2.pdf

  • Uploaded by: Mickey Koen
  • 0
  • 0
  • July 2020
  • PDF

This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA


Overview

Download & View 2.pdf as PDF for free.

More details

  • Words: 18,564
  • Pages: 92
Just bought a new #Gucci handbag Study about eWOM and Impression Management on Instagram

Master of Science in Marketing Faculty of Economics and Business Administration Vrije Universiteit Amsterdam

Student: Lotte Lobé Student number: 2582355 Supervisor: E.F.J. Lancée Academic year: 2016 - 2017 Submission date: July 1st, 2017

1

ABSTRACT

Instagram users are keeping themselves busy with impressing others by showing off purchased products. Customers who talk about products online is one of the most effective forms of marketing (i.e. eWOM). Especially luxury products seem to be suitable for impression management activities, which is an interesting fact for luxury brands. This study adds value by finding out if self-monitoring and impression management are drivers of eWOM about luxury products via Instagram. The innovative perspective of the research is making the distinction between privately and publicly used products. Data was generated using an online questionnaire targeting women with an Instagram account. It was expected that someone’s level of self-monitoring influences impression management behaviour positively, which results in more actively use of eWOM. According to prior literature, this effect should be bigger for products that are meant for public usage. Result did not confirm the mediation effect, but exploratory analyses concluded a significant relationship between impression management and eWOM. As hypothesized, eWOM outcomes turned out higher for products that are used in public. The main take-away for marketers is that impression management motivations drive consumers to talk about a luxury product via Instagram. This works probably better for products that are meant for public usage.

Electronic worth of mouth • Self-monitoring • Impression management • Instagram • Private vs. public product usage

2

INTRODUCTION

Do you ever scroll through your Instagram account and wonder why people share updates concerning an expensive watch, expressing how #happy they are with a purchase? Or are you that type of person who tries to gain status and impress your followers with showing of luxury goods? From a luxury brand’s perspective, (potential) consumers who talk about your products via social media platforms are extremely important for marketing related goals as brand awareness, brand preference, purchase intentions and loyalty (Buttle, 1998; Cheung and Thadani, 2012; Cheung, Lee & Rabjohn, 2008; Gruen, Osmonbekov & Czaplewski, 2006). Luxury car brand MINI describes this insight with the metaphor ‘MINI-lovers maxi mouths’, explaining the impact that satisfied customers have on fellow consumers by making an online brand-related statement (Yeh and Choi, 2011). But what drives customers to talk about a brand’s luxury products via social network sites, and what situational context might triggers this or specifically not? Following trends regarding social media marketing and content strategy, platforms focusing on the concept of sharing visuals are favourable among consumers (Thomas, 2012). Practical examples of such platforms are Instagram, YouTube and Snapchat which are rising in ranking for most popular social media (Dreamgrow, 2017; Lifewire, 2017; Smart Insights, 2017). The founder of Instagram, Kevin Systrom, created the platform in order to give people the opportunity to share their life trough pictures. Nowadays, Instagram grown into a business platform and integrated marketing tool upon which companies and consumers generate brand and product related content (Instagram, 2016). According to Internet Live Stats (2016), there are around 784 pictures placed on Instagram every second. Considering present-day marketing opportunities, the current research is interested in traits and drivers that predict

3

engaging in electronic-worth-of-mouth about luxury products via Instagram. When studying traits related to behaviour on social network sites, one seems to act as a better ‘social chameleon’ than the other, which can be explained by someone’s level of selfmonitoring. In contradiction to low self-monitoring individuals, high self-monitoring social media users are more able to control behaviour in order to meet the requirements of a social situation (Hogan, 2010; Lennox and Wolfe, 1984; Snyder, 1979). Another topic of interest when investigating behaviour on Instagram is the tendency that some people have to make a good impression towards others, which deals with the concept of ‘impression management’ (Goffman , 1978; Rosenfeld, Giacalone & Riodan, 1995). It seems to be really important to generate enough followers and likes to achieve a desired social identity. Remarkably, luxury products play an important role compared to non-luxury goods if it comes to gaining this status (Hudders, 2012; Yang and Mattila, 2017). This study relates the constructs self-monitoring and impression management, testing if someone’s level of self-monitoring has an influence on the way he or she impresses the online audience. Furthermore, this research aims to investigate the effect of impression management on eWOM activities via Instagram. Besides the influence of traits and psychological constructs on eWOM behaviour, several papers aroused the interest of placing this study in a certain context. The distinction will be made between eWOM about products that are used in public situations (outside the house) and products that are used in private situations (at home) (Graeff, 1996; Kulviwat, Gordon and Obaid Al-Shuridah, 2009; Ratner and Kahn, 2002). Testing eWOM behaviour and its potential drivers for both situations. The following research question summarizes the aim of the study:

4

Does someone’s (relatively high) level of self-monitoring leads to (more) use of impression management and therefore to (more) actively engagement in eWOM about luxury products via Instagram, testing this relationship for both privately and publicly used products?

To answer the main question, the following sub-questions are formulated: 

Which drivers of eWOM can be found in previous literature and what are potential new ones?



What is the difference between low and high self-monitoring people and how does this trait influences impression management?



Is there a relationship between someone’s impression management behaviour and eWOM activity?



What is the difference between privately used and publicly used products and how does this influence the potential indirect relationship that is addressed by this research? Answering the research questions brings theoretical and managerial relevance to the

field. Many research has been conducted about traits and drivers stimulating eWOM behaviour (Choi and Kim, 2014; Chu and Kim, 2011; Henning-Thurau Gwinner, Walsh & Gremler, 2004; Kim, Jang & Adler, 2015; Lee, Noh & Kim, 2013). This study views this potential relationship from innovative perspective, inspired by a considerable amount of studies testing the influence of products that are meant for public vs. private usage. This distinction seems to have an effect in different offline situations, but is never tested in the context of eWOM via Instagram before. Although previous research has featured great attention on eWOM via social network sites, the main emphasis is on platform Facebook, which has a different approach than Instagram (Chen et. al, 2013; Choi & Kim, 2014; Zywica & Danowski, 2008). Furthermore,

5

previous studies tend to research eWOM from a broader perspective, focusing for instance on eWOM drivers in general. This study has a more demarcated positioning regarding the platform (Instagram), product type (luxury) and context (public vs. private product usage). The theoretical relevance can be seen as deepening of previous studies about drivers of eWOM, combined with an extension of public vs. private products usage research to the online world. This study aims to bring managerial relevance to the field of luxury brand marketing. As mentioned before in the MINI success story, the effect of eWOM is a worthwhile and influential marketing strategy to consider. Important to be aware of is that the concrete action comes from the consumer and not from the marketing department. Therefore, it is useful to study traits and drivers discussed in this paper that trigger eWOM via Instagram. The second relevant insight provided by the current study is the potential difference between eWOM about publicly vs. privately used products. In the following sections the literature review gives deeper insight into the main constructs from which hypotheses are formulated. A within subjects design is conducted to test the expectations, followed by analyses of data from an online survey. The research concludes with a general discussion in which the study will be perceived from a broader perspective in terms of limitations and avenues for further research regarding eWOM.

6

THEORETICAL FRAMEWORK

Electronic worth of mouth about luxury products

Electronic worth of mouth refers to the online form of word of mouth (i.e. WOM). Marketers have recognized the importance of this phenomenon in which consumers affect other consumers in decision making and purchase behaviour. One of the earliest researches conducted about product adoption and the role of product-related conversation is from Arndt (1967). He defined WOM as product related person-to-person communication between communicator and receiver, without experiencing a commercial intention. Day (1997) inferred that personal sources are viewed as more trustworthy than forms of adverting spread by companies. This statement is supported by a research from Engel, Kegerreis & Blackwell (1996), saying that WOM has a greater influence on (potential) customers than printed ads, sales persons and radio advertisement. According to Buttle (1998), the power of worth of mouth is greater than influencing purchase behaviour only. He found that WOM has an effect on variance of conditions as awareness, expectations, attitude and product or brand perception. Besides the positive consequences of WOM, a considerable amount of research has been performed about negative WOM. Several researchers concluded that the effect of negative WOM is bigger than for positive WOM (Richins, 1983; Bachleda and Berrada-Fathi, 2016). According to Sweeney, Soutar & Mazzarol (2005), the type of WOM and its influence is mediated by emotion. Results indicate that negative WOM is more emotional and shows feelings of dissatisfaction. Because of the intensity of the emotions, the sender is more likely to influence the receiver’s opinion about the brand or product. In contrast, positive WOM is described as more considered and cognitive. Research of Richins (1983) also associates

7

feelings of dissatisfaction with the tendency to generate in negative WOM; the higher the feelings of dissatisfaction, the stronger the tendency to engage in negative WOM. Remarkably, Charlett, Garland & Marr (1995) conducted an exploratory research about this subject and did not found significant evidence in differences for negative vs. positive WOM. When studying the evolution of WOM the article of Kozinets, Valck, Wojnicki & Wilner (2010) brings interesting findings to the light. As explained earlier, the simplest form of WOM is personal brand related consumer to consumer communication, named the organic inter-consumer influence model. From there onwards the evolution started with a transformation to the linear marketer influence model, which was applicable during the postWorld War II era. From a marketers’ perspective, the interest arose for influencing WOM processes and the understanding of it. The role of so called opinion leaders was extremely important in convincing other consumers about brands and products. These opinion leaders were viewed as trustworthy by other consumers and therefore an important target group for marketers. Traditional marketing tools as printed advertisements and promotions were used to stimulate positive WOM between opinion leaders and other consumers. The third, and also most recent, discussed WOM related process by Kozinets et. al (2010) is the network coproduction model. Due to the internet consumers can exchange brand related information via network system. Market messages are no longer one-way oriented, but flow among online networks, groups and consumer communities. Thus, nowadays consumers communicate product or brand related information through an electronic landscape in forms of reviews, ratings, updates and posts, called electronic-worth-of-mouth, further referred to as eWOM. This process is defined as a positive or negative statement about a brand or product made by a (potential) customer (HenningThurau, Gwinner, Walsh & Gremler, 2004). According to Hu and Ha (2015), eWOM should not be limited to internet based statements created by consumers. The researchers interpret

