with no particular insights into Generation Z engagement in Social Network environment. The author sheds the light on engagement creating content and define 26 categories, which describes common pictorial material. The findings show that sexual, humorous and hedonic pictures are positively correlated with Facebook engagement. The crucial finding is that negative message character dominates significantly over the positive, what implies that Generation Z manifests the reverse trend in terms of passing along online content. This phenomenon gives direction to further research which should be conducted in order to form grounded knowledge. What is more, the author highlighted major correlations between categories what constitute an essential information for marketers, targeting Generation Z by visual marketing campaigns.

(4)

3

List of Tables and Figures

Tables Table 1: Comparison of Generation X, Y and Z (Levickaitė 2010)……….……11

Table 2: Scheme of picture categories……….…….33

Table 3: Cohen’s kappa for an inter-coder agreement about the scheme of 26 categories…….…….34

Table 3: Positive significant Pearson correlations between categories……….…...39

Table 5: Negative significant Pearson correlations between categories……….…….40

Table 6: Pearson correlation for sexual category and Facebook engagement………...41

Table 7: Pearson correlation for romantic category and Facebook engagement……….…....42

Table 8: Pearson correlation for hedonism category and Facebook engagement……….……...43

Table 9: Pearson correlation for humorous category and Facebook engagement………...44

Table 10: Overview of tested hypotheses………44

Figures Figure 1: Conceptual model for the study………..……22

Figure 2: The outline of Facebook fan page, used for the study………24

Figure 3: Fan page community’ characteristic (age)………...25

(6)

5

Figure 5: Example of image……….…27

Figure 6: Example of text……….27

Figure 7: Example of imagetext………..…....28

Figure 8: Sample posts ordered by Likes, Comments & Shares (Engagement rate)……...…...29

Figure 9: Sample descriptions of pictures conducted by Observer 1 and Observer 2………...32

Figure 10: Sample of KWIC Concordance analyses presented in Excel sheet………..…32

Figure 11: Sample of binary coding for pictures with the highest Engagement……….…....…....35

Figure 12: Sample of binary coding for pictures with the lowest Engagement…………....……..35

Figure 13: Categories dominance among pictures with the highest Engagement……….…...36

Figure 14: Categories dominance among pictures with the lowest Engagement…….………...36

Figure 15: Compilation of dominant categories among pictures with the highest and the lowest Engagement………..………37

(7)

6

Chapter 1: Introduction

1.1 Background

Generation Z (Gen Z) describes individuals born since 1995, fully integrated with online environment (Lichy 2012). Gen Z preferences in terms of online message content have been overlooked but can be significantly important for the companies. Organizations can benefit from knowing the type of engagement-creating messages, by conducting marketing campaigns that enhance customers’ communication and, in turn, brand recognition. Moreover, this may contribute to positive electronic word of mouth (eWOM) and creation of viral message (Cruz and Fill 2008). According to Smith (2012), targeting Generation Z is not an easy task and often results in impasse, because young users are not willing to build a relation with the company and manifest engagement. Even though Generation Z has been already studied, there are still lack of findings in terms of engagement-generating content concerning particularly representatives of Generation Z. Similar situation is visible in the field of electronic word of mouth and viral marketing, where studies are conducted mainly on Generation Y, without taking into account samples of the youngest generation (Ahn 2012). Therefore, an exploratory research is needed to examine the online content preferences among Generation Z.

Examining online content preferences requires determining general portrait of Generation Z and finding its most commonly used means of communication. Levickaitė (2010) states that Generation Z differs substantially from other generations and is the one fully integrated with technological devices and the Internet (Levickaitė 2010). Online environment is therefore considered as a natural place to spot Gen Z and conduct the research. Among multiple online tools, Generation Z took a liking to Social Networking Sites (SNSs) (Lenhart 2015), where instant communication with friends and family as well as self-expression and entertainment is possible. Through their hybrid characteristic (Mangold and Faulds 2009), social networking

(8)

7 sites not only connects individuals, but enable them to act with companies as well. Therefore, social networks constitute a platform for both leisure and business. The most popular Social Networking Site among Gen Z representatives is Facebook (Peslak 2011), where companies have already set their presence. Facebook enables posting various online content ranging from links and text to videos and pictures. This content causes users engagement, by liking, commenting and sharing; nevertheless, studies on Facebook shows that most popular as well as most involving are pictures, because they grab user’s attention at the first sight (Kwok and Yu 2013). As a result, the study is being narrowed to pictures as a basic tool used both in marketing and social networking environment. Therefore, for the naming purpose, the term content will be directly related to pictures. According to Paivio (1986), pictures convey the meaning better than words, thus by analyzing pictures, the preferences can be outlined (Paivio 1986, Yoon and Han 2012). The picture will be therefore a medium, which indicates the online preferences of Generation Z.

1.2 Research goal

Gen Z is a rising generation, starting to play a significant role in the online world. Consequently, understanding their online content preferences is necessary for companies to build reliable, long-lasting relations in social networks. While previous research has outlined Generation Z characteristic (Levickaitė 2010), specified the most commonly used means of communication (Lenhart 2015), motives to use them (Baek 2011, Ong et al. 2011, Martín and García-Sánchez 2013) and pictures as a most engaging medium (Kwok and Yu 2013), current study will de facto point out major online preferences of Generation Z in terms of pictorial content in particular. In other words, most engaging pictures will be highlighted, specifically described and analyzed. This is going to be achieved by using both qualitative and quantitative methods,

(9)

8 with a particular emphasis on netnography research conducted on Facebook. Therefore, the study will contribute to the existing literature on Generation Z, eWOM and visuals.

1.3 Research question and sub questions

This study aims to answer the following research question:

“What kind of pictorial content contributes to increased Facebook engagement among representatives of Generation Z?”

In order to provide an answer, the literature review will shed a light on main concepts related to the topic and attempt to answer the following sub questions:

 What is Generation Z? Why it is important to study Generation Z behavior?

 What are Social Networking Sites (SNSs)? What are the main motives for Generation Z to use SNSs? What is the most favorable SNS of Generation Z and why?

 What is engagement and what does it imply? What type of content SNS users tend to engage in?

 What is the role of pictures in marketing and Social Media?

