Why we use different social media : how motivations shape the use of various types of social media

WHY WE USE DIFFERENT SOCIAL MEDIA:

How Motivations shape the Use of Various Types of Social Media

Author: Willem Maijvis

Student Number: 10871306

Date of submission: 24-06-2016 (Final Version)

Program: MSc. in Business Administration – Marketing Track

STATEMENT OF ORIGINALITY

This document is written by Willem Maijvis 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.

ACKNOWLEDGMENTS

I would like to thank my supervisor dr. Frederik Situmeang for his help during the thesis process, my dad and my friend Karin Koevoet for their help during the pretest and the translating of the survey items, and my friends Arthur van Rhijn, Robert van de Peppel, and Wouter van der Zanden for their help with statistical issues and their critical feedback. Furthermore I would like to thank my parents, my sister, and all my friends for their support during the thesis writing process, and all the respondents who filled in my survey.

ABSTRACT

In order to examine the research question “How do motivations for social media use influence the amount of use of various social media platforms?” a survey-based quantitative research was performed. Subsequently, a comprehensible segmentation of social media users based on their motivations was made.

Results show that Socialization motivation and Information-seeking motivation are significant predictors of the amount of Average social media platform use. Comparing different types of social media though, it can be observed that Text-based social media use is only significantly influenced by Information-seeking motivation, whereas Image-based social media use is influenced by both Information-seeking motivation and Socialization motivation. Besides this, younger users are significantly more likely to use Mixed social media. Females are significantly more likely to use Image-based social media, while males are more frequent users of Text-based and Video-based social media platforms.

A K-means cluster analysis was performed for the segmentation. Four segments were identified: Agnostics (scoring low on all motivations), Devotees (scoring high on all motivations), Finders (scoring low on Socialization, and high on Entertainment and Information-seeking motivation), and Socializers (scoring high on Socialization and Entertainment, but low on Information-seeking motivation). There were significant differences between these segments in their Average amount of social media platform use, Image-based social media use, and levels of Brand Engagement.

These findings raise several opportunities for further research, and may serve to improve social media marketing efforts by identifying attractive consumer segments.

TABLE OF CONTENTS

LISTS OF FIGURES AND TABLES ... vi

CHAPTER 1. INTRODUCTION ... 1

CHAPTER 2. THEORETICAL FRAMEWORK ... 4

2.2. Motivations for social media use ... 4

2.2.1. Socialization motivation ... 6

2.2.2. Information-seeking motivation ... 8

2.2.3. Entertainment motivation ... 9

2.3. Conceptual framework ... 10

2.4. Segmentations of social media users in earlier research ... 10

CHAPTER 3. DATA AND METHOD ... 15

3.1. Method ... 15

3.2. Sample ... 15

3.3. Measures ... 16

3.3.2. Motivations for social media use ... 16

3.3.2. Amount of use of social media type ... 16

3.3.3. Brand Engagement ... 17

3.3.1. Demographics ... 17

CHAPTER 4. RESULTS ... 18

4.1. Descriptive statistics ... 18

4.2. Hypothesis Testing ... 19

4.2.1. Effects on Average amount of social media platform use ... 19

4.2.2. Effects on amount of use of Mixed social media ... 21

4.2.3. Effects on amount of use of Image-based social media ... 21

4.2.4. Effects on amount of use of Text-based social media ... 23

4.2.5. Effects on amount of use of Video-based social media ... 24

4.2.6. Effects on Brand Engagement ... 25

4.3. Cluster Analysis and Segmentation ... 26

4.3.1. K-means Cluster Analysis ... 26

4.3.2. Description of resulting segments ... 28

CHAPTER 5. DISCUSSION ... 35

CHAPTER 6. CONCLUSIONS ... 42

6.1. Conclusions ... 42

6.2. Limitations and recommendations for further research ... 43

REFERENCES ... 45

Appendix A: Original measures ... 49

LIST OF FIGURES

Figure 1: Conceptual framework ... 10

Figure 2: Means of average amount of social media platform use per segment ... 30

Figure 3: Mean amount of social media use per platform per segment ... 31

Figure 4: Mean Brand Engagement per segment ... 33

LIST OF TABLES Table 1: Means, Standard Deviations, Correlations ... 19

Table 2: Hierarchical regression model of Average amount of social media platform use ... 20

Table 3: Hierarchical regression model of amount of use Mixed social media ... 21

Table 4: Hierarchical regression model of amount of use of Image-based social media ... 22

Table 5: Hierarchical regression model of amount of use of Text-based social media ... 23

Table 6: Hierarchical regression model of amount of use of Video-based social media ... 24

Table 7: Hierarchical regression model of Brand Engagement ... 25

Table 8: Final cluster centers of K-means cluster analysis ... 26

Table 9: Social media user segments and number of cases in each segment ... 28

CHAPTER 1. INTRODUCTION

Social media play an important role in our daily lives. In 2014, 80% of Dutch Internet users made use of social media (Kloosterman and van Beuningen, 2015). Through social media, we stay connected with our friends, family, and acquaintances; show who we are to the world; stay updated with what is happening in the world around us; discover what activities we can do in our city; find out about what our favorite celebrities, brands, and companies are doing; entertain ourselves if we find ourselves bored, and many other things.

An abundance of social media platforms exist, each with varying functions, identities, user populations, symbolisms, manners of speech, inside jokes, and communities. However, much research on social media has taken a generalizing stance towards social media, treating them as a whole (Wilson et al., 2012). Hanna et al. (2011) argue that social media platforms differ and cannot be approached in a standalone manner, and provide a way to understand and conceptualize the different online social platforms as an interconnected ecosystem involving digital and traditional media. Campbell et al. (2014), who typified social media user groups explicitly state the need for further research similar to their own on other social platforms: “First, our focus in the scenario experiment was on one type of social media, Twitter. Further research should attempt to understand other social media applications that exist – Facebook is also routinely used to present consumers with marketing messages” (Campbell et al., 2014, p. 447).