8

statements created by companies, which are shared or reposted by consumers, also as eWOM. Other papers define eWOM activities driven by companies itself rather as electronic worth of mouth marketing (WOMM), since it is stimulated and controlled using the marketers mouth (Stokes and Lomax, 2002; Sernovitz, Godin & Kawasaki, 2006; Williams and Buttle, 2011). Important to mention is that the current study focuses on eWOM that is fully initiated by the consumer itself. Furthermore, Hu and Ha divided eWOM into four classifications. The first category is specialized eWOM, which concerns with customer reviews on rating websites designed for product comparisons (e.g. TripAdvisor). The second form is affiliate eWOM, referring to customer reviews affiliated with retail websites. The name of the third category is social eWOM and includes all product or brand related communications via social network sites (e.g. Facebook, Instagram and Myspace). The last category is about eWOM on discussion forms as blogs, so called miscellaneous eWOM. The current research is specifically interested in the third category which concerns with social eWOM. Social media specific eWOM activities (e.g. via Instagram) are for instance expressed in forms of liking, sharing and posting (Alhabash, McAlister, Lou & Hagerstrom, 2015; Reynolds-McIlnay and Taran, 2010). Social eWOM is according to Daugherty and Hoffman (2014) an emerging online dimension of sharing brand or product related content. Since social media becomes more and more part of the marketing mix, a considerable amount of research has been conducted about drivers and motivations for participating in eWOM activities. The first relevant article is from Muntinga, Moorman & Smit (2011), discussing interview results from consumers who engaged in brand related social media use. They developed the COBRA concept to understand different participations levels and motivations. According to them, the three main categories in social network eWOM are consuming,

9

contributing and creating information. Motivations can be the need for finding or creating information, entertainment and social interaction. Another driver is the feeling of experiencing empowerment when having an influence on companies or other consumers. The fourth motivation is remuneration, which deals with gaining an economical reward (e.g. money) for participating in spreading brand related information. The last driver is personal identity and covers social media gratification that is related to the self. Another research regarding determinants of consumers’ participation in social eWOM is set out by Chu and Kim (2011). Results show that engaging in product focused eWOM is positively associated with trust, tie strength and normative / informational influence between users. The study suggests that especially social relationships between users are predictors of eWOM. Referring to the research from Hennig-Thurau et. al (2004) again, major eWOM drivers arise from the desire for social interaction, economic drive, care for other consumers and strengthen own pride. Moreover, self-enhancement came out as a reason to spread the worth electronically. The article of Lee and Kim (2012) examines self-construal as a psychological driver for engaging in eWOM. In this study the perception of how an individual experiences ‘the self’ in relation to others, predicts eWOM behaviour. Another psychological construct positively related to eWOM is ‘self-presentation’, which refers to achieving a desired social online identity. Underlying motivations are showing others that you are powerful in sharing information and capable of purchasing good products. Furthermore, individuals try to connect a brand image to their personal image in order to present themselves in a positive way. So, eWOM engagers are gaining recognition of personal values from others (Choi and Kim, 2014; Kim, Jang & Adler, 2015; Lee, Noh & Kim, 2013). When looking at the effect of product type on engaging in eWOM, luxury products seems to be extra suitable for creating a positive impression in others’ mind. According to a

10

research from Hudders (2012), consumer purchase luxury goods rather from psychological benefits than from an utilitarian perspective. The consumer buys exclusivity and more importantly, status, which brings the opportunity to impress the social environment of the individual. Several papers make the distinction between utilitarian and luxury goods. Luxury goods are not necessary but make life more pleasant for the customer. The study of Dubois and Duquesne (1993) confirmed the more or less logical statement that the demand of luxury goods increases as income rises. He also concluded that the need for luxury goods are cultural related. Chen, Fay & Wang (2011) conducted a study about luxury products and consumerreview behaviour as a form of eWOM. The researchers found that ‘status concerns’ is the main driver for posting reviews about a luxury good (e.g. expensive car). This effect was significantly higher compared to utilitarian products. The research from Yang and Mattila (2017) confirms this by evaluating the relationship between product type, need for status and WOM intentions. Results indicate that parvenus (i.e. people who have a high need for gaining status) are more likely to engage in WOM about purchased luxury products than people who find status less important. When further studying on the topic eWOM behaviour about luxury products, the article of Thoumrungroje (2014) brings a relevant insight. Outcomes suggest that the intensity of someone’s social media activity is positively related to an increase in buying luxury products, which is mediated by eWOM. This means that the more time an individual spends on social media, the more he will be exposed to eWOM about luxury products, which results in more purchasing. \

11

Publicly and privately used luxury products

An innovative angle when researching eWOM about luxury products is the potential influence of using that particular product in a public or private situation. A ‘public’ product is exposed among other people who are aware of you using the product, in contradiction to private products that are mostly used at home (Baerden and Etzel, 1982; Bourne, 1957). The interest arose from different marketing related papers that made the distinction between public and private product use or consumption. For example, the results of Kulviwat, Bruner and Al-Shuridah (2009) indicate that there is a positive relationship between social influence and adoption intention for innovative products. This relationship is stronger when the innovation is meant for using in public (vs. private). The research of Ratner and Kahn (2002) gives an explanation about why public product use seems to have a bigger effect on different variables (e.g. variety seeking) than private product use. In public situations individuals are tending to process a favourable impression to others, which is perfectly in line with earlier described literature about the construct impression management (Goffman, 1978; Rosenfeld et. al, 1995). The paper of Berger and Ward (2010) confirms that the social environment is taken into account for products that are used in public. Results indicate that for those kind of luxury products, people prefer less explicit brand signals to differentiate from the mainstream. Also Bourne (1957) stated that opinions and influence of others are considered for public products. Within the context of WOM, a relatively old article about product evaluation related to public and private usage gives relevant insights (Graeff, 1996). Remarkably, more recent research about this topic is scarce. The relationships between the variables image congruence (i.e. match between brand image and desired personal image), self-monitoring and brand evaluation were investigated in the public vs. private context. The research concluded that

12

image congruence has a positive influence on brand evaluation for high self-monitoring people, when the product is used in public. This means that when high-self monitoring people feel closely related to the brand, they evaluate the brand’s publicly used product more positively than a product that is meant for private consumption. Next paragraph gives further explanation about the trait self-monitoring which is part of the conceptual model.

Self-monitoring

How does this social situation wants me to be and how can I act like that person? Individuals tend to convey behaviour that is suitable to the situational context they are in. Self-monitoring is considered as a trait which deals with controlling behaviour during a social interaction (Snyder, 1979). Practical examples are behaving appropriately during a work related occasion or adjusting voice textures and facial expressions. Self-monitoring can also trigger negative behaviour as lying if that is what the situation is calling for (Gangestad and Snyder, 2000). Most papers about self-monitoring state that individuals differ in expressive control and make therefore the distinction between low and high self-monitoring people. Relatively high self-monitoring people are the most flexible and able to adjust ‘the self’ among different situations and people. These individuals acquire information from their surrounding and spend time and effort on paying to social cues. In contradiction, low self-monitoring people act less flexible to personal interactions and rely more on the authentic self when facing social situations. They rather seek social information from others to discover which behaviour is appropriate (Arkin, Gabrenya, Appelman and Cochran, 1979; Brown and Kaldenberg, 1997; Gangestad and Snyder, 2000; Goffman, 1978; Kilduff and Day, 1994; Lennox and Wolf, 1984; Mill. 1984; Snyder, 1997).

13

The current research is interested in what kind of behaviour is perceived as appropriate and desirable for social situations that occur online (e.g. social networks as Instagram). Hogan (2010) stated that in these situations friends and followers form the social audience, whereby behaviour is expressing in posting status updates, liking, chatting and sharing. Research conducted about self-monitoring and Facebook has shown that there is a significant relationship between self-monitoring and extraversion on Facebook (Hall and Pennington, 2013). This means that Facebook situations are asking for more extravert behaviour, upon which high self-monitoring users respond. Furthermore, this paper confirms previous research with concluding that low self-monitoring users conveyed a more authentic online self. Another supporting research is from Zywica and Danowski (2008), demonstrating that online attractiveness can be achieved by showing outgoing behaviour. Engaging in eWOM might be a way for self-monitoring individuals to show extraversion and outgoing behaviour via for instance the contribution and creation of information (Muntinga et. al, 2011). This statement is confirmed in different papers which appoint that extraversion is a motivation to contribute in positive eWOM activities (Chiu, Hsieh, Kao & Lee, 2007; Gholamisaman, 2012; Husnain, Qureshi & Akhtar, 2016; Krämer and Winter, 2008; Soetarto, Yap & Sweeney, 2009). So, when an individual scores relatively high on the trait self-monitoring, he responds flexible to the social environment and is able to adjust his behaviour to what the situation is calling for. Literature suggests that for social network platforms presenting an outgoing and extravert self is preferred. It is also stated that eWOM behaviour is considered as a form of extraversion. Therefore it is expected that people who score relatively high on selfmonitoring, are more likely to engage in earlier discussed social eWOM activities as liking, sharing or posting about a luxury product via Instagram, which leads to the following hypothesis:

14

H1. The higher people score on self-monitoring, the more likely they are to engage actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram. As explained earlier in this study, it is expected that product evaluations are different for products that are used in public vs. private situations (Berger and Ward, 2010; Bourne (1957); Graeff, 1996; Kulviwat et. al, 2009; Ratner and Kahn 2002), which brings us to hypotheses 2a and 2b: H2 a. The higher people score on self-monitoring, the more likely they are to engage actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is bigger for products that are meant for using in public (in comparison with private)

H2 b. The higher people score on self-monitoring, the more likely they are to engage actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is smaller for products that are meant for using in private (in comparison with public)

Impression management (2.0)

People seem to be highly interested in how others evaluate and perceive them. The topic impression management has been studied in the fields of sociology, social psychology and marketing. Goffman (1978) is one of the first researchers who defined the phenomenon as a presentation of the self-provided the link between the way we perceive ourselves and the way we want others to perceive us. He stated that individuals control and manipulate the information they communicate to achieve a desired identity. More concrete, impression management is the process in which individuals try to create a positive impression in others

15

mind (Rosenfeld et. al, 1995; Schlenker, 1980; Tetlock and Manstead, 1985). Research from Leary and Kowalski (1990) conceptualized the concept impression management in two components. The first component involves the degree to which people are motivated to control the images others create of them, namely impression motivation. The second process is impression construction, which can be divided in five factors explaining impressive image building. The first factor is self-concept, which refers to how people see themselves. Self-concept is strongly related to the second factor which deals with the distinction between a desired and undesired identity. Furthermore, someone’s role in a group and related expectations (role constraints) influences the impression of an individual. The fourth factor deals with the consequences on the process of both current and potential social images. The last concept of the impression construction theory is target value and addresses how significant others value the given impression. Jones and Pittman (1982) determined five impression management strategies that people are most likely to use. The research distinguishes intimidating others, exemplification (i.e. behave towards requirements), supplication (i.e. advertise limitations to be seen as more needy), ingratiation (i.e. make use of flattery) and self-promotion. Most articles focus on the self-promotional aspect of impression management, which deals with showing competence by highlighting ones abilities and accomplishments (Ellis, West, Ryan & DeShon, 2002; Rudman, 1998; Singh, Kumra & Vinnicombe, 2002; Stevens and Kristof, 1995). Related topics to self-promotion are entitlement, whereby an individual claims a positive outcome, and enhancement which is about making convincing statements that upgrade ‘the self’ (Delery and Kacmar, 1998). Since the development of social network sites, impression management as a research topic is expanded to the online world. Platforms as Instagram and Facebook enable users to highly control impressions and promote the self, in comparison to face-to-face