1.4 Contribution

Theoretical

The literature concerning Generation Z and its online behavior has been scarce (Lichy 2012, Ahn 2012), probably because of the fact that the phenomenon has emerged recently. While previous studies focused on Generation Z motives and general characteristic, this study will take a look at online content preferences from eWOM perspective. Therefore, major contribution will build upon previous findings and develop significantly the marketing literature. That would be done in a threefold way. First of all, field dedicated to Generation Z will gain new information about online preferences in terms of pictorial messages. Secondly,

(10)

9 eWOM and viral marketing will gain new insights into the new user group and their engagement manner. Last but not least, pictorial studies will be extended by information about engagement creating patterns.

Managerial

The newly emerging Generation Z starts to gain importance in a digital world. Therefore, more and more organizations tend to include it as their target group and create promising virtual relations. This may be done through social media and social networking sites (SNSs) in particular. Social Media Marketing Industry Report (Stelzner 2011) reveals that increasing number of companies turn to establish their presence at SNSs. Firms aim to create loyal communities, distribute content and collect relevant feedback which influence brand perceptions and provide brand support (Michaelidou et al. 2011). Nonetheless, Weinberg and Pehlivan (2011) emphasize that there is still high degree of uncertainty with respect to budgetary spending to SNSs in terms of return on marketing investment (ROMI). Therefore, knowing the message recipients, their preferences and characteristics, plays a crucial role in creating tailored marketing campaigns and efficient allocation of financial resources. Nevertheless, due to lack of studies on Generation Z online content preferences, companies need to experiment and expose themselves to the risk of failure. This study aims to mitigate the risks by providing illustrated description of online content preferences presented by Generation Z, which in turn facilitates electronic word of mouth and viral spread of marketing message.

1.5 Thesis outline

The study will start with describing the existing literature about Generation Z characteristics, Social Networking Sites as their main online destination, Facebook engagement in general and pictures as a medium which conveys the message. Therefore, Chapter 2 will outline the research environment and create necessary links between each category. As a result, the literature gap

(11)

10 will be identified. Chapter 3 provides the conceptual framework for the study, where both hypotheses and conceptual model will be developed. Chapter 4 outlines the methodology of the study, where research design and sample measurement will be described. Study analysis and results will be delivered in Chapter 5. Chapter 6 and Chapter 7 will focus on general discussion and conclusion respectively.

Chapter 2: Literature review

2.1 Generation Z

Generation Z (Gen Z) stands for individuals born since 1995, who constitute the youngest human generation known so far (Lichy 2012). Gen Z has been entirely integrated with the Internet, therefore it is often referred as “net savyy youth” (Levin and Arafeh, 2002), “homo-zappiens” (Veen and Vrakking, 2006) or “born digital” ” (Palfrey and Gasser, 2008). Mission and Ministry (2015) describes Generation Z as 21st_{century world’s generation – the dot-com}

kids, digital natives, Generation Media. According to Levickaitė (2010), representatives of Gen Z show up-ageing trend: they grow up faster and start education earlier. Due to facilitated Internet access, marketing exposure starts relatively early in comparison to previous generations (Levickaitė 2010). Gen Z is characterized by multi-tasking: moves swiftly from one task to another, putting emphasis on speed rather than accurateness. Being just a few clicks away from the ocean of knowledge, Generation Z is considered as entirely digital, where their social interactions turn to be digital as well (Levickaitė 2010).

According to scholars, Generation Z is significantly different from previous generations (Selwyn 2009, Levickaitė 2010, Lichy 2012), what gives an important argument for research.

Levickaitė (2010) provides a complete comparison of last three generations (X,Y,Z) in terms of social, technological and historical environment, which has been illustrated in Table 1.

(12)

11

Table 1: Comparison of Generation X, Y and Z (Levickaitė 2010).

Gen Born Social environment Technological environment Historical environment X 1960s - 1974 Cultivation of staying at home, delaying career;

Experiencing the rise of computer, videogames and

the Internet; seeing it as commercial objectives;

Fall of Berlin Wall, collapse of USSR and

Yugoslavia; Y 1975 - 1990s Behaviors and attitudes shaped by events and leaders of

this time;

Communicating though technology;

Rejection of past norms and assumptions, more involved in technologies

than daily events; Z 1995 - 2000s Instantly online, critical thinking manifestation; Fully integrated communication via technological devices;

Social networking, world without limits of time and

space;

According to the information given, Gen Z is the first Generation fully embedded in cyber social environment and online world. As opposed to Gen Y, they are instantly online, already integrated with technological devices. Levickaitė (2010) highlights that Generation Y is more conservative, preferring real life meeting with friends and family rather than just online interaction. Gen Y constitute a transitional generation, a bridge between conservative Gen X and independent Gen Z (Levickaitė 2010).

From a marketing perspective, Gen Z stands out from other generations as well (Tapscott and Williams 2008, Selwyn 2009, Eleftheric-Smith 2012). Eleftheric-Smith (2012) highlights its independency, suspiciousness and critical thinking. Tapscott and Williams (2008) note that Gen Z are not passive consumers, but has a need for customization and co-creation. They are portrayed as autonomous and highly sociable with a particular need to be a part of Internet social environment, embodied in a form of social media (Lenhart 2015).

Nevertheless, studies on Gen Z in online environment has been scarce (Ahn 2012). Eleftheric-Smith (2012) points out that many firms experience a stalemate in communication with Gen Z.

(13)

12 They tend to fall into a “cool” trap which in turn weaken the relationship by limiting Gen Z engagement. Indeed, Eleftheric-Smith (2012) states that “marketers have to offer something extra rather than asking for engagement”. One of the way is to build credible relationships, not necessarily focused on being “cool”. It is about the right balance between what is trendy and what triggers passion and develop trust (Selwyn 2009). Nevertheless, many companies found this statement too abstract and cannot translate it for the successful marketing campaign (Eleftheric-Smith 2012). Therefore, more research is needed to define Generation Z preferences, resulting in tailored marketing messages.

As stated before, Generation Z is considered as fully integrated with online environment, with a particular emphasis on social media (Levickaitė 2010, Lenhart 2015). According to the researchers, the greatest popularity gain Social Networking Sites (SNSs), which became main online destination for the youngest generation (Selwyn 2009, Peslak 2011, Levickaitė 2010, Lenhart 2015). The next section aim to shed a light on the topic of SNSs with the emphasis on Gen Z activities.