In a cross-national social media user segmentation Alarcón-del-Amo et al. (2015), identified several segments on the basis of their activity on social media platforms, and examined the differences between these segments in Spain and the Netherlands: Introvert Users, Novel Users (Spain only), Versatile Users, and Expert Communicator Users. What is important to note here is that there appears to be a significant difference not only in the motivations for use per segment, but also in the number of social media platforms used. Moreover, this is not only the

case between countries, but also between segments, and between segments for each country (Alarcón-del-Amo et al. 2015, p. 11). More specifically, a positive relation can be observed between the number of social media platforms used and activity, and Entertainment and Socializing motivation and activity. Researches such as those by Campbell et al. and Alarcon-del-Amo thus put forward the question that if different platforms serve different uses, which motivations will predict what type of platform will be used? Furthermore, why does one marketing effort work better on one social media platform than on another? Do people use different social media platforms for different reasons? And if so, what are the implications of this? In this research paper, the influences of the motivations that people have for using social media, and the differences between platforms based on their basic point of form, function, and use will be considered: based on the differences between social media that use primarily text, images, videos, or a mix of these as primary modes of communication between users. The first research question therefore is formulated as follows:

RQ1: “How do motivations for social media use influence the amount of use of various social media platforms?”

With the rise of the Internet and in particular social media, it has become more important for marketers to analyze and connect with their customers. A popular approach to marketing strategy is to segment consumers based on distinctive features. Studies show that in general, a number of comparable consumer segments can be distinguished based on either or both their motivations and attitudes in interacting with social media and social media marketing content (Taylor et al., 2011; Foster et al, 2011; Lorenzo-Romero and Alarcón-del-Amo 2012; Campbell et al., 2014; Shao et al., 2015; Alarcón-del-Amo et al,. 2015).

These segmentations often only look at these groups within one social medium, while alternative research on this subject claims that the significant differences in personality traits

between users of different social media platforms (Hughes et al. 2012), and the use of various social media platforms are being driven by different motivations (Petrocchi et al. 2015).

As such, there is a need to analyze and differentiate between various consumer segments on the basis of their motivations, but also to study how these affect the amount of use of various types of social media platforms. The second research question is consequently formulated as:

RQ2: “What segmentation of social media users can be made based on Socialization, Entertainment, and Information-seeking motivations?”

By answering these two research questions, we will gain insight in what drives people to use social media, and, as unprecedented in earlier academic work, see how this influences the amount of use of different social media platforms. Furthermore, it will shed light on the relation between motivations for social media use, the type of social media platform use, and the receptivity of social media users towards social media marketing efforts.

CHAPTER 2. THEORETICAL FRAMEWORK

First, a more detailed look into the constructs used in the current research will be presented, along with their influence on the amount of use of social media platforms as found in previous research. The implications of earlier research and their subsequent theoretical impact will be used to make the key hypotheses that will be empirically tested in this research paper. These hypotheses will test whether and which motivations drive the use of different types of social media platforms, to answer the first research question.

As such, in addition to the use of these hypotheses as focal point of research earlier typologies and classifications of social media users (some of which based on the constructs used in the current research) are discussed and the stage will be set for the subsequent segmentation that will be performed after the initial hypothesis testing in order to answer the second research question.

2.2. Motivations for social media use

The main constructs used in this research paper —Socialization motivation, Entertainment motivation, and Information-seeking motivation— stem from Uses and Gratifications theory (U&G theory). U&G theory seeks out to understand how and why people use specific media for particular needs that partially derive from their own personality traits. In other words: “The main objective of Uses and Gratifications theory is to explain the psychological needs that shape why people use the media and that motivate them to engage in certain media-use behaviors for gratifications that fulfill those intrinsic needs” (Ko et al., 2005, p. 58). U&G theory has been applied to Internet and social media use since the rise of the Internet in the late 90s–early 2000s, and the rise of social media during the late 2000s–early 2010s throughout the years. For this

reason U&G theory is a relevant and actual theoretical framework in an effort to analyze people’s relationship with media and technologies.

Influence on choice of social media platform more than often is researched in the light of personality traits. However, U&G theory is built upon various kinds of personality traits: “Psychological characteristics, social context, and attitudes and perceptions influence people’s motives and behavior (Rubin, 1993; 1994)” (Paparachissi and Rubin, 2000). In example, Seidman (2013) explores various links between personality traits and motivations for the use of social media.

A seminal article exploring this subject is ‘Predictors of Internet Use’ by Paparachissi and Rubin (2000). They found empirical confirmation to identify motivations for use in the specific environment of the Internet using U&G theory, with influences by a robust variety of earlier work on the subject matter. Paparachissi and Rubin identify five different motivations for the use of Internet:

1. Interpersonal utility motivations (inclusion, affection, social interaction), sometimes called Socialization or Social interaction

2. Pass time motivations

3. Information(-seeking) motivations

4. Convenience motivations (related to time control, convenience, economy etc.) 5. Entertainment motivations

Furthermore, Paparachissi and Rubin state that audience activity (i.e., Internet usage) is central to U&G research, and motivations are key components of audience activity. This is also researched in an article by Ko et al., (2005), who indicate that Information-seeking motivation, Convenience motivation, and Social Interaction motivation significantly influence the duration of time using a

Website. Additionally, they show that the duration of time spent on a Website mediates the attitude towards a brand and purchase intentions (Ko et al. 2005, p. 65-66).

A study concerning uncovering the motivations for social media use specifically is that of Park et al. (2009). They examined the motivations of Facebook users and employing factor analysis identified the existence of four primary motivations for using Facebook: Information-seeking, Socializing, Entertainment and Self-status seeking motivations. These four determinants all vary according to the socio-demographic status of the Facebook users.

Sundar and Limperos (2013) criticize the disparity of the use of different motivations for Internet and social media use in existing research. For example, they state that Paparachissi and Mendelson (2011) use ‘Pass time’, ‘Relaxation’ and ‘Entertainment’ motivations – but that these load as one motivation in their analysis: these three might be simplified as Entertainment motivation. In this meta-analysis they posit a framework of articles that have researched motivations of Internet and social media use, and differentiate which construct are used in which study (Sundar and Limperos, 2013, p. 508). Their overview shows that the most used motivations in social media related research are Socialization, Information-seeking, and Entertainment motivation. We agree with the argument by Sundar and Limperos, which for example combines a multitude of determinants into one ‘Entertainment motivation’ variable. This in combination with the findings of Paparachissi and Rubin (2000), resonating in Park et al.’s (2009) research, leads to concluding that these three will be used as the main motivations for social media use in this research.