16

communication. Desired social identities are built from idealized photos, updates and comments, whereby also someone’s friend list and number of connections are part of the online identity. Impression management is a major motive to participate in social network sites and some researchers even introduced the term ‘impression management 2.0’ (Cunningham, 2013; Krämer and Winter, 2008; Mehdizadeh, 2010). Besides earlier explained drivers as self-enhancement and self-promotion, the current research is interested in traits playing a role in impression management. Several papers link the earlier discussed trait self-monitoring to impression management behaviour (Caldwell and O'Reilly, 1982; Delery and Kacmar, 1998; Schlenker, 1980; Snyder, 1997; Turnley and Bolino, 2001). The paper of Turnley and Bolino (2001) suggests that high self-monitoring individuals make better use of self-promotional impression management behaviour. Relatively older literature from Caldwell and O'Reilly (1982) confirms this with concluding that individuals who are highly sensitive to social cues, are more likely to engage in forms of impression management. Referring to the first and second hypotheses of the current research, it is expected that relatively high self-monitoring people are better in adjusting behaviour to what the social situation is calling for. Considering earlier presented literature, this would result in more outgoing and extravert behaviour (e.g. eWOM) on social network sites. The underlying mechanism might be the desire for an impressive online identity. Apparently, it is important to enhance and promote ‘the self’ by showing other consumers that you are a person of expertise, good taste and great influence (Chu and Kim, 2011; Henning-Thurau et. al, 2004; Kim, Jang & Adler, 2015; Matta and Frost, 2011; Taylor, 2010; Xiao-chai and Bei-lei, 2009). Considering these statements, the following third hypothesis is formulated:

H3. The construct impression management mediates the effect of someone’s (relatively high)

17

self-monitoring level on engaging (more) actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram. Being consistent in the research context, it is expected that product evaluations are different for products that are used in public vs. private situations (Berger and Ward, 2010; Bourne (1957); Graeff, 1996; Kulviwat et. al, 2009; Ratner and Kahn 2002), which leads to the fourth hypotheses: H4 a. The construct impression management mediates the effect of someone’s (relatively high) self-monitoring level on engaging (more) actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is bigger for products that are meant for using in public (in comparison with private)

H4 b. The construct impression management mediates the effect of someone’s (relatively high) self-monitoring level on engaging (more) actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is smaller for products that are meant for using in private (in comparison with public)

Conceptual framework

Model 1 is a visual representation of hypotheses that derived from the literature review. The direct (H1 & H2) and indirect effect (H3 & H4) on eWOM (liking, sharing and posting behaviour) about luxury products will be tested for products that are used in public vs. private situations.

18 Model 1 Conceptual framework

19

METHODOLOGY

Participants

The final sample consisted of 121 participants women only. The ages of the respondents ranged from 18 to 52 years (M = 23.77, SD = 4.61). Most of the participants were higher educated combining their study with having a job (see table 1). All participants were using an Instagram account actively, otherwise they were not allowed to respond the survey. A few respondents were excluded because of two empty fill-ins and one case with an unrealistic age of 98. One man was deleted out of the dataset.

Table 1 Participants characteristics (within subjects design) N= 121

Variable

Frequency

Gender (M : F)

0 : 121

Education level Mavo Havo/Vwo MBO Higher educated Other

1 9 5 104 2

Working status Work Work and study Study only No work, no study

44 46 28 3

Time on Instagram Less than 10 minutes 10 – 30 minutes 31 – 60 minutes 61 – 80 minutes More than 180

20 44 35 6 2

20

Design

The study measured a quantitative research design whereby participants were exposed to both public and private product usage situations via an online survey. Before setting out the main survey, a pretest was conducted to check whether the products demonstrated in the survey were indeed perceived as for public or private usage. The questionnaires were online available addressing Qualtrics links and could be filled in via computes, laptop and mobile devices (e.g. smartphone, tablet). Data was analyzed using the statistical program SPSS.

Procedure

As mentioned before, the study started with a pretest. This survey was filled in by participants who did not participate in the main questionnaire to prevent practice effects. Main survey respondents were approached personally and via social media channels as Facebook messenger and Whatsapp. The participation was voluntary, anonymous and without reward. Preparatory conditions for the respondents were stated; they had to be woman, older than 16 and active on Instagram. The survey started with a relatively short cover story explaining the time to complete, emphasizing the target group and contact details of the researcher. Furthermore, the respondents were generally informed about the aim of the study telling that the survey examines Instagram behaviour. To make sure that respondents were active on Instagram, the first question aimed to control for this. Besides that, respondents were asked about the time spending on Instagram per day. The following part of the questionnaire started with an Instagram post of a woman presenting a luxury product that is meant for using in public situations (Gucci handbag). The

21

Instagram post included a description expressing how happy this person is with her new purchase. After being exposed to public condition, respondents were tested on eWOM behaviour and asked how likely they were to like, share or post an update like this. The same procedure was tested for an Instagram post concerning a privately used product (Gucci toiletry bag). Importantly, both products were from the same brand (Gucci) and same price category (450 euro’s) to keep the level of luxury equal. After all respondents were exposed to both public and private conditions, questions were asked about their impression management behaviour on Instagram. After the Instagram related part of the survey, respondents were confronted with questions measuring the trait self-monitoring in general. The survey ended with demographical items regarding gender, age, education and working status. A representation of the full survey can be found in appendices A.

Materials and measurements

Independent variable self-monitoring The construct self-monitoring is measured according to a classic scale of Lennox and Wolfe (1984). The name of this 7 items scale is ‘ability to modify self-presentation’ and one of the most popular measures since (Briggs and Cheek, 1986; Gangestad and Snyder; 2000). The scale is a revision and significant improvement of the first self-monitoring scale from Snyder (1974). Snyder’s scale is narrowed down, as it measured irrelevant constructs as social anxiety. In a recent study of Lee and Lim (2010), the Cronbrach’s Alpha of the Lennox and Wolfe (1984) self-monitoring scale resulted in α = .84. Each of the 7 statements was rated on a 5 point scale from strongly disagree to

22

strongly agree. The scale is testing the degree of someone’s ability to adjust behaviour according to the social situation (appendices B). Mediator impression management To measure someone’s impression management behaviour on Instagram, the scale items of Bearden and Rose (1990) were slightly rebuilt to make them more cohesive with the subject of social media (appendices B). The 5 point (strongly disagree to strongly agree) scale measures the degree to which a person is aware of the self and how this impresses others. Reported alphas of Bearden and Rose (1990) in different studies are α =.83, α = .74 and α = .79. The scale turned out to be successful in more recent studies as well, in for example the impression management related paper of Malär, Krohmer, Hoyer, and Nyffenegger (2011). There are existing scales created to examine social media and impression management, but none of them seemed suitable for this study. For example the scale of Vitak (2005), which measures impression management on the social media platform Facebook. The scale items were focusing on the likelihood of deleting or adjusting a recently posted status. Thinking beyond topic, these items could also test someone’s feelings of insecurity related to the platform. Dependent variable eWOM For both conditions (public and private) the following three questions were asked to measure eWOM behaviour; ‘How likely are you to like this post?, ‘how likely are you to share this post’? and ‘how likely are you to post an Instagram update like this yourself’? The items were measured on a 5 point likert scale. Scales as presented above are types of behavioural intention scales (Davis and Warshaw, 1992). These items are often used in marketing research to indicate consumers attitudes and perceptions, as for instance purchase or eWOM intentions (Garretson and burton; 2010, Kalwani and Silk, 1982; Kwon, Trail & James, 2007).

23

RESULTS

Pretest

Participants. The pretest sample consisted of 30 participant (women only) with an Instagram account. The ages of the respondents ranged from 19 to 53 years (M = 24.63, SD = 6.01). T-Test. A paired-sample T-test was conducted to compare a privately used product with a publicly used product. The private product reported significantly higher levels of private characteristics (M= 2.50, SD= 1.38) and the public product reported significantly higher levels of public characteristics (M= 4.83, SD= .46), t(29) = -8.84, p = .00)

Factor analyses and reliability (main study)

Factor analyses. A Principal Axis Factor (PAF) with a varimax rotation of 25 was conducted for 14 items belonging to the self-monitoring and impression management scales. First of all, the Bartlett’s Test of Sphericity turned out to be significant (p= .00) and the Kaiser-Meyer-Olkin measure of sampling adequecy was factorable (KMO = .710). The analyse loaded 4 factors capturing 62% of the variance that is in the original 14 variables (Eigenvalue < 1). Four items of the self-monitoring scale where significantly more loading on other factors (item 3, item 5, item 6, item 7). A new variable (SM_FA) was computed without these items to check whether this had a positive impact on the results. By doing this, no extra

24

significant outcomes were found. Therefore, the decision is made to keep the scales according to the classic 7 items which have proven to be successful. Reliability. The variables self-monitoring and impression management were both measured with 7 item scales. A Cronbrach’s alpha scale reliability test has been conducted to assess the internal consistency of the items and therefore the reliability (Cronbach, 1951). The assumption of testing one-dimensional items only has been met. According to De Heus, Leeden & Gazendam (1995), outcomes from α ≥ .70 onwards are interpret as reliable whereby α ≥ .80 should be interpret as very reliable. The results of the tests confirm that the impression management scale (α = .801) and self-monitoring scale (α = .723) are reliable.

Hypotheses testing

H1. The higher people score on self-monitoring, the more likely they are to engage actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram.

For testing this basis hypothesis, no distinction was made between the different forms of eWOM via Instagram (like, share, post) and contexts (private and public). All items measuring any form of eWOM were computed together into one variable. A simple linear regression analyses was calculated to predict eWOM based on self-monitoring. No significant findings came out of the analysis (R2 = .001, F(1,119)= .157, p = .69). It can be concluded that for this study self-monitoring did not predicted eWOM significantly (β = .05 , p = .69).