2.2 Social Networking Sites

Social Networking Site (SNS) is defined as a web-based service allowing individuals to create a profile which is embedded in an organized network. An individual may generate a list of profiles, which belong to others, in order to set connections. These connections enable to exchange information in the form of text, pictures, videos, links etc. (Boyd and Ellison 2007). An initial research suggests that SNSs satisfied the need of companion and emotional support (Joinson 2008) by keeping in touch with friends and family (Urista et al. 2008). Recent studies show that people turn into SNSs mainly for entertainment and enjoyment, continuing cultivating relations and maintaining social support (Baek et al. 2011). Nowadays, SNSs are the most visited web pages on the Internet, still extending over new users (Lee et al. 2014).

(14)

13 Social Networking Sites gained prominence among Gen Z representatives, mainly by enabling peer connections without adult surveillance, facilitating identity and exploring social relations (Ong et al. 2011). SNSs enable young users to co-create the website content, fulfilling the need of creation and customization (Heinrichs et al. 2011). Many SNSs websites allow to register from 13 years old and according to Lenhart (2015), 76% of teens aged 13-17 are active SNSs users. Comparing to adults, young users are twice as likely to poses a social networking account (Peslak 2011).

Selwyn (2009) emphasizes that the youngest generation uses SNSs for promotion and self-expression instead of listening and learning from others. This statement is shared by Ralph et al. (2011), who write that SNSs create a venue for self-expression and identity creation. Ong et al. (2011) study the concept of narcissism and contribute that more narcissistic adolescents update their social profiles more often than their less narcissistic peers do. The authors emphasize that the need for self-expression generates self-interest, what reorganize the adolescents’ value hierarchy making self-expression more important than contacting friends (Ong et al. 2011). McAndrew and Jeong (2012) repeat the need for friend connections and expressing self, however do not state the order of importance. Martín and García-Sánchez (2013) conducted a questionnaire among young people and concluded that the main reason why they use SNSs was an entertainment purpose. That was confirmed by Chen (2013) who stresses out that the crucial benefit from using SNSs is the perceived enjoyment. As a result, the literature notes a change in Gen Z motives for using SNSs: from nurturing and maintaining existing relationships, enhancing one’s reputation to entertainment (Tosun 2012). This specifies the direction for further studies and companies operations.

Facebook

Among many different Social Networking Sites, Facebook is considered as the most popular and the largest social networking tool (Peslak 2011; Coulter and Roggeveen 2012). Founded in

(15)

14 2004 by Mark Zuckerberg, its main goal was to improve communication between Harvard University students (Hansson et al. 2013). Over the years, Facebook gained popularity among adults and teens, becoming the world’s most successful social network (Hansson et al. 2013).

Lenhart (2015) reports that Facebook is most used SNS among representatives of Generation Z. 71% of teens use Facebook, 52% go for Instagram and 41% take a liking for Snapchat (Lenhart 2015). Furthermore, Lenhart (2015) notes the relationship between Facebook frequency use and parental income. The higher the family income, the less likely Gen Z is using Facebook. This interdependence was emphasized by Peslak (2011) as well.

Studies show that Facebook attracts user’s attention by building on curiosity, enjoyment and excitement; people are interested in each other, pass along various information and play games with each other (Palmer and Koenig-Lewis 2009, Hansson et al. 2013). Facebook enables instant communication with peers and family as well as serves as a self-representation tool by posting pictures and textual descriptions (Baek 2011).

The use of Facebook attracts not only young users but also companies, making it an important platform for leisure and business (Mangold and Faulds 2009, Chu and Kim 2011; Araujo and Neijens 2012; Hansson et al. 2013). Mangold and Faulds called Facebook “a hybrid element of promotional mix”, because it enables companies to contact with and drives the conversation between customers. Michaelidou et al. (2011) highlight major motives why firms use SNS, including Facebook. Most of all, they aim to build direct relationships with customers, create communities, collect feedback, identify new business opportunities, distribute content, increase traffic to their websites which in the end result in brand support (Michaelidou et al. 2011). Araujo and Neijens (2012) study brand presence at SNSs and highlight the importance of user’s engagement which measures company’s performance and constitutes a dialogue between users and the company (Araujo and Neijens 2012).

(16)

15 2.3 Facebook engagement

Facebook engagement is determined by likes, shares and comments made by network users (How is engagement rate defined? 2014). Engagement reflects emotional arousal: Facebook users manifest their affective evaluation of posts or comments, which can be either positive, negative or taking into account both (Berger and Milkman 2011, Alhabash and McAlister 2014). Facebook users’ engagement implies information-passing behavior: other users see what person has liked, what comment has been made and what content has been shared. Therefore, the piece of information can be captured and passed on.

In this sense, engagement contributes to electronic word of mouth (eWOM), an interpersonal virtual communication about a product, which include its pros and cons (Hennig-Thurau et al. 2004, Cheng and Zhou 2010). Because of user-generated opinions, eWOM enhances customers’ trust significantly more than purely marketing information provided by companies (Gruen et al., 2005). While spreading rapidly, eWOM may transform into viral message, reaching a significant number of people (Cruz and Fill 2008, Ferguson 2008). In the marketing perspective, this phenomenon is framed as “viral marketing” or “viral advertising”. Using low capital expenditures, viral marketing aims to maximize the reach, achieving the level of awareness comparable with TV advertisement or higher (Cruz and Fill 2008; Kaplan and Haenlein 2011). In these sense, successful viral marketing campaign becomes a holy grail for most of the companies. Therefore, it is crucial to determine what type of content boosts customer’s engagement (Cruz and Fill 2008, Chiu et al. 2013).

Engagement-creating content

The literature states that people engage in messages which create emotional arousal (Chu 2011, Berger 2011, Kaplan and Haenlein 2011, Alhabash et al. 2013). Chu (2011) emphasizes emotions as a crucial element of information-passing behavior. She points out that surprising

(17)

16 content facilitates the process of sharing and talking about the message. Moreover, a positive message influences an attitude toward the ad, brand and passing on online content. This statement is shared by Berger (2011), who concludes that positive content is more viral than negative content. Similarly, high-arousal content is more prone to get viral that low-arousal content, where high-arousal reflects both positive and negative emotions. Anger or anxiety (high-arousal) obtain more engagement than sadness (low-arousal), awe (high-arousal) is more influential than joy (low-arousal) (Berger 2011).

Chen and Berger (2013) shed a light on controversy, which according to many scholars (Mangold and Faulds 2009; Alhabash et al. 2013) leads to users’ engagement. They found out that this statement is only true at a low and moderate level of controversy. Beyond moderate level, additional dose of controversy increases discomfort what actually decreases likeliness of discussion (Chen and Berger 2013).