2.2.1. Socialization motivation

Social media platforms appear to be mainly designed to grow and foster individuals’ social network. Paparachissi and Rubin in their 2000 article already demonstrate the relation between ‘Interpersonal utility motivation’ (Socialization motivation) and its significance in predicting

amount of Internet use. Hughes et al. (2012) research how preference for Facebook and/or Twitter is influenced by personality traits, and found that in contrast to Twitter, preference for Facebook was predicted by higher Sociability, Extraversion, and Neuroticism; which are precursors to aspects of Socialization motivation as shown by Seidman (2013). Research by Chen (2015) examined the relation between motivations for social media use and the frequency of social media use for female bloggers. She found that ‘Recreation’ motivation (Entertainment motivation) outweighed Information-seeking and ‘Engagement’ motivation (Socialization motivation) in predicting the frequency of social media use in general. However, when differences between Facebook, Twitter, and a cluster of other social media platforms were considered, the results indicated that Facebook use was highly correlated with ‘Engagement’ motivation (Socialization) and Twitter use with Information-seeking motivation. With Mixed (meaning mixing text, image, and video) platforms’ (like Facebook or MySpace) focusing on building a personal profile and sharing experiences with others through messages and pictures, the use of this type of platforms may be predicted by Socialization motivation. The same can be observed for Image-based social platforms (like Instagram); building a visual identity online plays a key role here, and therefore may serve a similar purpose as Mixed social media as Facebook. Thus, the author hypothesizes that:

H1a: Socialization motivation positively influences Mixed social media use.

2.2.2. Information-seeking motivation

Consumers increasingly use social media platforms to acquire information (Shao et al. 2015). A research into the media consumption of teenagers in the Netherlands revealed that social media are increasingly used as their primary mode of staying updated with the world, overthrowing traditional news sources (Stroom Mediacommunicatie, 2015). Text-based platforms such as web forums and Twitter may focus more on information sharing and discussing ideas than other platforms. As mentioned above, the research by Chen (2015) pointed at the correlation of Twitter use with Information-seeking motivation, contrasted with Facebook use. An article by Santana and Hopp (2016) researched the different uses of social media in a sample of journalists. Their results show that journalists place more value on Twitter than on Facebook, for various reasons concerned with the features of the platforms and their use in sharing detailed and relevant information (for journalists). Twitter was found to be better for journalistic research and sharing of information, whereas Facebook was considered to be a more of a socializing or networking platform. In the research by Hughes et al. (2012), the personality trait Need for Cognition (meaning to be inclined towards cognitively stimulating information and activity) predicted a higher preference for Twitter.

Besides Text-based social media, Information-seeking motivation may also predict the use of Video-based social media: “How-to” video searches on YouTube increased with approximately 70% in 2014-2015, with more than 100 million hours of how-to content watched in 2015 in North-America alone. Additionally, “91% of smartphone users watch YouTube how-to videos when completing and executing tasks” (Google Think, 20151_{). Luchman, Bergstrom,} and Krulikowski (2014) found that social media that “associate with content, skills,

information/knowledge building - but not engaging in social contact (...) were YouTube (, MMORPGs, and Wikipedia)” (Luchman, Bergstrom, and Krulikowski, 2014, p. 138).

Based on these works, the author hypothesizes that individuals who score high on Information-seeking motivation will display a higher amount of use of Text-based platforms, and Video-based platforms:

H2a: Information-seeking motivation positively influences Text-based social media use. H2b: Information-seeking motivation positively influences Video-based social media use.

2.2.3. Entertainment motivation

Social media platforms also serve as a way to pass time and entertain oneself, to “seek experiential, hedonic, and entertainment value” (Shao et al. 2015, p. 1074). Another finding by Hughes et al. was that, surprisingly, informational use of Facebook was negatively correlated with Conscientiousness and Need for Cognition (NFC). Hughes et al. explain this by the informational use of Facebook being a better fit for Entertainment motivations: “The negative correlation with both Conscientiousness and NFC may suggest that in contrast to socializing, informational uses of Facebook may well be indicative of procrastination, a lack of self-discipline and diligence” (Hughes et al., 2012, p. 567). Bearing this in mind, the author of this thesis hypothesizes that segments that score high on Entertainment motivation will display a higher amount of use of Mixed social platforms:

2.3. Conceptual framework

The hypotheses based on earlier research on the relation between motivations for social media use and social media platform are represented visually in a conceptual framework below:

Figure 1: Conceptual framework

2.4. Segmentations of social media users in earlier research

A number of segmentations on the bases of either or both the motivations and attitudes/behaviors towards social media use have been performed, providing an overview of different combinations of these motivations and attitudes/behaviors and the ratios in which they present themselves within a social media platform. As Foster et al. (2011) state: “Market segmentation is foundational to developing effective marketing strategies because it is based on the belief that, in order to meet customer needs effectively, all groups of users cannot be viewed as having identical characteristics. Much of the previous research on online communities treats users as a homogenous group, or as varying from one another on a single dimension. By gaining insight into the scope and characteristics of segments within the online community, the authors can assist

Socialization motivation Information-seeking motivation Entertainment motivation Mixed SM Image-based SM Text-based SM Video-based SM H1 + H1b + H2a + H2b + H3 + H1a: +

marketers in developing effective brand management strategies for reaching and influencing specific groups of customers using online technology.” Forrester Research has since as early as 2007 performed segmentation efforts to better understand distinctions between social media users on the basis of their online activities, and found that for example Spectators (low creating activity but high usage) are more Entertainment motivated.

Foster et al. (2011) use interactive participation (Socialization) and information needs (Information-seeking motivation) to identify four distinct segments:

1) Socializers (high interactive participation, low information needs)

2) Minimally involved (low interactive participation, low information needs) 3) SMT Mavens (high interactive participation, high information needs) 4) Info Seekers (low interactive participation, high information needs)

Furthermore, they find empirical support for their hypothesis that “social media users are more appropriately viewed as distinct segments on the basis of their social media usage and participation behaviors” (Foster et al., 2011, p. 14). They also encourage more segmentation research in the area of social media.

Lorenzo-Romero and Alarcón-del-Amo (2012) segment users of social media based on the use of social media and level of active engagement with marketing-related activities (such as posting product reviews). They identify three segments: Introvert users (low activity), Versatile users (mixed activity), and Expert communicator users (high activity). The ratios found among the usage of the social media in aforementioned segmentations indicate that Information-seeking and Socializing are among the main motivations for the use of social media.

A key article to the further expansion of this kind of research, and arguably the most extensive and detailed consumer segmentations as of yet, is the recent research by Campbell et al. (2014), because, in contrast to other consumer segmentations, they use more than one segmentation base. They segment consumers on basis of their reactions to social media marketing, but with Information-seeking, Entertainment, and Convenience motivations as identified by Paparachissi and Rubin in 2000 (and demographics) as covariates. Based on these, they can identify five different segments:

1) Passives, who score low on all the attitudinal variables regarding social media marketing, relatively high Entertainment motivation, and low Convenience motivation.