25

Table 2 coefficient of self-monitoring (IV) on eWOM (DV)

Private & Public Used Product

eWOM (DV)

β

p

.05

.69

R2

.001

H2 a. The higher people score on self-monitoring, the more likely they are to engage actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is bigger for products that are meant for using in public (in comparison with private)

H2 b. The higher people score on self-monitoring, the more likely they are to engage actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is smaller for products that are meant for using in private (in comparison with public)

Hypothesis 2 is designed in order to test the effect from self-monitoring on eWOM in both public (H2a) and private (H2b) contexts. Practically this means that statistics are carried out for each situation which means twice. Before testing the direct effect of self- monitoring on eWOM in both situations, a Ttest was conducted comparing the variables eWOM public (liking, sharing, posting computed

26

together) and eWOM private (liking, sharing, posting computed together). A paired-sample Ttest reported a significant difference between eWOM private (M= 1.61, SD= .64) and eWOM public (M= 2.00, SD= .74), t(120) = -6.77, p = .00). These results confirm that respondents engage more in eWOM activities about a luxury product when this product is used in public situations. Now the significant difference between eWOM public and eWOM private is determined and significantly confirmed, the variable self-monitoring can be included in the process. Testing the effect from self-monitoring on eWOM (private vs. public) is carried out in two steps. First, the items measuring eWOM (liking, sharing and posting) were computed together for both public and private contexts. This results in testing self-monitoring on the variable eWOM private (liking, sharing, posting computed) and eWOM public (liking, sharing, posting computed). eWOM private. A simple linear regression analyses was calculated to predict eWOM private based on self-monitoring. No significant findings came out of the analysis (R2 = .00, F(1,119)= .002, p = .88). It can be concluded that for this study self-monitoring did not predicted eWOM private significantly (β = -.02 , p = .88). eWOM public. A simple linear regression analyses was calculated to predict eWOM public based on self-monitoring. No significant findings came out of the analysis (R2 = .005, F(1,119)= .624, p = .43). It can be concluded that for this study self-monitoring did not predicted eWOM public significantly (β = .11 , p = .43). Remarkably and as expected, the coefficient (β = .11) turned out to be higher for the public condition (table 3).

27 Table 3 coefficients of self-monitoring (IV) on eWOM private and eWOM public

Private Used Product

eWOM (DV)

β

p

-.02

.88

Public Used Product

R2

.00

β

p

R2

.11

.43

.005

The second step of testing hypothesis 2 deals with testing self-monitoring separately on all six items that measures eWOM. This is carried out by 3 regression analyses for the private condition (on liking, sharing and posting) and 3 regression analyses for the public condition (on liking, sharing and posting). It is predicted that the higher people score on selfmonitoring, the more likely they are to engage in any researched form of eWOM. It is expected that this effect is bigger for eWOM about luxury products that are used in public situations. As can been seen in table 4, all coefficients resulted higher for the public context. This could be interpret as remarkable but not as a significant outcome.

28

Table 4 coefficients of self-monitoring (IV) on eWOM private and eWOM public for all items

Private Used Product

Public Used Product

β

P

β

p

Like

-.126

.508

.138

.532

Share

.024

.751

.039

.676

Post

.049

.780

.147

.475

eWOM variables (DV)

H3. The construct impression management mediates the effect of someone’s (relatively high) score of self-monitoring on engaging (more) actively in eWOM about a luxury product via Instagram.

From this point onwards, hypotheses are dealing with the potential mediation role of the construct impression management. For testing the third hypothesis, no distinction was made between the different forms of eWOM via Instagram (like, share, post) and conditions (private vs. public). The in total 6 items were computed together into one variable. A mediation analyses using 5,000 boot-strapped samples (Hayes 2012; PROCESS SPSS macro; model 4) is conducted. Conclusion of the analysis is that impression management does not mediate the effect of self-monitoring on eWOM significantly (β= .03, SE = .04, bootstrap CI:.03 to .13).

29

Besides testing the indirect effect, Post-Hoc analyses has been conducted using PROCESS to generate broader statistical insight. Model 2 shows beta’s for each relationship of the conceptual model. Most remarkable finding: results show a positive and significant effect from impression management (mediator) on eWOM (β= .22, SE = .08, p = .00, bootstrap CI: .06 to .37).

Model 2 Mediation of impression management on eWOM

H4 a. The construct impression management mediates the effect of someone’s (relatively high) self-monitoring level on engaging (more) actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is bigger for products that are meant for using in public (in comparison with private)

H4 b. The construct impression management mediates the effect of someone’s (relatively high) self-monitoring level on engaging (more) actively in eWOM (A. liking, B. sharing, C. posting) about a luxury product via Instagram, whereby this effect is smaller for products that are meant for using in private (in comparison with public) Testing this hypothesis has been realized in two steps. In consistency with hypothesis 2, for step one the items measuring eWOM (liking, sharing and posting) were computed

30

together for both private and public conditions. This results in testing the mediation on the dependent variables eWOM private (computed liking, sharing, posting) and eWOM public (computed liking, sharing, posting). eWOM private. Items measuring eWOM private were computed into one variable. A mediation analyses using 5,000 boot-strapped samples is conducted. Conclusion of the analysis is that impression management does not mediate the effect of self-monitoring on eWOM private significantly (β= .01, SE = .02, bootstrap CI: -.01 to .08). Post-Hoc analyses using PROCESS did not result in significant relationships, as can be seen in model 3.

Model 3 Mediation of impression management on eWOM Private

eWOM public. First of all, items measuring eWOM public were computed into one variable. A mediation analyses using 5,000 boot-strapped samples is conducted. Conclusion of the analysis is that impression management does not mediate the effect of self-monitoring on eWOM public significantly (β= .05, SE = .06, bootstrap CI: -.05 to .19). Post-Hoc analyses has been conducted using PROCESS. Model 4 shows outcomes for each relationship of the conceptual model. Most remarkable finding: results show a positive and significant effect of impression management (mediator) on eWOM public (β= .36, SE = .09, p = .00, bootstrap CI: .17 to .54).

31

Model 4 Mediation of impression management on eWOM public

Step two of testing hypothesis 4 is running analyses for each item separately which measured eWOM. More concrete, this means the indirect effect on eWOM liking, eWOM sharing and eWOM posting for both private and public conditions (total of 6 analyses). eWOM private. Using mediation analysis PROCESS, it is concluded that impression management does not mediate the effect of self-monitoring on eWOM private liking significantly (β= .003, SE = .03, bootstrap CI: -.04 to .08). Testing the same for eWOM sharing, resulting in impression management that does not mediate the effect of self-monitoring on eWOM private sharing significantly (β= -.004, SE = .01, bootstrap CI: -.05 to .01). Conducted PROCESS for eWOM private posting, it is concluded that impression management does not mediate the effect of self-monitoring on eWOM posting significantly (β= .03, SE = .05, bootstrap CI: -.03 to .17). Post-Hoc analyses (model 5) resulted in a marginally significant relationship between impression management and eWOM private posting behaviour (β= .23, SE = .12, p = .07, bootstrap CI: -.02 to .48).

32 Model 5 Mediation of impression management on eWOM private all items

eWOM public. Using mediation analysis PROCESS, it is concluded that impression management does not mediate the effect of self-monitoring on eWOM public liking significantly (β= .07, SE = .08, bootstrap CI: -.06 to .25). Testing the same for eWOM public sharing, resulting in impression management that does not mediate the effect of self-monitoring on eWOM sharing significantly (β= .01, SE = .02 , bootstrap CI: -.01 to .08). Conducted PROCESS for eWOM public posting, it is concluded that impression management does not mediate the effect of self-monitoring on eWOM posting significantly (β= .09, SE = .09 , bootstrap CI: -.10 to .28) Conducting Post-Hoc analyses, significant results were found for impression management on eWOM liking behaviour (β= .44 , p= .01, SE = .15 , bootstrap CI: .13 to .74). Also the effect of impression management on eWOM posting behaviour turned out to be significant as can be seen in model 6 (β= .57 , p= .00, SE = .14 , bootstrap CI: .29 to .84).

33 Model 6 Mediation of impression management on eWOM public all items

34

DISCUSSION

Discussion of the results

The current research represents one of the first attempts to study eWOM activities and impression management on Instagram. The core finding of the current research is the significant mean difference between eWOM about publicly used products and eWOM about privately used products. Furthermore, the in total four hypotheses conceptualized the broader aim of the study and can be divided in two main objectives. The first objective formulates the relationship between the variables self-monitoring and eWOM. Hereby, three different forms of eWOM (i.e. liking, sharing, posting) were examined for both privately and publicly used products. It was expected that results would show a significant relationship between selfmonitoring and eWOM, with bigger coefficients for publicly used products. The outcome of this direct effect was not significant. Remarkably and as expected, coefficients turned out higher for the public context. Despite that no hard conclusions can be drawn from this, the noticed difference between privately and publicly used products is in line with prior literature (Baerden and Etzel, 1982; Berger and Ward, 2010; Bourne, 1957; Kulviwat et. al, 2009; Ratner and Kahn, 2002). The second expectation related to the first objective stated that self-monitoring is positively related to any form of eWOM. Negative outcomes were found for eWOM private (table 3) and eWOM private liking (table 4). If these results were significant, the conclusion would suggest that high self-monitoring people are less likely to spread eWOM about a privately used product. A potential explanation can be found in the privacy concerns people have regarding Instagram and social media in general. Social media exist to share information

35

about the self, but there are some pitfalls and privacy boundaries where users could struggle with (Ellison, Vitak, Steinfield Gray and Lampe, 2011; Fogel and Nehmad, 2009; O'Keeffe and Clarke-Pearson, 2011). According to Young and Quan-Haase (2009) people experience concerns because of the ‘unwanted audience’ they get confronted with on social media. A possible consideration could be that publicly used products are viewed by an unwanted audience anyway. Sharing online information about a product that people use in private situations, might be one step over the privacy line. The second objective aimed to identify impression management as a moderating factor between self-monitoring and eWOM about luxury products. In terms of consistency, the distinction was made between privately and publicly used products. There was again determined a difference between these types of products. Remarkably, no significant confirmation resulted from the mediation analyses. Potential explanations can be found in exploratory analyses that showed a significant relationship between the mediator impression management and dependent (sub)variables eWOM (model 2), eWOM public (model 4), eWOM public like and eWOM public post (model 6). A marginally significant result was found for eWOM private post (model 5). In contradiction with the hypotheses, results from the Post-Hoc analyses suggest that impression management is driven by a trait different from self-monitoring or drives on its own. One explanation could be the trait extraversion which is discussed in the literature review. Several papers indicate that people with this characteristic process easier in impression management behaviour (Barrick and Mount, 1996; Krämer, 2008; Kristof-Brown, Barrick & Franke, 2002; Weiss and Feldman, 2006). Another and more extreme alternative is the trait narcissism, which is related to the construct impression management in a considerable amount of research (Kapidzic, 2013; Mehdizadeh, 2010; Ong, Ang, Ho, Lim, Goh, Lee & Chua, 2011).