Ho and Dempsey (2010) state that passing along is positively influenced by the entertaining content and socialization. This includes humorous (Kaplan and Haenlein 2011; Tucker 2011) and sexual messages (Alhabash et al. 2013; Doornwaard et al. 2014). Golan and Zaidner (2008) contend that 91 per cent of viral messages contain an element of humor followed by sexual appeals with 28 per cent. Tucker (2011) points out that humorous ads build positive attitude toward the brand as well as increase virality. Moreover, she enriches the field of controversy with the finding that outrageous messages increase virality, nevertheless reduce ad persuasiveness. Funny visuals did not record reduced persuasiveness. Carter (2012) highlights the importance of funny visuals, which gain significant interest of Facebook users.

In terms of sexuality, recent research carried out by Doornwaard et al. (2014) provides important insights about sexual and romantic references displayed on Facebook by Dutch adolescents, aged 11 to 18 years old. According to the results, half of the participants were

(18)

17 actually the displayers of romantic and sexual content. In comparison with non-displayers, they were older, more sexually experienced and more active on the social platform. Furthermore, romantic references obtained an advantage over sexual references (204 to 67 references) in terms of Facebook profile posting.

Other important factors which increase users’ engagement are credibility (Coulter and Roggeveen 2012), utilitarian and hedonic values (Chiu et al. 2013) as well as altruism and self-concept (Kaplan and Haenlein 2011). Kaplan and Haenlein (2011) emphasize the value of true stories which tend to be more interesting than typical corporate advertising. Moreover, they point out that viral messages should be memorable, what may be especially obtained with visuals (Starch 1966).

Regarding the category sorts (e.g. text, pictures, videos or links) the literature on Social Networking Sites remains consistent in that matter. Studies show that posts with photo or text obtain more users’ engagement than posts with links or embedded videos (Kwok and Yu 2013). Other authors emphasize that sharing pictures became one of the main activities users display on Facebook (Palmer and Koenig-Lewis 2009, Hansson et al. 2013). Studies on Twitter, other social networking site, present that pictures gain most engagement among hashtag, digit, quote, picture and video (What fuels Tweet’s engagement? 2014). Therefore, taking closer look at pictures becomes necessary for further discussion.

2.4 Pictures

“A picture is worth a thousand words.” First time this idiom appeared in a newspaper in 1911 as a form of advertising (Speakers Give Sound Advice. 1911) and meant that complex idea can be illustrated in a still picture. In the literature, picture has been known as involving, attention-getting device enhancing message persuasiveness (MacKenzie 1986, Miniard et al. 1991, Zhang and Zinkhan 2006); Miniard et al. outlined that product attitudes may be changed or

(19)

18 formed by pictures. Throughout the years the power of image has become an indispensable tool for marketers, used in offline and online advertising (Yoon and Han 2012). Pictorial material gains significant interest in Social Media and seem to elicit most engagement (What fuels Tweet’s engagement? 2014). Because of the fact that pictures are most frequently used as a posting content in Social Networks, this study will directly relate the term content with term

picture. The next paragraph will open the literature review about pictures and their influence

on message processing.

According to the researchers, pictures have a meaningful influence on human mind and memory (Starch 1966; Paivio 1971; Marks 1973; Morris and Stevens 1974; Bower et al. 1975; Lutz and Lutz 1977; Childers and Houston 1984; Mitchell 1995). Starch (1966) discovered that when printed ad includes a picture, objects from experiment show greater tendency to recall the ad as it was without the picture. In his study about visual imagery, Marks (1973) provides a proof that images play a crucial role in remembrance. Morris and Stevens (1974) confirm this theory and add that when a group of images are linked together, the recall is greater.

Over time, the literature was enriched by a new term called “picture superiority effect”, deeply examined by Childers and Houston in 1984. The effect says that concepts are much more memorable when they are shown as pictures instead of words. Moreover, pictures are considered to convey the meaning more explicitly than words (Paivio 1986). Plentiful studies have proven that pictures are more readily recognized than words (Paivio 1969; Lutz 1978). W.J.T. Mitchell (1995) notes that the "pictorial turn" supplants the "linguistic turn", highlighting the greatness of visuals with the beginning of late twentieth century.

While conducting a research on picture and words, scientists begin to realize a significant relationship between pictorial and verbal materials (Knowlton 1966; Bower et al. 1975; Edell and Staelin 1983; Mitchell 1995; Fletcher 1998). In the first section of his paper, Knowlton

(20)

19 (1966) analyzes pictures in isolation from words. After that he embeds pictures in the verbal context. He concludes that learning is involved in the process of picture interpretation and verbal context determines picture’s meaning. Bower et al. (1975) states that people remember nonsensical, random pictures remarkably better if there is an interpretation what they are about. Edell and Staelin (1983) confirmed this data with a statement that pictorial messages in advertisements are processed differently whether a reference point (interpretation) to encode the message is provided. A dot over the “i” is put by Mitchell (1995), who introduces a term called imagetext. Mitchell assumes that visual and verbal experience, pictures and words are woven together. Every medium becomes a “mixed medium” and form one concept. Mitchell’s findings confirm the superiority effect of picture over the written word, but more importantly, highlight the inconceivable relationship formed by image and text. Recent studies conducted by Yoon and Han (2012) confirm the rise of favorable thoughts done by picture and text together in the Internet environment. As a result of using both picture and text, the information is processed better and suggests to create greater people engagement than text or picture alone (Yoon and Han 2012).

According to abovementioned literature, pictures has significant influence on human mind and memory, conveys meaning better than words. The “pictorial turn” is particularly visible in Social Networks, where imagetext became a popular tool to express ideas, motives and preferences (Yoon and Han 2012). Based on that, it is justified to use pictures as a tool to define Generation Z online content preferences.

2.5 Literature gap

The literature review has shown that Generation Z differs substantially from other generations (Levickaitė 2010, Lichy 2012). Most of the researches conducted studies on Generation X and Generation Y, living behind representatives of Generation Z (Levickaitė 2010). As a “net savyy

(21)

20 youth” (Levin and Arafeh, 2002), Generation Z shows remarkable online activity, nevertheless little is known about its online content preferences (Ahn 2012, Doornwaard et al. 2014). Current research on eWOM and viral marketing addresses general tendencies in engagement-creating content, without particular emphasis on user’s age or generation.