2) Talkers, who score high on Brand Engagement and WOM referral intention, low on purchase intention, and high Information-seeking motivation.

3) Hesitants, who score low on the attitudinal variables regarding social media marketing, and low Information-seeking motivation.

4) Actives, who score the highest on all the attitudinal variables regarding social media marketing, and high Information-seeking motivation.

5) Averse, who score lowest on the attitudinal variables regarding social media marketing, and above average Convenience motivation.

The study by Campbell et al. provides a detailed look into combinations between motivations and the attitudes and behaviors that are driven by them, with Information-seeking and Entertainment motivation having the highest influence on attitudes towards social media marketing, and Convenience motivation having only a marginal effect on the attitudes. With regards to social media users’ receptivity to social media marketing Campbell et al. state: “Specifically, Information motivation is a strong determinant of segment membership for those

consumers most likely to be impacted by social network marketing (Actives and Talkers). This suggests that positive effects from social network marketing interactions are linked to a consumer’s desire for information” (Campbell et al., 2014, p. 445). Therefore, a higher Information-seeking motivation can be expected to result in higher Brand Engagement. However, it must be noted that the research by Campbell et al. in fact did not measure Socialization motivation, which in the article by e.g. Foster et al. is shown to have a significant impact on various behavioral outcomes.

Shao et al. (2015) segmented Facebook users into different groups based on Socialization motivation, Entertainment motivation, and Information-seeking motivation in combination with various demographic variables, and found results similar to Foster et al. (2011). Four segments were identified, namely Devotees, Agnostics, Socializers and Finders. Devotees displayed high scores on all three motivations. Agnostics scored low on the three motivations. Socializers scored high on Socialization motivation and Entertainment motivation, and low on Information-seeking motivation. Finders scored low on Socialization motivation, and high on Entertainment motivation and Information-seeking motivation.

Because the same constructs used by (in part) Foster et al. (2011) and Shao et al. (2015) are used in this research paper to base the segmentation on (Socialization motivation, Entertainment motivation, and Information-seeking motivation), comparable results in the composition of social media user segments are to be expected. As both these researches found statistical evidence pointing in the direction of a number of four distinct segments based on the constructs used, this is also the number (K) of means (and therefore the number of resulting segments) that will be used in the K-means cluster analysis. This will make interpretation of the segments meaningful and comprehensible, while also easing the comparison of this research paper’s findings with the aforementioned researches by Foster et al. (2011) and Shao et al.

(2015). This research paper however extends the segmentation on basis of social media users’ motivations by also bridging the relation between these motivations with the amount of use of various social media platforms. This is done by researching the impact of segment membership and corresponding motivational differences on the use of not only social media in general, but also between different social media platform types. Furthermore, it is researched how this affects the segment members’ receptivity to social media marketing and brand activity in the form of Brand Engagement.

CHAPTER 3. DATA AND METHOD 3.1. Method

To research the influence of the motivations for social media use on the amount of use of different platforms in a quantitative way, a cross-sectional survey was performed, of which the data was downloaded and analyzed using SPSS.

3.2. Sample

The population for this research was (Dutch) social media users (defined as people having at least one account on a social media platform). This is a large and diverse population, thus heterogeneous sampling was of importance in order to represent the population as good as possible. A convenience sample was used to acquire as many respondents as possible: this to strengthen the quality of the statistical analysis. The measures in the research were not time or date-restricted, so time and date of response should not have influenced the interpretation of the questions. The survey was distributed online using the Qualtrics survey platform, and promoted through the author’s network. A gift voucher was handed out randomly to increase incentive to fill in the survey.

The survey consisted of three parts: one to measure the motivations to use social media, one to measure the use of different types of social media platforms, and one to measure (online) Brand Engagement. After pre-testing, some of the subjects stated that they would find it to be more natural and easy for respondents to first fill in their use of different types of social media, and then their motivations; so, these two parts were switched in the final survey design. The language used in the survey was Dutch, and measures were parallel translated.

3.3. Measures

3.3.2. Motivations for social media use

The constructs Socialization motivation, Entertainment motivation, and Information-seeking motivation were measured through 11 items on five-point Likert-scales (Socialization motivation 5 items, Entertainment motivation 3 items, Information-seeking motivation 3 items) adopted from research by Shao et al. (2015) (which were based on Park et al. (2009)), ranging from 1=Strongly disagree to 5=Strongly agree. The items were parallel translated to Dutch by the author and two others (both proficient in academic English). The translations were put side by side and discussed to decide the final version of the translation of the items. An overview of these items can be found in appendix A and B.

3.3.2. Amount of use of social media type

The constructs Amount of use of predominantly Text-based, Image-based, Video-based and Mixed social media platforms respectively (providing example platforms with each construct), were measured on newly developed five-point Likert scales (to match the items for Motivations) ranging from 1=never, 2=less than 1 hour per week, 3=1 to 4 hours per week, 4=5 to 9 hours per week, 5=10 or more hours per week (e.g. “How often do you use a predominantly text-based social media platform (such as Twitter or web forums) in an average week?”). The scale used for these items was adopted from other social media research by Ip and Wagner (2008) and Foster et al. (2011) to ensure a good fit with actual amount of social media use.

From the pretest, it became clear that in the questions on amount of social media platform use, respondents might interpret the example platform given being the topic of the question. Therefore, two examples of the type of platform were provided to the participant in order to underline the intended nature of the question, versus a specific platform. A mean of these four

variables was later calculated to transform into a newly created variable Average amount of social media platform use.

3.3.3. Brand Engagement

As an addition to the measures used in the segmentation, Brand Engagement was also measured as covariate, to examine relations between social media motivations and reactions to social media marketing efforts (as in Campbell et al., 2014). In order to measure Brand Engagement, 7 items on a 5 pt. Likert-scale ranging from 1=Strongly disagree to 5=Strongly agree adopted from Keller (2001) (also used by Campbell et al., 2014) were included in the survey. The relationship between motivations for social media use and Brand Engagement is particularly interesting for future research, since Campbell et al. (2014) hint at this relation, but has never been researched in combination with amount of use of different social media platforms. The implications of this construct in relation to the others will also be discussed in the results.

3.3.1. Demographics

To research the influence of gender and age, and to profile segments accordingly, the questionnaire asked respondents their demographics gender (binary categorical variable) and age (ratio variable) at the end of the survey.