36

Limitations and directions for future research

This study has a strong focus which comes along with the first limitation. The theory reviews more than one trait, but only self-monitoring was implemented in the final research design. For further research it is suggested to include the traits extraversion and narcissism. The second limitation addresses cultural differences within the sample. Participants came from different countries as The Netherlands, Germany and Italy. The paper about Hofstede’s cultural dimensions discusses the potential influence of the difference between individualism and collectivism. Impression management is about centralizing the self, which is much more appreciated in individualistic cultures than in societies living according to the collectivism (Hofstede, 1984). One suggestion for future research is to focus on a specific culture to avoid influences. It might be even more interesting to test the same survey among different countries to find out what the influence of the cultural dimension would be. The third limitation results from a critical evaluation about the online questionnaire. The researcher’s Instagram account was used to create visual examples that were part of the survey (appendices A). The username (i.e. lottelob) is mentioned and could have a positive or negative influence when answering the questions concerning with someone’s likelihood of sharing or liking that update. Fourthly, it is recommended to implement one more survey question which covers products that are meant for using in both private and public situations (e.g. luxury watch). This will provide a more complete research design and therefore stronger statistical power. The final recommendation deals with discussion around comparing beta’s in order to make (careful) statements about differences between the public and private contexts. Despite the fact that far from all research outcomes were based on this approach and research models

37

behind the statistics were similar, it is recommended to exclude this in general unusual approach in further research. An interesting alternative would be redesigning the model whereby public vs. private acts as one variable and moderator. The most suitable analysis would be moderation via PROCESS (Hayes, 2012). Unfortunately, the current dataset does not allow to view the research project from this approach already.

Theoretical implication

The current study both challenges and supports prior literature. Firstly, the study challenges the earlier proven relationship between self-monitoring and impression management. These conclusions were mostly drawn from an offline research environment (Caldwell and O'Reilly, 1982; Delery and Kacmar, 1998; Schlenker, 1980; Snyder, 1997; Turnley and Bolino, 2001). Apparently, within the online context of Instagram, selfmonitoring is not driving someone’s impression management activities. The second theoretical addition is the significant relationship between impression management and different forms of eWOM. No available prior research was found saying explicit that impression management drives eWOM behaviour on Instagram. This study achieved to bring a relatively new theoretical insight with statistical proof. The difference between public and private product usage is partly supported and partly challenged. Existing theory about this topic is relatively old and tested on differences mostly in an offline context (Berger and Ward, 2010; Bourne (1957); Graeff, 1996; Kulviwat et. al, 2009; Ratner and Kahn 2002). The present study has shown that in an online environment as Instagram there is a difference in eWOM intensity when comparing the means of eWOM public and eWOM private. Taking the full research model into consideration including the variables self-monitoring and impression management, there are differences, but without

38

significant beta’s. Therefore no hard conclusions can be drawn from this whereby outcomes can be interpret as remarkable only.

Managerial implication

The most useful finding for marketers is that someone’s need for impressing other people drives them to talk about your luxury products via Instagram. The role of impression management is relevant to consider when discussing the promotional part of launching a product in the market. A possible strategy is to approach Instagram users who are standing out in their impression management behaviour (e.g. high amount of followers, frequently viewed profile) to spread to worth positively about your products. This concept is also beneficial for the influencer because it gives her/him the opportunity to impress followers and gain status by showing off that particular product. Social media influencers are described as an independent third party who influence and shape the audience via pictures, blogging and video’s (e.g. vlogging). This relatively new interned-based marketing has proven to be successful within different fields as lifestyle, beauty and sports (Booth and Matic, 2011; Gillin, 2009; Graham, McGaughey & Freberg, 2011). It is important to keep the difference between privately and publicly used products in mind when setting out these strategies. It is expected that stimulating eWOM about a luxury car (public product) is more effective than triggering eWOM about a luxury coffee machine (private product).

39

APPENDICES

A) Survey

40

41

42

43

44

45

B) Scales

I. Self-monitoring scale Lennox and Wolfe (1984), 7 item likert scale ‘Ability to modify self-presentation’ 1. I have found that I can adjust my behaviour to meet the requirements of any situation I find myself in 2. I have the ability to control the way I come across to people, depending on my impression I wish to give them 3. Once I know what the situation calls for, it’s easy for me to regulate my actions accordingly 4. In social situations, I have the ability to alter my behaviour if I feel that something else is called for 5. If I feel that the image I am portraying isn’t working, I can readily change it to something that does 6. I have trouble changing my behaviour to suit different people and different situations 7. Even when it might to be my advantage, I have difficulty to putting up a good front

II. Original Impression Management scale Bearden and Rose (1990), 7 item likert scale ‘Self-consciousness in public’ 1. I am concerned about my style of doing things 2. I am concerned about the way I present myself 3. I am self-conscious about the way I look 4. I usually worry about making a good impression 5. One of the last things I do before leaving my house is look in the mirror 6. I am concerned about what other people think of me 7. I am usually aware of my appearance

III. Adjusted Impression Management scale In order to make suitable for Instagram 1. I am concerned about my style of doing things on Instagram 2. I am concerned about the way I present myself on Instagram 3. I am self-conscious about the way I look on Instagram 4. I usually worry about making a good impression on Instagram 5. One often check my Instagram profile to see what it looks like

46

6. I am concerned about what other people think of me when looking on my Instagram profile 7. I am usually aware of how I present myself on Instagram

C) Output SPSS / PROCESS

I. Prestest output What is your gender? Cumulative Frequency Valid

Female

Percent

30

Valid Percent

100,0

Percent

100,0

100,0

Descriptive Statistics N

Minimum

What is your age?

30

Valid N (listwise)

30

Maximum

19

Mean

53

Std. Deviation

24,63

6,009

Paired Samples Statistics Mean Pair 1

N

Std. Deviation

Std. Error Mean

Toiletry bag (private)

2,50

30

1,383

,253

Handbag (public)

4,83

30

,461

,084

Paired Samples Test Paired Differences 95% Confidence Interval of

Mean Pair 1 Toiletry bag (private) Handbag (public)

-2,333

Std.

Std. Error

Deviation

Mean

1,446

,264

the Difference Lower -2,873

Sig. (2-

Upper -1,793

t -8,836

df

tailed) 29

,000

47

II. Main study output Particpants

Gender Cumulative Frequency Valid

Female

121

Percent

Valid Percent

100,0

Percent

100,0

100,0

Descriptive Statistics N

Minimum

Age

121

Valid N (listwise)

121

Maximum

18

Mean

52

Std. Deviation

23,77

4,609

What is your highest level of education? Cumulative Frequency Valid

Percent

Valid Percent

Percent

Mavo

1

,8

,8

,8

Havo/vwo

9

7,4

7,4

8,3

Mbo (level 1 t/m 4)

5

4,1

4,1

12,4

104

86,0

86,0

98,3

Other

2

1,7

1,7

100,0

Total

121

100,0

100,0

Higher educated (hbo/university)

What is your current working status? Cumulative Frequency Valid

Percent

43

35,5

35,8

35,8

I study and have a job

46

38,0

38,3

74,2

Study only

28

23,1

23,3

97,5

3

2,5

2,5

100,0

120

99,2

100,0

1

,8

121

100,0

Total

Total

Valid Percent

I have a job only

No job, no study

Missing

Percent

System

48

Do you have an Instagram account at the moment? Cumulative Frequency Valid

Yes

Percent

121

Valid Percent

100,0

Percent

100,0

100,0

How many minutes do you spend on Instagram per day? Cumulative Frequency Valid

Percent

Valid Percent

Percent

Less than 10 minutes

20

16,5

16,5

16,5

10 - 30 minutes

44

36,4

36,4

52,9

31 - 60 minutes

35

28,9

28,9

81,8

61 - 120 minutes

14

11,6

11,6

93,4

121 - 180 minutes

6

5,0

5,0

98,3

More than 180 minutes

2

1,7

1,7

100,0

121

100,0

100,0

Total

Factor analysis KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett's Test of Sphericity

,710

Approx. Chi-Square

555,102

df

91

Sig.

,000

Total Variance Explained Extraction Sums of Squared

Rotation Sums of Squared

Loadings

Loadings

Initial Eigenvalues Compon

% of

Cumulative

Variance

%

Total

% of

Cumulative

Variance

%

Total

% of

Cumulative

Variance

%

ent

Total

1

3,500

25,003

25,003

3,500

25,003

25,003

2,731

19,508

19,508

2

2,724

19,461

44,464

2,724

19,461

44,464

2,337

16,696

36,204

3

1,318

9,416

53,880

1,318

9,416

53,880

1,943

13,880

50,084

4

1,138

8,129

62,009

1,138

8,129

62,009

1,670

11,925

62,009

5

,946

6,758

68,767

49

6

,851

6,077

74,844

7

,769

5,490

80,334

8

,641

4,576

84,910

9

,489

3,494

88,405

10

,452

3,232

91,637

11

,406

2,903

94,540

12

,321

2,291

96,831

13

,258

1,840

98,671

14

,186

1,329

100,000

Extraction Method: Principal Component Analysis.

Component Matrix

a

Component 1

2

3

4

To what extent do you agree or disagree with the following statements? - I am concerned about my style of

,695

-,393

-,092

-,453

,681

-,405

-,094

-,419

,568

-,140

-,452

,125

,558

-,495

,295

,081

,527

-,390

-,018

,329

posting updates on Instagram To what extent do you agree or disagree with the following statements? - I am concerned about the way I present myself on Instagram To what extent do you agree or disagree with the following statements? - I am self-conscious about the way I look on Instagram To what extent do you agree or disagree with the following statements? - I usually worry about making a good impression on Instagram To what extent do you agree or disagree with the following statements? - I often check my Instagram profile to see what it looks like

50

To what extent do you agree or disagree with the following statements? - I am concerned about what other

,426

-,521

,474

,095

,600

,016

-,195

,565

SM_3_Recode

,344

,422

,122

-,358

SM_7_Recode

,222

,422

-,477

-,017

,509

,499

,010

-,096

,512

,643

,047

-,188

people think of me when looking on my Instagram profile To what extent do you agree or disagree with the following statements? - I am usually aware of how I present myself on Instagram

The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - I have found that I can adjust my behaviour to meet the requirements of any situation I find myself in The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - I have the ability to control the way I come across to people, depending on my impression I wish to give them

51

The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the

,514

,441

-,073

,329

,346

,640

,323

,036

,169

,305

,626

,127

following statements? Once I know what the situation calls for, it’s easy for me to regulate my actions accordingly The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - In social situations, I have the ability to adjust my behaviour if I feel that something else is called for The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - If I feel that the image I am portraying isn’t working, I can readily change it to something that does work Extraction Method: Principal Component Analysis. a. 4 components extracted.