Therefore, Chapter 3 will refer to the existing gap and shed a light on hypothetical answers.

Chapter 3: Conceptual Framework and hypotheses development

3.1 Key motives and engagement-creating content

Ho and Dempsey (2010) conducted a study on motives which determine passing along online content. Their conclusion is that marketing message has a greater chance of success when it resonates with key motivations of targeted audience (Ho and Dempsey 2010). Therefore, applying these results to current study, the literature gap concerning Gen Z engagement-creating content may be bridged by confronting the key motivations of Generation Z with general findings about engagement-creating content from the literature. Therefore the key motivations of Gen Z to use SNSs will be revised and combined with general engagement-creating features.

As stated by Baek (2011), Generation Z seeks to satisfy the need of companion and relationship maintenance with peers and family. Ong et al. (2011) underline the need for self-expression, which in turn is highly correlated with self-interest. García-Martín and García-Sánchez (2013) contributes that Generation Z looks for entertainment and enjoyment in their SNS use. Therefore, the main motives of Generation Z are summarized as follows: maintaining relationships, self-expression and entertainment.

(22)

21

Maintaining relationships

The literature on young adolescences highlights that sexual and romantic references are engagement-creating content among Gen Z (Doornwaard et al. 2014). This contributes to the relationship area in terms of looking for the partner and building relations between man and woman. Doornwaard et al. (2014) report that half of the study respondents showed interest in sexual and romantic content, with a clear indication of romantic premises. Therefore, the following hypotheses are predicted:

Hypothesis 1: Sexual content contributes to increased Facebook engagement among Gen Z representatives.

Hypothesis 2: Romantic content contributes to increased Facebook engagement among Gen Z representatives.

Self-expression

Regarding self-expression, the literature about engagement-creating content has been scarce, emphasizing primarily hedonic values (Chandon et al. 2000, Chiu et al. 2013). According to Chiu et al. (2013), hedonic values cause greater willingness to share message with others, building on self-interest and gaining pleasure. The authors emphasize that hedonic outcome may take the form of self-expression and be manifested by sharing content expressing one’s diversion (Chiu et al. 2013).

Hypothesis 3: Hedonic values contribute to increased Facebook engagement among Gen Z representatives.

Entertainment

Kaplan and Haenlein (2011) connect passing along online content with humorous messages. This fulfills the need of entertainment manifested by the youngest generation. Humorous

(23)

22 H3

H4

content draws user’s attention (Zhang and Zinkhan 2006) and gives the importance to message arguments. It is said that humorous content increases the favorable attitude towards brand and ad. According to the researchers, humorous messages are prone to get viral (Kaplan and Haenlein 2011; Tucker 2011; Chiu et al. 2013).

Hypothesis 4: Humorous content contributes to increased Facebook engagement among Gen Z representatives.

3.2 Conceptual Framework

The conceptual model of the study is illustrated in a Figure 1. In summary, the hypotheses which are going to be tested look as follows:

H1 - Sexual content contributes to increased Facebook engagement among Gen Z representatives. H2 – Romantic content contributes to increased Facebook engagement among Gen Z representatives. H3 - Hedonic values contribute to increased Facebook engagement among Gen Z representatives. H4 - Humorous content contributes to increased Facebook engagement among representatives of Gen Z.

Figure 1: Conceptual model for the study.

H2 H1 Facebook engagement among Generation Z Sexual content Humorous content Romantic content Hedonic values

(24)

23 According to the model proposed, sexual and romantic content, hedonic values as well as humorous content will positively influence Facebook engagement displayed by representatives of Generation Z (H1, H2, H3, H4). Therefore, sexual, romantic, humorous content and hedonic values constitute a range of independent variables, where Facebook engagement becomes a dependent variable.

Chapter 4: Methodology design

In order to derive an engagement creating content from pictorial material, both qualitative and quantitative methods will be used. Therefore, the study will follow integrated qualitative-quantitative research design called generalization design (Srnka and Koeszegi 2007). Previous studies suggest that generalization model successfully achieves two goals. Firstly, it provides considerable insights into a research area by addressing problems from discovery-oriented perspective. Secondly, it ensures scientific rigor with generation of reliable data (Weingart et al. 2004, Srnka and Koeszegi 2007). The next section will shed a light on the qualitative part of methodology design.

4.1 Qualitative analysis

4.1.1 Applying netnographic methodology

The analyses will start with Netnography, that is an open-ended practice based on participation and observation of distinctive meanings, artifacts and practices of a specific social group (Kozinets 2002). In other words, Netnography focuses on qualitative data collection conducted through the Internet, and is commonly used in social media research (Bartl and Stockinger 2013). According to the literature, Netnography stands for metaphorical and logical data interpretation (Spiggle 1994, Kozinets 2002). Due to the fact that it is often used to provide

(25)

24 generalization, it serves for constituting the grounded knowledge (Kozinets 2002). Therefore, Netnography has particular application to exploratory study, where online content preferences according to pictorial material will be derived.

4.1.1.1 Entrée description

The netnography research will be carried out on Facebook fan page, gathering 1200 polish representatives of Generation Z and around 1600 people in general. Fan page contains pictures posted between 1st December 2014 and 30th March 2015. The pictures differ between imagetext as well as text and image alone. The fan page was created for the widely understood entertainment purpose and in fact became author’s research platform to study young people pictorial content preferences. The author of the research has an administrative rights to the fan page, what provides an opportunity to use Facebook Insights known as general page metrics providing information about user’s activity on the fan page (About Page Insights, 2016). The fan page outline is presented in Figure 2.

(26)

25 Fan page community consists of ca. 1200 representatives of Generation Z, male and female aged 13-17 years old. It covers 74% of all users who liked the fan page. Within this number, there is 21% of girls and 53% of boys (see Figure 3). The community has been created by Facebook advertising campaigns, targeting both masculine and feminine polish representatives of Generation Z. The predominance of male audience occurred naturally, without author’s intervention.

The majority of fan page fans lives in Poland, nevertheless some of them are inhabitants of United Kingdom, Germany and other. Due to the small amount of representatives from other counties, only Polish origin will be taken into account. According to Figure 4, most of fans live in major Polish cities, including Warsaw, Łódź, Poznan, Katowice, Kraków, Czestochowa etc. Due to the extended list of cities provided by Facebook, only the most significant were included in the figure. The same situation applies to countries and languages. Most of the fans speak Polish and English (see Figure 4).