CHAPTER 4. RESULTS 4.1. Descriptive statistics

A total of 245 respondents filled in the survey, of which 220 (90%) fully completed it. These

n=220 cases were downloaded into an SPSS file and used for the statistical analysis. As regards

to demographic descriptive, 63.6% of respondents were female (140 cases) and 36.4% were male (80 cases). The median age was 24 (Mage=26.08, SDage=8.04 years) with the youngest respondent being 16 and the oldest respondent being 57 years old. The distribution of age was positively skewed: a large amount of the cases ranging between 20 and 30 years of age. This can probably be attested to the fact that the author’s network that was primarily addressed for responses falls into this age category. Nonetheless, social media are used more by persons below age 30 in general (Pew Research Social Networking Factsheet 2014).

The distribution of motivations was as follows: people displayed on average highest Entertainment motivation (Mentertainment=4.17, SDentertainment=.76), then Information-seeking motivation (Minfoseeking=3.22, SDinfoseeking=1.17), and lastly Socialization motivation (Msocialization=2.96, SDsocialization=.76).

Amount of use of social media platforms differed between types of platforms: in general, Mixed social media platforms were used most with a median of 4=‘5 to 9 hours per week’ (MMixed=3.81, SDMixed=0.89), then Video-based social media platforms with a median of 3=‘1 to 4 hours per week’ (Mvideo=2.93, SDvideo=1.08), then Image-based social media platforms with a median of 3=‘1 to 4 hours per week’ (Mimage=2.59, SDimage=1.23), and the least used type of social media platforms were Text-based social media platforms with a median of 2= ‘less than 1 hour per week’ (MText=2.23, SDText=1.08).

Table 1: Means, Standard Deviations, Correlations Variables Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1. Gender (1=female) .64 .48 - 2. Age 26.01 8.04 -.05 - 3. Socialization Motivation 2.96 .76 .11 .04 - 4. Entertainment Motivation 4.17 .76 .14* _-.14* _.24** _- 5. Information-seeking Motivation 3.22 1.17 .04 -.07 .17 * _.28** _-

6. Text-based social media 2.23 1.08 -.14 .03 .10 .13 .17* - 7. Image-based social media 2.59 1.23 .31** _{-.07 .26}** _{.10 .30}** _.19** _-

8. Video-based social media 2.93 1.08 -.24 -.12 .07 .10 .01 .08 -.00 - 9. Mixed social media 3.81 .89 .08 -.15* _{.13 .11 -.05 .07 .16}* _.22** _-

10. Brand Engagement 2.48 .95 .01 -.04 .28** .24** .58** .14* .38** .02 .06 - *. Correlation is significant at the 0.05 level (2-tailed).

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

4.2. Hypothesis Testing

To investigate the relations between the motivations for social media use and Average amount of social media use per platform (mean of the variables of different types), and the amount of use of specific types of social media platforms, six hierarchical multiple regressions were performed. In the first step, the covariates Gender and Age were entered, after which the motivations for social media use were added to the model. The six regressions looked into the effects of the motivations for social media use on the amount of social media in general, and the use of each different type of social media platform, and their effects on Brand Engagement, respectively, after controlling for Gender and Age.

4.2.1. Effects on Average amount of social media platform use

First, the effects on the Average amount of social media platform use were examined. Taking the mean of the various types of social media platforms use, this was taken as a new dependent variable. In the first step, Gender and Age were entered, resulting in model 1, not significant F (2, 217) = 1.1887; p=.154, explaining 1.7% of total variance. After adding the motivations for social

media use, the model 2 was significant, with F (5, 214) = 5.256; p<.001, explaining 10.9% of the total variance. This explained an additional 9.2% of total variance after controlling for Gender and Age (R2 change =.092, F change (3, 214) = 7.39, p<.001). Socialization motivation was the best predictor of the Average social media platform use (β=.204, p=.003), followed by Information-seeking motivation (β=.143, p=.036). No other predictors were statistically significant.

Table 2: Hierarchical regression model of Average amount of social media platform use

R R2 _R2 _{Change B Std. Error β t} Step 1 .13 .02 Gender .01 .09 .01 .11 Age -.01 .01 -.13 -1.93 Step 2 .33 .11*** .09*** Gender -.04 .09 -.03 -.48 Age -.01 .01 -.12 -1.82 Socialization Motivation .17 .06 .20** 3.03 Entertainment Motivation .07 .06 .08 1.20 Information-seeking Motivation .08 .04 .14* 2.11

Dependent variable: Average amount of social media platform use Note: *Statistical significance: *p <.05; **p <.01; ***p <.001

4.2.2. Effects on amount of use of Mixed social media

After this, the effects of the three motivations for social media use on the use of Mixed social media were tested. In the first step, Gender and Age were entered, this model was not significant F (2, 217) = 2.854; p=.06, explaining 2.6% of total variance. Thus, with these statistics the model did not predict the amount of use of Mixed social media well. When the three motivations were entered, the model was significant F (5, 214) = 2.478; p=.033, with 5.5% of total variance explained by the model. Looking at the influence of the predictors, none of the motivations were significant predictors of the amount of use of Mixed social media. Only Age had a significant (negative) influence (β=-.143; p=.035) on the amount of use of Mixed social media.

Table 3: Hierarchical regression model of amount of use Mixed social media

R R2 _R2 _{Change B Std. Error β t} Step 1 .16 .03 Gender .13 .12 .07 1.02 Age -.02 .01 -.14 -2.11 Step 2 .23 .06* .03 Gender .09 .12 .05 .69 Age -.02 .01 -.14* -2.12 Socialization Motivation .14 .08 .12 1.77 Entertainment Motivation .10 .08 .09 1.22 Information-seeking Motivation -.08 .05 -.11 -1.51

Dependent variable: Amount of use of Mixed social media. Note: *Statistical significance: *p <.05; **p <.01; ***p <.001

4.2.3. Effects on amount of use of Image-based social media

Next, the effects of the motivations on the amount of use of Image-based social media were tested. In the first step, Gender and Age were entered, with model 1 significant F (2, 217) = 11.826; p<.001 and explaining 9.8% of total variance. In the second step, the three motivations were entered. Model 2 was significant F (5, 214) = 11.79; p<.001, explaining 21.6% of total variance of amount of use of Image-based social media. The added variables explained an additional 11.8% more variance of the amount of use of Image-based social media, after controlling for Gender and Age (R2 change =.118, F change (3, 214) = 10.708, p<.001). Three out of five predictors were statistically significant: Gender (β=.284, p<.001) had the strongest effect, then Information-seeking motivation (β=.269, p<.001), and last Socialization motivation (β=.201, p=.002).