52

Rotated Component Matrix

a

Component 1

2

3

4

To what extent do you agree or disagree with the following statements? - I am concerned about my style of

,105

,898

,119

,143

,081

,874

,137

,150

,015

,433

,587

-,175

-,026

,445

,300

,601

-,105

,298

,571

,337

-,060

,328

,172

,738

,149

,053

,828

,078

posting updates on Instagram To what extent do you agree or disagree with the following statements? - I am concerned about the way I present myself on Instagram To what extent do you agree or disagree with the following statements? - I am self-conscious about the way I look on Instagram To what extent do you agree or disagree with the following statements? - I usually worry about making a good impression on Instagram To what extent do you agree or disagree with the following statements? - I often check my Instagram profile to see what it looks like To what extent do you agree or disagree with the following statements? - I am concerned about what other people think of me when looking on my Instagram profile To what extent do you agree or disagree with the following statements? - I am usually aware of how I present myself on Instagram

53

SM_3_Recode

,611

,208

-,131

-,079

SM_7_Recode

,296

,061

,263

-,543

,669

,138

,192

-,116

,813

,123

,106

-,166

,520

-,075

,540

-,069

The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - I have found that I can adjust my behaviour to meet the requirements of any situation I find myself in The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - I have the ability to control the way I come across to people, depending on my impression I wish to give them The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? Once I know what the situation calls for, it’s easy for me to regulate my actions accordingly

54

The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the

,769

-,177

,076

,083

,494

-,257

-,049

,466

following statements? - In social situations, I have the ability to adjust my behaviour if I feel that something else is called for The questions below are about your behaviour in general, so not about Instagram specifically. To what extent do you disagree or agree with the following statements? - If I feel that the image I am portraying isn’t working, I can readily change it to something that does work Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 8 iterations.

Component Transformation Matrix Component

1

2

3

4

1

,509

,605

,574

,216

2

,780

-,454

-,052

-,428

3

,311

-,227

-,357

,851

4

-,191

-,614

,735

,215

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.

55

Reliability self-monitoring scale Reliability Statistics Cronbach's Alpha Based on Cronbach’s

Standardized

Alpha

Items ,723

N of Items ,736

7

Reliability impression management scale Reliability Statistics Cronbach's Alpha Based on Cronbach’s

Standardized

Alpha

Items ,801

N of Items ,797

7

Hypothesis 1

Model Summary

Model

R

1

,036

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,001

-,007

,61930

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,060

1

,060

Residual

45,641

119

,384

Total

45,701

120

F

Sig. ,157

,692

b

a. Dependent Variable: eWOM_all b. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

,060

df

Mean Square 1

,060

F

Sig. ,157

,692

b

56

Residual

45,641

119

Total

45,701

120

,384

a. Dependent Variable: eWOM_all b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant) SM_var

Coefficients

Std. Error

Beta

1,639

,411

,045

,114

t

Sig.

3,992

,000

,397

,692

,036

a. Dependent Variable: eWOM_all

Hypothesis 2

T-test Paired Samples Statistics Mean Pair 1

N

Std. Deviation

Std. Error Mean

eWOM_pri

1,6061

121

,64118

,05829

eWOM_pu

1,9945

121

,74160

,06742

Paired Samples Correlations N Pair 1

eWOM_pri & eWOM_pu

Correlation 121

Sig.

,591

,000

Paired Samples Test Paired Differences 95% Confidence Interval of the Std. Error Mean Pair 1

eWOM_pri eWOM_pu

-,38843

Std. Deviation ,63150

Mean ,05741

Difference Lower -,50210

Upper -,27476

t -6,766

df 120

57

Model Summary

Model

R

1

,014

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,000

-,008

,64381

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,009

1

,009

Residual

49,324

119

,414

Total

49,333

120

F

Sig. ,022

,882

b

a. Dependent Variable: eWOM_pri b. Predictors: (Constant), SM_var Coefficients

a

Standardized Unstandardized Coefficients Model 1

B

Coefficients

Std. Error

Beta

(Constant)

1,669

,427

SM_var

-,018

,119

t

-,014

a. Dependent Variable: eWOM_pri

Model Summary

Model 1

R ,072

R Square a

,005

a. Predictors: (Constant), SM_var

Adjusted R

Std. Error of the

Square

Estimate

-,003

,74276

Sig.

3,910

,000

-,149

,882

58

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,344

1

,344

Residual

65,652

119

,552

Total

65,996

120

F

Sig. ,624

,431

b

a. Dependent Variable: eWOM_pu b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant)

Coefficients

Std. Error

Beta

1,609

,492

,108

,137

SM_var

t

Sig.

3,268

,001

,790

,431

,072

a. Dependent Variable: eWOM_pu

Model Summary

Model

R

1

,061

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,004

-,005

1,027

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,465

1

,465

Residual

125,502

119

1,055

Total

125,967

120

F

Sig. ,441

,508

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to like (dubble tab) this Instagram post? b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model

B

Std. Error

Coefficients Beta

t

Sig.

b

59

1

(Constant)

2,464

,681

SM_var

-,126

,189

3,620

,000

-,664

,508

-,061

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to like (dubble tab) this Instagram post?

Model Summary

Model

R

1

,029

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,001

-,008

,403

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,016

1

,016

Residual

19,339

119

,163

Total

19,355

120

F

Sig. ,101

,751

b

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to repost this Instagram post? b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant)

Coefficients

Std. Error

Beta

1,089

,267

,024

,074

SM_var

t

,029

Sig.

4,077

,000

,318

,751

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to repost this Instagram post?

Model Summary

Model 1

R ,026

R Square a

,001

a. Predictors: (Constant), SM_var

Adjusted R

Std. Error of the

Square

Estimate

-,008

,954

60

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,071

1

,071

Residual

108,193

119

,909

Total

108,264

120

F

Sig. ,078

,780

b

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to post an Instagram update like this yourself? b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant)

Coefficients

Std. Error

Beta

1,453

,632

,049

,176

SM_var

t

,026

Sig.

2,299

,023

,280

,780

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to post an Instagram update like this yourself?

Model Summary

Model

R

1

,057

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,003

-,005

1,193

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,559

1

,559

Residual

169,507

119

1,424

Total

170,066

120

F

Sig. ,393

,532

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to like (dubble tab) this Instagram post? b. Predictors: (Constant), SM_var

b

61

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant)

Coefficients

Std. Error

Beta

2,195

,791

,138

,220

SM_var

t

Sig.

2,774

,006

,627

,532

,057

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to like (dubble tab) this Instagram post?

Model Summary

Model

R

1

,038

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,001

-,007

,510

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,046

1

,046

Residual

30,930

119

,260

Total

30,975

120

F

Sig. ,175

,676

b

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to repost this Instagram post? b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant) SM_var

Std. Error 1,083

,338

,039

,094

Coefficients Beta

t

,038

Sig.

3,204

,002

,419

,676

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to repost this Instagram post?

Model Summary

62

Model

R

1

,066

R Square a

Adjusted R

Std. Error of the

Square

Estimate

,004

-,004

1,114

a. Predictors: (Constant), SM_var

a

ANOVA Model 1

Sum of Squares Regression

df

Mean Square

,638

1

,638

Residual

147,692

119

1,241

Total

148,331

120

F

Sig. ,514

,475

b

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to post an Instagram update like this yourself? b. Predictors: (Constant), SM_var

Coefficients

a

Standardized Unstandardized Coefficients Model 1

B (Constant) SM_var

Std. Error 1,550

,738

,147

,205

Coefficients Beta

t

,066

Sig.

2,098

,038

,717

,475

a. Dependent Variable: Please answer the questions below according to your personal likelihood How likely are you to post an Instagram update like this yourself?

Hypothesis 3 ************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = eWOM_all X = SM_var M = IM_var Sample size 121

63

************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: eWOM_all Model Summary R p ,2448 ,0261

R-sq

MSE

F

df1

df2

,0599

,3641

3,7607

2,0000

118,0000

Model constant IM_var SM_var

coeff 1,0680 ,2160 ,0127

se ,4520 ,0796 ,1118

t 2,3629 2,7122 ,1140

p ,0198 ,0077 ,9094

LLCI ,1729 ,0583 -,2086

ULCI 1,9631 ,3737 ,2341

************************** TOTAL EFFECT MODEL **************************** Outcome: eWOM_all Model Summary R p ,0363 ,6924

R-sq

MSE

F

df1

df2

,0013

,3835

,1572

1,0000

119,0000

Model constant SM_var

coeff 1,6390 ,0452

se ,4105 ,1141

t 3,9924 ,3965

p ,0001 ,6924

LLCI ,8261 -,1806

ULCI 2,4519 ,2711

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,0452 ,1141

t ,3965

p ,6924

LLCI -,1806

ULCI ,2711

Direct effect of X on Y Effect SE ,0127 ,1118

t ,1140

p ,9094

LLCI -,2086

ULCI ,2341

Indirect effect of X on Y Effect Boot SE IM_var ,0325 ,0388

BootLLCI -,0282

BootULCI ,1335

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0526 ,0622 -,0469 ,2069

64

Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0261 ,0310 -,0220 ,1065 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,7183 90,4209 ,0976 6268,8471 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var 2,5498 36,4571 1,1354 366,7167 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var ,0012 ,0074 -,0070 ,0282 Normal theory tests for indirect effect Effect se Z p ,0325 ,0318 1,0219 ,3068 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

Hypothesis 4 ************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = eWOM_pri X = SM_var M = IM_var Sample size 121 **************************************************************************

65 Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: eWOM_pri Model Summary R p ,0810 ,6779

R-sq

MSE

F

df1

df2

,0066

,4153

,3901

2,0000

118,0000

Model constant IM_var SM_var

coeff 1,4731 ,0741 -,0288

se ,4828 ,0851 ,1194

t 3,0513 ,8707 -,2409

p ,0028 ,3857 ,8100

LLCI ,5171 -,0944 -,2651

ULCI 2,4291 ,2425 ,2076

************************** TOTAL EFFECT MODEL **************************** Outcome: eWOM_pri Model Summary R p ,0136 ,8821

R-sq

MSE

F

df1

df2

,0002

,4145

,0221

1,0000

119,0000

Model constant SM_var

coeff 1,6689 -,0176

se ,4268 ,1186

t 3,9104 -,1486

p ,0002 ,8821

LLCI ,8238 -,2524

ULCI 2,5139 ,2172

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE -,0176 ,1186

t -,1486

p ,8821

LLCI -,2524

ULCI ,2172

Direct effect of X on Y Effect SE -,0288 ,1194

t -,2409

p ,8100

LLCI -,2651

ULCI ,2076

Indirect effect of X on Y Effect Boot SE IM_var ,0111 ,0212

BootLLCI -,0134

BootULCI ,0822

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0174 ,0330 -,0216 ,1281 Completely standardized indirect effect of X on Y