(27)

26

Figure 4: Fan page community’ characteristic (country, city, language).

4.1.1.2 Ensuring ethical standards

The place of research (Facebook fan page) is owned by the author and Facebook Inc. The study is settled in a public environment, where everyone can like/dislike a fan page. In order to ensure naturalistic character of the research, the author will not reveal her identity. For the sake of user anonymity, the name of the fan page will not be revealed.

4.1.1.3 Subject of analysis

The author is going to study 358 pictures which has been posted in a time period between 1st December 2014 and 30th March 2015 on a Facebook fan page. Initially, pictures were derived from English entertaining websites (November 2014), gathering representatives of Generation Z. These pictures were translated into Polish language with the caution to assure meaning consistency. Therefore, the back-translation was conducted (Ohnesorge 2004). Derived pictures gained most engagement, thus they were considered as appropriate to use in the study. Therefore, a pre-selection was conducted.

(28)

27 Pictures that have been posted on the fan page consist of image, text and imagetext. Examples containing English translation are included in Figure 5, 6, 7 respectively. In order to ensure meaning coherency (Ohnesorge 2004), sample pictures were translated with the back-translation.

Figure 5: Example of image.

(29)

28

Figure 7: Example of imagetext.

4.1.1.4 Procedure

The relevant data concerning Gen Z online content preferences are going to be obtained through several steps. First of all, 358 pictures will be ordered according to the Gen Z engagement. Due to administrative rights, the author has a possibility to use Facebook Insights, which enable to select particular information about the posted content such as Post Reach, Engagement and Page Likes. Of the particular interest to this study is Engagement rate constituted by Likes, Comments and Shares (Androutsopoulos 2013; How is engagement rate defined? 2014). Therefore, pictures will be ordered by Engagement rate in order to highlight the most engaging material. Figure 8 presents sample posts ordered by engagement, which is illustrated as a purple bar.

(30)

29

Figure 8: Sample posts ordered by Likes, Comments & Shares (Engagement rate).

The next step will aim to separate pictures according to the highest level of users’ engagement. Additionally, the author assumes to divide pictures with the lowest engagement in order to create a control group for comparative purposes.

Most engaging pictorial material will be used to create the scheme of relevant categories, which will determine the content character, and therefore contribute to answering the research question. In order to create the categories and obtain independent judgments the author decided to acquire a multi-person involvement procedure (Srnka and Koeszegi 2007) consisting of two observers. The observers will be chosen from authors surrounding (convenient sample), assuring possibility of verbal consultation after the procedure finished. Therefore, Observer 1 and Observer 2 will be exchange students, currently residing in Taipei, Taiwan. In order to

(31)

30 provide communication consistency, polish nationality will be selected. The observers will represent both women and men, what suggest more independent judgments.

The observers will be ask to describe each of the most engaging pictures both in terms of general outlook (how does the picture look in general?) and meaning transmission (what meaning does

the picture transmit?). As introduced in section 4.1.1, the observers will use netnography

analysis tool for the metaphorical and logical data interpretation (Kozinets 2002).

The author decided to introduce terms such as “imagetext”, “image”, “text”, “short text”, “long text”, “hedonic values”, “romantic” and “sexual” while explaining the purpose of the task. First of all, the explanation is necessary in order to introduce the terminology derived from the literature. Secondly, it will minimize the risk of misunderstandings and provide the assessment intersubjectivity, that is the universality of word meaning (Srnka and Koeszegi 2007).

After each of the most engaging pictures will be described, the author will start with data elaboration which is supposed to end with categories clarification. After quality check, the qualitative analysis will be finalized by binary coding, showing whether each category appears on the picture or not.

4.2 Quantitative analysis

The quantitative analysis aims to process the nominal data and provide statistical insights into the study. This will be obtained by descriptive statistics, based on the use of SPSS Statistics Software, KWIC Concordance, Microsoft Office Excel as well as Facebook Insights. Each of these tools provides numerical insight into the studied phenomenon, assuring scientific rigor and data reliability.

(32)

31

Chapter 5: Results

5.1 Qualitative analysis

5.1.1 Multi-person involvement procedure

Before the multi-level involvement procedure has been started, 358 pictures were organized according to Engagement rate. This enabled to separate most engaging and least engaging pictures, equal to the amount of 75 and 73 respectively.

The 75 most engaging pictorial materials were used in a multi-person involvement procedure consisted of two independent observers. The author conducted two individual meetings with Observer 1 and Observer 2 respectively. Each observer was asked to describe every picture in English, with respect to general outlook and message transmission. As stated in Chapter 4, the author explained the procedure with giving the expressions “imagetext”, “image”, “text”, “short text”, “long text”, “hedonic values”, “romantic” and “sexual” as sample answers. After the task was completed, each observer was asked to read again the given answers and remove possible repetitions.

5.1.2 Categorization

In the next stage, data were analyzed and interpreted using a deductive-inductive approach, suggested by Srnka and Koeszegi (2007) for text analyses. The author reviewed verbal descriptions (deductive step) using KWIC Concordance, an analytical tool for making word frequency list. The words were therefore organized in an alphabetical order with the corresponding word count. Outcomes from Observer 1 and Observer 2 were analyzed separately. Sample descriptions and KWIC Concordance analyses are included in Figure 9 and 10 respectively.

(33)

32

Figure 9: Sample descriptions of pictures conducted by Observer 1 and Observer 2.

Figure 10: Sample of KWIC Concordance analyses presented in Excel sheet.

The next step was to select similar words or groups of words from each outcome and create the relevant scheme of categories (inductive step). Crucial determinant for category creation was the word count, which indicated the word relevance. For example, if term called “imagetext” occurred 5 times in Figure 10 (the most), that would give the premise to build such a category.

(34)

33 The rounds of preliminary coding were conducted on the complete results from KWIC Concordance. Ordered outcomes of Observer 1 and Observer 2 were juxtaposed and repeatedly compared. This iterative process included changing the word construct (e.g. sexuality  sexual) as well as grouping or eliminating words (bad joke  negativity). When doubts concerning word meaning occurred, the verbal consultations with observers were held. The last preliminary coding round included paying attention on categories intersubjectivity, that is on universality of category meaning (Srnka and Koeszegi 2007). As a result, a scheme of 26 categories has been formulated and presented in Table 2. The presentation of categories includes a general division into outlook (how does the picture look in general?) and meaning transmission (what meaning does the picture transmit?).