Table 4: Hierarchical regression model of amount of use of Image-based social media

R R2 _R2 _{Change B Std. Error β t} Step 1 .31 .10*** Gender .78 .16 .31 4.73 Age -.01 .01 -.06 -.89 Step 2 .47 .22*** .12*** Gender .72 .16 .28*** 4.63 Age -.01 .01 -.06 -.95 Socialization Motivation .33 .10 .20** 3.17 Entertainment Motivation -.12 .11 -.08 -1.14 Information-seeking Motivation .28 .07 .27*** 4.24

Dependent variable: Amount of use of Image-based social media. Note: *Statistical significance: *p <.05; **p <.01; ***p <.001

4.2.4. Effects on amount of use of Text-based social media

Following, the effects of the motivations on the amount of use of Text-based social media were investigated. Again, in step 1 Gender and Age were entered. Model 1 was not significant F (2, 217) = 2.175; p=.116, explaining 2% of total variance. Entering the three motivations for social media use, model 2 was significant F (5, 214) = 3.013; p=.012), explaining 6.6% of total variance of the amount of use of Text-based social media. The added variables explained 4.6% more variance, after controlling for Gender and Age (R2 change =.046, F change (3, 214) = 3.520, p=.016). Two out the five predictors were statistically significant predictors of the amount of use of Text-based social media: Gender (β=-.163; p=.016) and Information-seeking motivation (β=.139; p=.046).

Table 5: Hierarchical regression model of amount of use of Text-based social media

R R2 R2 Change B Std. Error β t Step 1 .14 .02 Gender -.31 .15 -.138 -2.05 Age .00 .01 .019 .28 Step 2 .26 .07** .05** Gender -.37 .15 -.16** -2.44 Age .01 .01 .04 .56 Socialization Motivation .09 .10 .07 .94 Entertainment Motivation .14 .10 .10 1.39 Information-seeking Motivation .13 .06 .14* 2.01

Dependent variable: Amount of use of Text-based social media. Note: *Statistical significance: *p <.05; **p <.01; ***p <.001

4.2.5. Effects on amount of use of Video-based social media

Ensuing, the effects on the amount of use of Video-based social media were tested. In model 1, Gender and Age were entered, with the model being significant F (2, 217) = 9.005; p<.001, explaining 7.7% of the total variance.

Model 2 was significant F (5, 214) = 4.542, p=.001, explaining 9.6% of the total variance in the model. Adding the three motivations resulted in explaining 1.9% more of total variance after controlling for Gender and Age (but this was not a significant F change): R2 change =.019, F change (3, 214) = 1.523, p=.209. Out of the predictors, Gender was statistically significant (β=-.27; p<.001).

Table 6: Hierarchical regression model of amount of use of Video-based social media

R R2 _R2 _{Change B Std. Error β t} Step 1 .28 .08*** Gender -.56 .15 -.25 -3.79 Age -.02 .01 -.14 -2.10 Step 2 .31 .10** .02 Gender -.61 .15 -.27*** -4.09 Age -.02 .01 -.13 -1.96 Socialization Motivation .11 .10 .08 1.13 Entertainment Motivation .15 .10 .11 1.53 Information-seeking Motivation -.03 .06 -.03 -.41

Dependent variable: Amount of use of Video-based social media. Note: *Statistical significance: *p <.05; **p <.01; ***p <.001

4.2.6. Effects on Brand Engagement

Concluding, the effects of the motivations on Brand Engagement were analyzed, again with a hierarchical regression. In model 1, Gender and Age were entered, with model 1 being not significant F (2, 217) = .151; p=.86. This model explained .1% of total variance.

Model 2, with the three motivations Socialization, Entertainment, and Information-seeking motivation entered, was significant F (5, 214) = 25.109; p<.001, explaining 37% of the total variance. The additional variance explained by these three added predictors was 36.8%, a big step from model 1 (R2 change =.368, F change (3, 214) = 41.691, p<.001). In this regression analysis, the statistically significant predictors of Brand Engagement were Information-seeking motivation with a strong effect (β=.533; p<.001), and to a lesser degree Socialization motivation (β=.176, p=.002).

Table 7: Hierarchical regression model of Brand Engagement R R2 _R2 _{Change B Std. Error β t} Step 1 .04 .00 Gender .02 .13 .01 .15 Age -.00 .01 -.04 -.52 Step 2 .61 .37*** .37*** Gender -.07 .11 -.04 -.65 Age -7.89 .01 -.00 -.012 Socialization Motivation .22 .07 .18** 3.11 Entertainment Motivation .07 .07 .06 .98 Information-seeking Motivation .43 .05 .53*** 9.37

Dependent variable: Brand Engagement.

4.3. Cluster Analysis and Segmentation

In order to answer research question two (“What segmentation of social media users can be made based on Socialization, Entertainment, and Information-seeking motivations?”) a K-means cluster analysis was performed. The choice was made to perform a K-means cluster analysis with a number of four clusters. This because it made the interpretation of results more meaningful, and because this is a number of clusters identified in other studies (Foster et al. 2011; Shao et al. 2015; Campbell et al. 2015), which used similar constructs as the bases for their segmentations.

4.3.1. K-means Cluster Analysis

The bases of segmentation were the three independent variables Socialization motivation, Entertainment motivation, and Information-seeking motivation. Since these three constructs were all measured using five-point Likert scales, it was not necessary to standardize the scores for these variables. The maximum number of iterations was 10, after which no adjustments to cluster centers should be necessary. Cluster membership was saved as a new categorical variable (=1, 2, 3 or 4) in order to compare segments and to perform analysis with other variables after the cluster analysis. The Iteration History showed that after seven iterations, convergence was achieved due to no (or small) change in cluster centers, meaning that less than 10 iterations were needed to arrive at the desired number of clusters. The final cluster centers are reported in Table 8 below. Table 8: Final cluster centers of K-means cluster analysis

Cluster 1 2 3 4

Socialization Motivation _{2.23 3.29 2.45 3.06}

Entertainment Motivation _{2.17 4.50 3.89 4.31}

Information-seeking

The importance of each variable in determining the cluster was represented in the ANOVA. All three motivations had a significant impact on determining to which cluster a case was allocated. Information-seeking motivation was the most impactful (F=330.81, p<.001), Entertainment motivation second (F=74.77, p<.001), and Socialization motivation third (F=26.03, p<.001) in determining cluster membership. ANOVA revealed that the differences in means of these motivations differed significantly between the four segments: Socialization motivation F (3, 216) = 26.029, p<.001, Entertainment motivation F (3, 216) = 74.765, p<.001, Information-seeking motivation F (3, 216) = 330.809, p<.001.