66

IM_var

Effect ,0086

Boot SE ,0166

BootLLCI -,0105

BootULCI ,0650

Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,6323 5,128E+011 -237,6483 -,1647 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,3874 14,7957 -969,6419 -,0273 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var -,0003 ,0028 -,0132 ,0020 Normal theory tests for indirect effect Effect se Z p ,0111 ,0193 ,5777 ,5635 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 WARNING: Bootstrap CI endpoints below not trustworthy. or increase bootstraps -237,6483 -969,6419

Decrease confidence

Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = eWOM_pu X = SM_var M = IM_var Sample size 121

67

************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: eWOM_pu Model Summary R p ,3415 ,0007

R-sq

MSE

F

df1

df2

,1167

,4940

7,7914

2,0000

118,0000

Model constant IM_var SM_var

coeff ,6630 ,3580 ,0542

se ,5265 ,0928 ,1302

t 1,2592 3,8582 ,4166

p ,2105 ,0002 ,6777

LLCI -,3797 ,1742 -,2036

ULCI 1,7057 ,5417 ,3121

************************** TOTAL EFFECT MODEL **************************** Outcome: eWOM_pu Model Summary R p ,0722 ,4311

R-sq

MSE

F

df1

df2

,0052

,5517

,6242

1,0000

119,0000

Model constant SM_var

coeff 1,6092 ,1081

se ,4924 ,1368

t 3,2682 ,7901

p ,0014 ,4311

LLCI ,6342 -,1628

ULCI 2,5841 ,3789

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,1081 ,1368

t ,7901

p ,4311

LLCI -,1628

ULCI ,3789

Direct effect of X on Y Effect SE ,0542 ,1302

t ,4166

p ,6777

LLCI -,2036

ULCI ,3121

Indirect effect of X on Y Effect Boot SE IM_var ,0538 ,0604

BootLLCI -,0519

BootULCI ,1902

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0726 ,0808 -,0723 ,2484

68

Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0360 ,0402 -,0333 ,1276 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,4981 34,9863 -,3015 153,4635 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,9925 99,3087 ,1254 1089,4800 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var ,0039 ,0112 -,0070 ,0491 Normal theory tests for indirect effect Effect se Z p ,0538 ,0493 1,0917 ,2750 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = Ewom_pr1 X = SM_var M = IM_var Sample size

69 121 ************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: Ewom_pr1 Model Summary R p ,0620 ,7967

R-sq

MSE

F

df1

df2

,0038

1,0634

,2277

2,0000

118,0000

Model constant IM_var SM_var

coeff 2,4157 ,0184 -,1283

se ,7725 ,1361 ,1910

t 3,1271 ,1350 -,6719

p ,0022 ,8928 ,5030

LLCI ,8859 -,2512 -,5066

ULCI 3,9454 ,2879 ,2499

************************** TOTAL EFFECT MODEL **************************** Outcome: Ewom_pr1 Model Summary R p ,0608 ,5080

R-sq

MSE

F

df1

df2

,0037

1,0546

,4408

1,0000

119,0000

Model constant SM_var

coeff 2,4642 -,1256

se ,6808 ,1891

t 3,6199 -,6640

p ,0004 ,5080

LLCI 1,1163 -,5001

ULCI 3,8122 ,2489

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE -,1256 ,1891

t -,6640

p ,5080

LLCI -,5001

ULCI ,2489

Direct effect of X on Y Effect SE -,1283 ,1910

t -,6719

p ,5030

LLCI -,5066

ULCI ,2499

Indirect effect of X on Y Effect Boot SE IM_var ,0028 ,0296

BootLLCI -,0444

BootULCI ,0854

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI

70 IM_var

,0027

,0292

-,0451

,0839

Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0013 ,0147 -,0217 ,0430 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,0220 8,284E+011 -6,3888 ,4995 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,0215 7,4698 -2,4762 ,6995 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var -,0001 ,0027 -,0089 ,0034 Normal theory tests for indirect effect Effect se Z p ,0028 ,0270 ,1025 ,9184 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = Ewom_pr2 X = SM_var M = IM_var Sample size 121

71 ************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: Ewom_pr2 Model Summary R p ,0558 ,8318

R-sq

MSE

F

df1

df2

,0031

,1635

,1844

2,0000

118,0000

Model constant IM_var SM_var

coeff 1,1626 -,0277 ,0277

se ,3029 ,0534 ,0749

t 3,8379 -,5182 ,3704

p ,0002 ,6053 ,7117

LLCI ,5627 -,1334 -,1206

ULCI 1,7624 ,0780 ,1761

************************** TOTAL EFFECT MODEL **************************** Outcome: Ewom_pr2 Model Summary R p ,0291 ,7513

R-sq

MSE

F

df1

df2

,0008

,1625

,1009

1,0000

119,0000

Model constant SM_var

coeff 1,0895 ,0236

se ,2672 ,0742

t 4,0769 ,3177

p ,0001 ,7513

LLCI ,5603 -,1234

ULCI 1,6186 ,1706

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,0236 ,0742

t ,3177

p ,7513

LLCI -,1234

ULCI ,1706

Direct effect of X on Y Effect SE ,0277 ,0749

t ,3704

p ,7117

LLCI -,1206

ULCI ,1761

Indirect effect of X on Y Effect Boot SE IM_var -,0042 ,0131

BootLLCI -,0490

BootULCI ,0108

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,0104 ,0332 -,1315 ,0265

72 Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,0051 ,0168 -,0674 ,0127 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,1764 8,210E+010 -2,71E+011 ,0337 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var -,1499 54,5968 -466,1928 ,0656 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var -,0003 ,0032 -,0186 ,0018 Normal theory tests for indirect effect Effect se Z p -,0042 ,0111 -,3742 ,7082 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = Ewom_pr3 X = SM_var M = IM_var Sample size 121

73

************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: Ewom_pr3 Model Summary R p ,1705 ,1755

R-sq

MSE

F

df1

df2

,0291

,8908

1,7662

2,0000

118,0000

Model constant IM_var SM_var

coeff ,8410 ,2315 ,0143

se ,7070 ,1246 ,1748

t 1,1896 1,8581 ,0819

p ,2366 ,0656 ,9349

LLCI -,5591 -,0152 -,3319

ULCI 2,2412 ,4782 ,3605

************************** TOTAL EFFECT MODEL **************************** Outcome: Ewom_pr3 Model Summary R p ,0256 ,7801

R-sq

MSE

F

df1

df2

,0007

,9092

,0783

1,0000

119,0000

Model constant SM_var

coeff 1,4529 ,0491

se ,6321 ,1756

t 2,2987 ,2798

p ,0233 ,7801

LLCI ,2014 -,2986

ULCI 2,7045 ,3968

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,0491 ,1756

t ,2798

p ,7801

LLCI -,2986

ULCI ,3968

Direct effect of X on Y Effect SE ,0143 ,1748

t ,0819

p ,9349

LLCI -,3319

ULCI ,3605

Indirect effect of X on Y Effect Boot SE IM_var ,0348 ,0457

BootLLCI -,0257

BootULCI ,1665

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0367 ,0476 -,0278 ,1750

74

Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0182 ,0238 -,0125 ,0888 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,7086 5,004E+011 ,1221 508,6308 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var 2,4315 15,5544 1,4701 599,7449 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var ,0006 ,0043 -,0040 ,0167 Normal theory tests for indirect effect Effect se Z p ,0348 ,0385 ,9045 ,3657 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = Ewom_pu1 X = SM_var M = IM_var Sample size

75 121 ************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: Ewom_pu1 Model Summary R p ,2601 ,0160

R-sq

MSE

F

df1

df2

,0677

1,3437

4,2814

2,0000

118,0000

Model constant IM_var SM_var

coeff 1,0405 ,4367 ,0721

se ,8684 ,1530 ,2147

t 1,1982 2,8542 ,3356

p ,2332 ,0051 ,7377

LLCI -,6791 ,1337 -,3531

ULCI 2,7601 ,7397 ,4973

************************** TOTAL EFFECT MODEL **************************** Outcome: Ewom_pu1 Model Summary R p ,0574 ,5321

R-sq

MSE

F

df1

df2

,0033

1,4244

,3927

1,0000

119,0000

Model constant SM_var

coeff 2,1948 ,1377

se ,7912 ,2198

t 2,7742 ,6267

p ,0064 ,5321

LLCI ,6283 -,2975

ULCI 3,7614 ,5730

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,1377 ,2198

t ,6267

p ,5321

LLCI -,2975

ULCI ,5730

Direct effect of X on Y Effect SE ,0721 ,2147

t ,3356

p ,7377

LLCI -,3531

ULCI ,4973

Indirect effect of X on Y Effect Boot SE IM_var ,0657 ,0776

BootLLCI -,0571

BootULCI ,2485

Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI

76 IM_var

,0552

,0654

-,0500

,2082

Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0273 ,0324 -,0226 ,1062 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,4768 3,615E+011 -,1149 198,3918 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,9114 56,0863 ,1190 2553,3843 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var ,0024 ,0087 -,0059 ,0386 Normal theory tests for indirect effect Effect se Z p ,0657 ,0635 1,0343 ,3010 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = Ewom_pu2 X = SM_var M = IM_var

77 Sample size 121 ************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: Ewom_pu2 Model Summary R p ,1050 ,5196

R-sq

MSE

F

df1

df2

,0110

,2596

,6583

2,0000

118,0000

Model constant IM_var SM_var

coeff ,8931 ,0718 ,0285

se ,3817 ,0673 ,0944

t 2,3398 1,0682 ,3022

p ,0210 ,2876 ,7631

LLCI ,1372 -,0613 -,1584

ULCI 1,6489 ,2050 ,2154

************************** TOTAL EFFECT MODEL **************************** Outcome: Ewom_pu2 Model Summary R p ,0384 ,6761

R-sq

MSE

F

df1

df2

,0015

,2599

,1754

1,0000

119,0000

Model constant SM_var

coeff 1,0829 ,0393

se ,3380 ,0939

t 3,2044 ,4188

p ,0017 ,6761

LLCI ,4138 -,1466

ULCI 1,7521 ,2252

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,0393 ,0939

t ,4188

p ,6761

LLCI -,1466

ULCI ,2252

Direct effect of X on Y Effect SE ,0285 ,0944

t ,3022

p ,7631

LLCI -,1584

ULCI ,2154

Indirect effect of X on Y Effect Boot SE IM_var ,0108 ,0187

BootLLCI -,0088

BootULCI ,0785

Partially standardized indirect effect of X on Y

78

IM_var

Effect ,0213

Boot SE ,0344

BootLLCI -,0193

BootULCI ,1368

Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0105 ,0174 -,0087 ,0708 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,2748 10,3877 ,0071 166,3008 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,3788 4,2863 ,0530 145,6186 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var ,0007 ,0044 -,0020 ,0228 Normal theory tests for indirect effect Effect se Z p ,0108 ,0161 ,6691 ,5034 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