Table 2: Scheme of picture categories.

General division Category name

OUTLOOK

MEANING TRANSMISSION

Imagetext, Image, Text

*Long Text, Short Text (under 3 sentences) Animal, Celebrity, Drawing, View, Boy, Man, Girl, Woman

Controversial, Hedonism, Humorous, Sexual, Romantic, Cute, Disgusting

Positive, Negative

Anger, Fear, Sadness, Thrill

*Note: Long Text and Short Text constitute a subcategories for Imagetext and Text. 5.1.3 Quality check

In order to check the consistency of qualitative data, researchers suggest to conduct a Cohen’s

kappa coefficient statistic (Brennan and Prediger 1981). It constitute a reliability check which

measure the inter-coder agreement for qualitative categories. As Brennan and Prediger (1981) indicate, Cohen’s kappa is expressed in numbers from 0 to 1, where moderate agreement starts

(35)

34 from 0.41, substantial agreement is achieved from 0.61, and 0.81 begins almost perfect combination. The evaluation of agreement between Observer 1 and Observer 2 was conducted in SPSS Statistic Software. The total word count for each category served as a reference point for Cohen’s kappa analysis. As a results, the Cohen’s kappa for a scheme of pictures categories was equal to 0,756 what indicates a substantial agreement between observers (see Table 3). Therefore, there is no necessity for deleting categories.

Table 3: Cohen’s kappa for an inter-coder agreement about the scheme of 26 categories.

5.1.4 Binary coding

After obtaining the relevant categories, the next step included implementation of binary coding. This means that each picture was verified again by the author, but this time the categories served as a reference point for picture assessment. The binary coding was necessary in order to start the proceedings of quantitative analysis. Therefore, if the picture could be assigned to particular category, the “1” equivalent was ascribed. When the category was not applicable, then “0” number was attributed. The binary coding procedure was applied for 75 most engaging pictures as well as 73 least engaging pictorial materials (see Figure 11 and Figure 12). The inclusion of the least engaging pictures served as control group, which could be therefore used for further analysis. Furthermore, applying binary coding constituted a double check for the outcomes of two observers, which in turn were consistent with Cohen’s kappa inter-coder agreement.

(36)

35

Figure 11: Sample of binary coding for pictures with the highest Engagement.

Figure 12: Sample of binary coding for pictures with the lowest Engagement.

5.2 Quantitative analysis

The quantitative analysis starts with calculation of the categories which dominate among the most engaging pictures. The author continue to follow a general categories division into outlook and meaning transmission in order to provide more accurate insights. Both high engaging as well as low engaging pictures were taken into consideration.

(37)

36 5.2.1 Descriptive analysis

In order to describe nominal data obtained from binary coding, a bar graph was used. Figure 13, Figure 14 and Figure 15 illustrate the frequency of the identified categories and present the calculation results.

Figure 13: Categories dominance among pictures with the highest Engagement.

(38)

37

Figure 15: Compilation of dominant categories among pictures with the highest and the

lowest Engagement.

The results show the clear dominance of imagetext and short text in terms of outlook as well as

humorous and negative content with regard to meaning transmission, both in terms of pictures

with high and low engagement. This double occurrence confirms the dominant role of the four categories. The pioneering discovery concerns negative message character, which seems to create significant engagement among Generation Z. This finding shed a new light on Berger’s statement, where positive content outperform negative (Berger 2011), and make a significant contribution to the eWOM literature.

Moreover, the picture superiority effect over verbal material is present (image, drawing compared with text). Likewise, the imagetext outperform image and text alone (Mitchell 1995). The special note should be made that long text and short text constitute subcategories for

(39)

38 Clearly visible is the fact that Generation Z engages in content which creates emotional arousal (consistent with Chu 2011, Kaplan and Haenlein 2011, Alhabash et al. 2013). While comparing Figure 13 and Figure 14 (orange and yellow bars), it is evident that categories concerning

meaning transmission occur more frequently among pictures with high engagement than with

low engagement. Although, the difference is not that meaningful in terms of outlook, where the categories prevalence is more converged.

Focusing more on high engaging pictures, it can be stated that sexual pictures get more engagement than romantic pictures. What is more, categories such as controversial, hedonism and positive have a significant value and constitute the high engaging content. Furthermore, for the sample with a dominance of male users, placing the image of boy and man gain more engagement than including visual counterpart of girl or woman.

When it comes to low engaging pictures, factors which may harm the Gen Z engagement are long text as well as placing views and text only. Additionally, images without emotional aspect may be perceived as neuter and do not encourage young users to like, comment or share the visual content.

5.2.2 Exploratory analysis

Going a step further in quantitative analysis, the author would like to check the possible correlations between dominant categories among highly engaging pictures and therefore provide more insights into the studied phenomenon. These findings will outline the set of categories that, while taken together, increase the probability of engagement.

SPSS Statistic were used in order to define the Pearson correlation coefficient for the binary data. Table 4 highlights positive significant correlation between variables, what means that if one variable increases in value, second variable increases in value as well. Likewise, when one variable decreases in value, second variable follows its direction (Srnka and Koeszegi 2007).

(40)

39 This tendency is of particular interest of this study, because it provides significant implications for marketers who aim to create highly engaging pictorial messages addressed to Gen Z in online environment (see Table 4).

Table 4: Positive significant Pearson correlations between categories.

Pearson Correlation

Significant correlations at the 0.01 level (2-tailed)

Significant correlations at the 0.05 level (2-tailed)

0,861 Short Text & Imagetext Humorous & Short Text 0,279

0,546 Celebrity & Man Negative & Boy 0,269

0,553 Negative & Humorous Man & Humorous 0,267

0,490 Animal & Cute Controversial & Anger 0,257

0,454 Cute & Positive Thrill & Boy 0,251

0,444 Girl & Sexual View & Image 0,245

0,438 Girl & Hedonism Man & Negative 0,230

0,432 Man & Woman 0,389 Anger & Fear 0,389 Anger & Sadness 0,389 Boy & Woman 0,387 Negative & Imagetext 0,370 Disgusting & Fear 0,370 Negative & Short Text 0,368 Hedonism & Sexual

0,362 Controversial & Negative 0,338 Romantic & Cute

0,337 Positive & Image 0,329 Long Text & Text 0,317 Controversial & Thrill 0,306 Fear & Thrill

0,298 Humorous & Imagetext

In order to interpret data from Table 4, the author will follow the assumptions made by Srnka and Koeszegi (2007) who state that moderate relationship between variables starts from 0.4, strong relationship begins from 0.6 and very strong interdependence is visible from 0.8 correlation coefficient value. Therefore, the dark blue shadowed part of Table 4 illustrates the most significant correlations between categories. Hence, the verbal part of imagetext should be very short (under 3 sentences) in order to be easily processed and express the relevant meaning.