In order to give a more interpretative and meaningful view of the cluster or segment profiles, the mean of each variable between the four clusters/segments was calculated, resulting in Socialization motivation (2.76), Entertainment motivation (3.72) and Information-seeking motivation (2.64). For every cluster then, it was determined whether the cluster’s score for each variable was under or above this average. When under, the cluster received a (-) for this variable, when above, the cluster received a (+) for this variable.

4.3.2. Description of resulting segments

The outcome of this is represented in Table 9 below. The resulting characteristics of the segmentation replicate findings by Foster et al. (2011) and Shao et al. (2015) closely: a segment scoring low on all motivations, a segment scoring high on all motivations, a segment with high Information-seeking motivation and low Socialization motivation, and a segment with low Information-seeking motivation and high Socialization motivation. The segments found in this research paper are therefore named after the similar segments found in the study by Shao et al.: Agnostics, Devotees, Finders, and Socializers.

Table 9: Social media user segments and number of cases in each segment

Social media user segments 1) Agnostics 2) Devotees 3) Finders 4) Socializers

Socialization Motivation

_-

₊

_-

₊

Entertainment Motivation

_-

₊

₊

₊

Information-seeking Motivation

_-

₊

₊

-n of cases _{12 102 59 47}

(n=220)

● Agnostics The first segment was the smallest segment (n=12) making up for 5.4% of total respondents. This segment displays below average scores on all three motivations: Socialization motivation (M=2.23), Entertainment motivation (M=2.17) and Information-seeking motivation (M=1.58) compared to the other segments, and thus was named the Social Media Agnostics.

● Devotees The second segment was the largest segment (n=102) making up for 46.4% of total respondents. This segment displays above average scores on all three motivations: Socialization motivation (M=3.29), Entertainment motivation (M=4.50) and

Information-seeking motivation (M=4.12) compared to the other segments, and thus was named the Social Media Devotees.

● Finders The third segment made up about 26.8% of total respondents (n=59). This segment scores below average on Socialization motivation (M=2.45), but above average on Entertainment motivation (M=3.89) and Information-seeking motivation (M=3.32), compared to the other segments. Therefore, this segment was named the Social Media Finders.

● Socializers The fourth and final segment made up about 21.4% of total respondents (n=47). This segment scores above average on Socialization motivation (M=3.06) and Entertainment motivation (M=4.31), but below average on Information-seeking motivation (M=1.54), compared to the other segments. Thus, this segment was named the Social Media Socializers.

After this, analyses of variance (ANOVA) (and a χ2 cross-tabulation for Segment * Gender) were performed to analyze how segment membership related to various outcome variables.

Demographics

The χ2 analysis indicated there were no significant differences between segments regarding Gender: χ2 (3) = 3.265, p=.353. ANOVA showed that there neither were significant differences between segments regarding Age F (3, 216) =1.011; p=.389. Thus, there can be no assumptions made based on Gender or Age relating to the population. In the sample of this research, means for each segment indicated that Social Media Agnostics were on average a bit older (MageAgnostics=29.75) than the other three segments, which were on average around the total mean age of 26. Furthermore, on average, Social Media Agnostics were mostly male (40% female), Social Media Devotees mostly female (66% female), Social Media Finders mostly female (61% female) and Social Media Socializers mostly female as well (68% females).

However, it must be noted again, that the differences between the segments are not significant, and thus must be viewed with this in mind.

Amounts of use of (different types of) social media

Through ANOVA the differences in amounts of use of (different types of) social media between the four segments were analyzed. ANOVA shows that there are significant differences in the Average amount of social media platform use between the four segments F (3, 216) =4.360; p=.005. Overall, Social Media Devotees displayed the highest Average amount of social media platform use, followed by the Social Media Finders, then the Social Media Socializers. The segment that displayed the lowest Average amount of social media platform use was that of the Social Media Agnostics (Figure 2).

Figure 2: Means of average amount of social media platform use per segment

* 1=never, 2=less than 1 hour per week, 3=1 to 4 hours per week, 4=5 to 9 hours per week, 5=10 or more hours per week ANOVA was then carried out to test the differences between segments in the amount of use of different social media platforms. An overview of the means per segment, per social media type can be found in Figure 3

2.6 3.04 2.81 _2.73 1 1.5 2 2.5 3 3.5 4 4.5 5

Agnostics Devotees Finders Socializers

A mou n t of ave rage s oc ial me d ia p latfor m u se *

Figure 3: Mean amount of social media use per platform per segment

* 1=never, 2=less than 1 hour per week, 3=1 to 4 hours per week, 4=5 to 9 hours per week, 5=10 or more hours per week The ANOVAs that measured the differences between the segment’s scores for each type of social media platform pointed out that apart from the use of Image-based social media F (3, 216) = 7.295; p<.001, there were no significant differences between the means per segment. Thus, these differences cannot be extended to the population and thus no definitive conclusions can be made from this. However, for illustrational purposes, the descriptive statistics from this sample will be discussed.

First of all, we can see that in the amount of use of Mixed social media (Facebook, MySpace etc.), the Agnostics have, together with the Socializers, the highest mean amount of use. The Finders on the other hand have the lowest mean score for the amount of using Mixed social media. In the amount of use of Image-based social media (where there were significant differences between segments) we can observe that the Devotees have the highest mean score, followed by the Finders. The Agnostics and Socializers have lower scores on the use of this type of social media platform. The amount of use of Text-based social media shows a similar pattern;

3.92 _3.82 3.69 3.91 2.08 2.93 2.54 2.02 1.83 2.38 2.22 2 2.58 3.03 2.78 3 1 1.5 2 2.5 3 3.5 4 4.5 5

Agnostics Devotees Finders Socializers

A mou n t of u se of s oc ial me d ia p latfor m* Mixed Image-based Text-based Video-based

with the Devotees having the highest mean score, followed by the Finders, the Socializers, and the Agnostics. Finally, the amount of use of Video-based social media shows that the Devotees and the Socializers show equal levels of use, with Finders falling second, and the Agnostics displaying the lowest mean score. Again, it must be stated, that the differences between segment means were not significant (except for Image-based social media), and that therefore these findings cannot be generalized to the population of social media users in general.

Brand Engagement

Concluding, ANOVA was carried out to see if there were differences in Brand Engagement between the four segments. This resulted in the finding that there were in fact significant differences in Brand Engagement between the four segments F (3, 216) = 35.051; p<.001. A graphic overview can be found in Figure 4 below. The segment that displayed the highest amount of Brand Engagement was that of the Devotees (MBrandEng=3.01), followed by the Finders (MBrandEng=2.36), the Socializers (MBrandEng=1.73) and lastly the Agnostics (MBrandEng=1.61).

Figure 4: Mean Brand Engagement per segment

All of the above analyses are compiled in Table 10, encapsulating a more detailed and comprehensible description of segment characteristics. Similarly to the table depicting distributions of motivations per segment (Table 9) the average of each segment for the use of different types of social media and Brand Engagement was compared with the total mean of the segments, and, if below the average of the four segments, appointed a (-), and if above appointed a (+). 1.61 3.01 2.36 1.73 1 1.5 2 2.5 3 3.5 4 4.5 5

Agnostics Devotees Finders Socializers

Br an d En gage me n t

Table 10: Summary of segment characteristics

Social media user segments 1) Agnostics 2) Devotees 3) Finders 4) Socializers

Demographics

Percentage of total sample _{5.4% 46.4% 26.8% 21.4%}

Gender _{60% ♂ 66% ♀ 61% ♀ 68% ♀}

Mean Age _{29.8 25.8 25.5 26.5}

Motivations for social media use

Socialization Motivation***

_-

₊

_-

₊

Entertainment Motivation***

_-

₊

₊

₊

Information-seeking Motivation***

_-

₊

₊

_-

Amount of Social Media use

Average social media platform

use**

-

+

+

-

₊

_-

_-

₊

_-

₊

₊

_-

_-

₊

₊

_-

_-

₊

_-

₊

_-

₊

₊

_-

CHAPTER 5. DISCUSSION

The results of this research paper and their theoretical and practical implications will be discussed here. Regarding the various motivations for social media use and their influence on the amount of social media platform use, the following can be set forth. Socialization motivation followed by Information-seeking motivation significantly (positively) impacted the Average amount of use of social media per platform the most (section 4.2.1), even though Entertainment motivation shows a higher mean score. This shows that after all, social media are indeed for social use first. The need to connect with others and staying in touch with friends and acquaintances is the main motivation for the use of social media. This is in concurrence with results found by Ko et al. (2005), who demonstrate that Information-seeking motivation and Social Interaction motivation (but not Entertainment motivation) significantly influence the duration spent on a website, and Ryan and Xenos (2011) who find that for example, Extraversion results in higher Facebook use. In contrast our results are not in accordance with findings by Pettrochi et al. (2015), whose results report that Entertainment motivation was higher with overall social media use. Chen (2015) also finds that ‘Recreation’ motivation (similar to Entertainment motivation) outweighs ‘Engagement’ motivation (Socialization) and Information (-seeking) motivation in predicting the frequency of social media use for female bloggers. In this research paper we found no empirical evidence for this. Entertainment motivation in fact did not account for any statistically significant effects on the amount of use of social media in general nor for specific types of social media. This raises questions about this discrepancy between the findings in the current study and aforementioned research.

The results from section 4.2.2 indicate there is a significant negative relation between Age and the amount of use of Mixed social media; meaning younger respondents display more use of

Mixed social media. For this type of social media, none of the motivations were significantly influencing, and thus, H1 (‘Socialization motivation positively influences Mixed social media use’) and H5 (‘Entertainment motivation positively influences Mixed social media use’) are not supported. In general, younger people use social media more than older people (Pew Research sheet 2014), and seeing that Mixed social media had the highest average amount of usage, this general effect could be explained by this. Younger respondents show a significantly higher amount of use of Mixed social media such as Facebook and MySpace, with these platforms forming a key role in staying connected to their social network.

Based on the results of 4.2.3 we have demonstrated that H1b (‘Socialization motivation positively influences the amount of use of Image-based social media’) is supported, but Gender (female) and Information-seeking motivation are better predictors of the amount of use of Image-based social media. Gender is thus the best predictor for this type of social media. Pew Research Center’s annual online socio-demographics reports that of active web users, 44% of American females used Pinterest, versus 16% of males. 31% of American females used Instagram, versus 24% of males2. Park et al. (2009) show findings that reveal a (though very weak) significant correlation between females and Information-seeking motivation. Shao et al. (2015) also found that “females were significantly more highly motivated by (...) information needs than males” while they researched the effects of motivations on Facebook use (Shao et al., 2015, p. 1078). The results of 4.2.3 also show that users of Image-based platforms were Information-seeking and Socialization motivated. Especially on Instagram, many brands are present where they share information about their products and services, thus being a rich source of information for people.

From the results of 4.2.4, we can infer that H2a (‘Information-seeking motivation positively influences the amount of use of Text-based social media’) is supported, with Gender

(male) being a slightly better predictor for the amount of use of Text-based social media platforms. According to Pew Research Center, males are more represented on online discussion forums (20% of online males, vs. 11% of online females), and Twitter (22% of online males, vs. 15% of online females)3. Furthermore, the hypothesis that a higher amount of Information-seeking motivation significantly influences the amount of use of Text-based social media is accepted, in agreement with studies by Hughes et al. (2012), Chen (2015), and Santana and Hopp (2016). Text-based social media platforms allow users to use text to accurately describe ideas, or receive information in more detail than say, an image could. Furthermore, this type of platform allows for more discussions between users through textual conversation.

The results of 4.2.5 show a significant negative relation between gender and the amount of use of Video-based social media, meaning significantly more males make use of Video-based social media as opposed to females. Apart from this, no motivations had a statistically significant influence. As such, H2b (‘Information-seeking motivation positively influences the amount of use of Video-based social media’) was not supported.

Results from section 4.2.6 indicate that Information-seeking and Socialization motivation are significant predictors of Brand Engagement. These results are in correspondence with the findings by Ko et al. (2005), who found that these two motivations drove audience activity, which was a mediator for Brand Attitude. This is also in line with the findings of Campbell et al. (2014) who showed that in their segmentation, the segments with higher Information (-seeking) motivation also displayed higher Brand Engagement: “Specifically, information motivation is a strong determinant of segment membership for those consumers most likely to be impacted by social network marketing (Actives and Talkers). This suggests that positive effects from social network marketing interactions are linked to a consumer’s desire for information” (Campbell et