(41)

40 More interestingly, humorous content is moderately correlated with negative content what sheds a new light on Generation Z cultural preferences and sense of humor.

On the other hand, Table 5 presents negative significant correlations between variables, where increase of one variable results in decrease of the second variable and vice versa. This applies to Image & Imagetext, Short Text & Image and Positive & Humorous, where relationships are formed above the moderate level. Again, the relationship between Positive & Humorous categories pick up the attention. This correlation states that while positive character of the picture increases, the humorous value decreases.

Table 5: Negative significant Pearson correlations between categories.

Pearson Correlation -0,945 Image & Imagetext Long text & Short Text -0,291 -0,862 Short Text & Image Controversial & Positive -0,271 -0,471 Positive & Humorous Controversial & Cute -0,253 -0,359 Positive & Imagetext Cute & Imagetext -0,252 -0,359 Negative & Image Controversial & Animal -0,249 -0,339 Positive & Short Text Girl & Drawing -0,245 -0,337 Image & Humorous Controversial & Romantic -0,238 -0,331 Cute & Negative

-0,310 Humorous & View -0,299 Positive & Boy

The above findings provide a significant added value for marketers as well as shed a new light on the literature. Electronic WOM articles tend to follow the direction of positive content which creates most engagement. For Generation Z, however, this tendency seem to be reversed and create a starting point for further research.

5.3 Hypotheses testing

Four hypotheses will be tested based on the above findings as well as a direct correlation between the category and Facebook engagement. Therefore, Pearson correlation between

(42)

41

sexual, romantic, hedonism, humorous category and engagement will be calculated in order to

provide solid base for decision about supporting or rejecting the hypothesis.

H1 - Sexual pictorial content contributes to increased Facebook engagement among Gen Z representatives.

According to Figure 13, sexual category is placed higher than romantic category but lower than

controversial, hedonism and positive category. Figure 13 does not provide a clear answer about sexual content contribution into Facebook engagement, therefore more insights into this case

should be provided. While looking at Pearson correlation coefficient between categories (see Table 4), the positive relationship between sexual and hedonism is visible, what means that

sexual category is positively correlated with substantially higher category. In order to dispel

doubts, Table 6 provides information about Pearson correlation coefficient between sexual category and Facebook engagement.

Table 6: Pearson correlation for sexual category and Facebook engagement.

Following the outcomes from Table 6, Pearson correlation coefficient for sexual category and Facebook engagement is positive, supported by significance at the 0.01 level (2-tailed). This means when sexual content increase, Facebook engagement also increase. Consequently, H1 is supported.

(43)

42 H2 - Romantic pictorial content contributes to increased Facebook engagement among Gen Z representatives.

Figure 13 presents relatively low contribution of romantic category into Facebook engagement. Its position in the low parts of the graph implies negligible presence in the most engaging pictorial content. As stated in Table 5, romantic category has negative relationship with

controversial category. Because of the fact that controversy has relatively high significance, it

means that the higher the controversial category is on the graph, the lower will be the romantic category. Moreover, the relation between romantic category and Facebook engagement was tested using Pearson correlation coefficient. Findings are presented in Table 7.

Table 7: Pearson correlation for romantic category and Facebook engagement.

With regard to Table 7, Pearson correlation coefficient for romantic category and Facebook engagement is positive, but does not show any statistical significance. This means that increase or decrease of romantic content does not significantly influence Facebook engagement. Therefore, H2 is rejected.

H3 - Hedonic values contribute to increased Facebook engagement among Gen Z representatives.

As stated in Figure 13, hedonic values were considered as hedonism category. Figure 13 illustrates that hedonism reflects comparable level of occurrence as controversial and positive category. What is more, while looking at Figure 14 hedonism is being placed directly below the

(44)

43

humorous category leaving far behind controversial and positive content. This implies the

superiority of hedonism over controversial and positive category. In order to finally check the relationship between hedonism and Facebook engagement, Pearson correlation coefficient was provided and presented in Table 8.

Table 8: Pearson correlation for hedonism category and Facebook engagement.

According to Table 8, Pearson correlation coefficient for hedonism category and Facebook engagement is positive, supported by significance at the 0.01 level (2-tailed). This implies the fact that when hedonic values increase, Facebook engagement also increase. As a result, H3 is supported.

H4 - Humorous pictorial content contributes to Facebook engagement among Gen Z representatives.

According to Figure 13, humorous category is one of the dominant categories which influence Gen Z Facebook engagement. Humor plays a significant role both in terms of high and low engaging pictures (see Figure 15). This constitutes a dual confirmation that humorous content influences positively the Facebook engagement among Generation Z representatives and as a result contributes to its growth. According to Pearson correlation coefficient, humorous category is directly correlated with other dominant categories such as imagetext and negative, what strengthen the initial findings from descriptive statistics. To formulate final conclusion,

(45)

44 Pearson correlation coefficient for humorous category and Facebook engagement is provided (see Table 9).

Table 9: Pearson correlation for humorous category and Facebook engagement.

As stated in Table 9, Pearson correlation coefficient for humorous category and Facebook engagement is positive, supported by significance at the 0.01 level (2-tailed). This means when humorous content increase, Facebook engagement also increase. Therefore, H4 is supported.

5.4 Overview of the Hypotheses

An overview of tested hypotheses is presented in Table 10. The study fully supported three out of four hypotheses. One hypothesis has been rejected.

Table 10: Overview of tested hypotheses.

Hypotheses overview Outcome

H1 - Sexual pictorial content contributes to increased Facebook engagement among Gen Z representatives.

Supported

H2 - Romantic pictorial content contributes to increased Facebook engagement among Gen Z representatives.

Rejected

H3 - Hedonic values contribute to increased Facebook engagement among Gen Z representatives.

Supported

H4 - Humorous pictorial content contributes to increased Facebook engagement among Gen Z representatives.

Pictures, Facebook and Generation Z : an exploratory study of Generation Z and its online content preferences

Master Thesis Marketing