************* PROCESS Procedure for SPSS Release 2.16.3 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com Documentation available in Hayes (2013). www.guilford.com/p/hayes3 ************************************************************************** Model = 4 Y = Ewom_pu3 X = SM_var M = IM_var

79

Sample size 121 ************************************************************************** Outcome: IM_var Model Summary R p ,1072 ,2420

R-sq

MSE

F

df1

df2

,0115

,4823

1,3826

1,0000

119,0000

Model constant SM_var

coeff 2,6432 ,1504

se ,4604 ,1279

t 5,7416 1,1759

p ,0000 ,2420

LLCI 1,7317 -,1029

ULCI 3,5548 ,4036

************************************************************************** Outcome: Ewom_pu3 Model Summary R p ,3577 ,0003

R-sq

MSE

F

df1

df2

,1280

1,0962

8,6578

2,0000

118,0000

Model constant IM_var SM_var

coeff ,0554 ,5653 ,0621

se ,7843 ,1382 ,1939

t ,0707 4,0906 ,3204

p ,9438 ,0001 ,7492

LLCI -1,4977 ,2916 -,3219

ULCI 1,6086 ,8390 ,4462

************************** TOTAL EFFECT MODEL **************************** Outcome: Ewom_pu3 Model Summary R p ,0656 ,4746

R-sq

MSE

F

df1

df2

,0043

1,2411

,5144

1,0000

119,0000

Model constant SM_var

coeff 1,5497 ,1472

se ,7385 ,2052

t 2,0985 ,7172

p ,0380 ,4746

LLCI ,0874 -,2591

ULCI 3,0120 ,5534

***************** TOTAL, DIRECT, AND INDIRECT EFFECTS ******************** Total effect of X on Y Effect SE ,1472 ,2052

t ,7172

p ,4746

LLCI -,2591

ULCI ,5534

Direct effect of X on Y Effect SE ,0621 ,1939

t ,3204

p ,7492

LLCI -,3219

ULCI ,4462

Indirect effect of X on Y Effect Boot SE IM_var ,0850 ,0922

BootLLCI -,0962

BootULCI ,2735

80 Partially standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0765 ,0830 -,0889 ,2433 Completely standardized indirect effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,0379 ,0414 -,0400 ,1253 Ratio of indirect to total effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var ,5778 8,775E+011 -,2475 81,7311 Ratio of indirect to direct effect of X on Y Effect Boot SE BootLLCI BootULCI IM_var 1,3683 41,1605 ,2997 1624,0072 R-squared mediation effect size (R-sq_med) Effect Boot SE BootLLCI BootULCI IM_var ,0035 ,0106 -,0087 ,0435 Normal theory tests for indirect effect Effect se Z p ,0850 ,0773 1,1001 ,2713 ******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 5000 Level of confidence for all confidence intervals in output: 95,00 NOTE: Kappa-squared is disabled from output as of version 2.16. ------ END MATRIX -----

BIBLIOGRAPHY

Alhabash, S., McAlister, A. R., Lou, C., & Hagerstrom, A. (2015). From Clicks to Behaviors: The Mediating Effect of Intentions to Like, Share, and Comment on the Relationship between Message Evaluations and Offline Behavioral Intentions. Journal of Interactive Advertising, 15(2), 82-96. Arkin, R. M., Gabrenya Jr, W. K., Appelman, A. S., & Cochran, S. T. (1979). Selfpresentation, self-monitoring, and the self-serving bias in causal attribution.

81

Personality and Social Psychology Bulletin, 5(1), 73-76. Arndt, J. (1967). Role of product-related conversations in the diffusion of a new product. Journal of marketing Research, 291-295. Bachleda, C., & Berrada-Fathi, B. (2016). Is negative eWOM more influential than negative eWOM?. Journal of Service Theory and Practice, 26(1), 109-132. Barrick, M. R., & Mount, M. K. (1996). Effects of impression management and selfdeception on the predictive validity of personality constructs. Journal of applied psychology, 81(3), 261. Bearden, W. O., & Etzel, M. J. (1982). Reference group influence on product and brand purchase decisions. Journal of consumer research, 9(2), 183-194. Berger, J., & Ward, M. (2010). Subtle signals of inconspicuous consumption. Journal of Consumer Research, 37(4), 555-569. Briggs, S. R., & Cheek, J. M. (1986). The role of factor analysis in the development and evaluation of personality scales. Journal of personality, 54(1), 106-148. Bolino, M. C., & Turnley, W. H. (1999). Measuring impression management in organizations: A scale development based on the Jones and Pittman taxonomy. Organizational Research Methods, 2(2), 187-206. Booth, N., & Matic, J. A. (2011). Mapping and leveraging influencers in social media to shape corporate brand perceptions. Corporate Communications: An International Journal, 16(3), 184-191. Bourne, F. S. (1957). Group Influence in Marketing and Public Relations, in Some Applications of Behavioral Research,(eds.) Rensis Likert and Samuel P. Hayes, Paris, France: Unesco. Browne, B. A., & Kaldenberg, D. O. (1997). Conceptualizing self-monitoring: links to materialism and product involvement. Journal of Consumer Marketing, 14(1), 31-44.

Buttle, F. A. (1998). Word of mouth: understanding and managing referral marketing. Journal of strategic marketing, 6(3), 241-254. Caldwell, D. F., & O'Reilly, C. A. (1982). Responses to failure: The effects of choice and responsibility on impression management. Academy of management journal, 25(1), 121-136.

82

Charlett, D., Garland, R., & Marr, N. (1995). How damaging is negative word of mouth. Marketing Bulletin, 6(1), 42-50. Chen, C. Y., Chen, T. H., Chen, Y. H., Chen, C. L., & Yu, S. E. (2013). The spatio-temporal distribution of different types of messages and personality traits affecting the eWOM of Facebook. Natural hazards, 65(3), 2077-2103. Chen, Y., Fay, S., & Wang, Q. (2011). The role of marketing in social media: How online consumer reviews evolve. Journal of Interactive Marketing, 25(2), 85-94. Cheung, C. M., Lee, M. K., & Rabjohn, N. (2008). The impact of electronic word-of-mouth: The adoption of online opinions in online customer communities. Internet research, 18(3), 229-247. Cheung, C. M., & Thadani, D. R. (2012). The impact of electronic word-of-mouth communication: A literature analysis and integrative model. Decision support systems, 54(1), 461-470. Chiu, H. C., Hsieh, Y. C., Kao, Y. H., & Lee, M. (2007). The determinants of email receivers' disseminating behaviors on the Internet. Journal of Advertising Research, 47(4), 524534.

Choi, J., & Kim, Y. (2014). The moderating effects of gender and number of friends on the relationship between self-presentation and brand-related word-of-mouth on Facebook. Personality and Individual Differences, 68, 1-5. Chu, S. C., & Kim, Y. (2011). Determinants of consumer engagement in electronic word-ofmouth (eWOM) in social networking sites. International journal of Advertising, 30(1), 47-75. Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. psychometrika, 16(3), 297-334. Cunningham, C. (2013). Social networking and impression management: selfpresentation in the digital age. Rowman & Littlefield. Daugherty, T., & Hoffman, E. (2014). eWOM and the importance of capturing consumer attention within social media. Journal of Marketing Communications, 20(1-2), 82-102. Davis, F. D., & Warshaw, P. R. (1992). What do intention scales measure?. The Journal of General Psychology, 119(4), 391-407. Day, G. S. (1971). Attitude change, media and word of mouth. Journal of Advertising

83

Research. Delery, J. E., & Kacmar, K. M. (1998). The Influence of Applicant and Interviewer Characteristics on the Use of Impression Management1. Journal of Applied Social Psychology, 28(18), 1649-1669. Dubois, B., & Duquesne, P. (1993). The market for luxury goods: Income versus culture. European Journal of Marketing, 27(1), 35-44. Ellis, A. P., West, B. J., Ryan, A. M., & DeShon, R. P. (2002). The use of impression management tactics in structured interviews: a function of question type?. Journal of Applied Psychology, 87(6), 1200. Ellison, N. B., Vitak, J., Steinfield, C., Gray, R., & Lampe, C. (2011). Negotiating privacy concerns and social capital needs in a social media environment. In Privacy online (pp. 19-32). Springer Berlin Heidelberg.

Engel, J. F., Kegerreis, R. J., & Blackwell, R. D. (1969). Word-of-mouth communication by the innovator. The Journal of Marketing, 15-19. Fogel, J., & Nehmad, E. (2009). Internet social network communities: Risk taking, trust, and privacy concerns. Computers in human behavior, 25(1), 153-160. Freberg, K., Graham, K., McGaughey, K., & Freberg, L. A. (2011). Who are the social media influencers? A study of public perceptions of personality. Public Relations Review, 37(1), 90-92. Gangestad, S. W., & Snyder, M. (2000). Self-monitoring: Appraisal and reappraisal. Psychological bulletin, 126(4), 530. Garretson, J. A., & Burton, S. (2000). Effects of nutrition facts panel values, nutrition claims, and health claims on consumer attitudes, perceptions of disease-related risks, and trust. Journal of Public Policy & Marketing, 19(2), 213-227. Gholamisaman, E. (2012). Social and personal factors influencing individuals participation in eWOM on social networking sites. BOOK OF, 161. Gillin, P. (2009). The new influencers: A marketer's guide to the new social media. Linden Publishing. Goffman, E. (1978). The presentation of self in everyday life (p. 56). Harmondsworth. Graeff, T. R. (1996). Image congruence effects on product evaluations: The role of

84

self‐monitoring and public/private consumption. Psychology & Marketing, 13(5), 481499. Gruen, T. W., Osmonbekov, T., & Czaplewski, A. J. (2006). eWOM: The impact of customer-to-customer online know-how exchange on customer value and loyalty. Journal of Business research, 59(4), 449-456. Hall, J. A., & Pennington, N. (2013). Self-monitoring, honesty, and cue use on Facebook: The relationship with user extraversion and conscientiousness. Computers in Human Behavior, 29(4), 1556-1564. Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling. White paper, http://www.afhayes.com/public/process2012.pdf, accessed on 23/05/2014 Hennig-Thurau, T., Gwinner, K. P., Walsh, G., & Gremler, D. D. (2004). Electronic word-ofmouth via consumer-opinion platforms: what motivates consumers to articulate themselves on the internet?. Journal of interactive marketing, 18(1), 38-52. Heus, P. D., & Leeden, R. van der, & Gazendam, B.(1995). Toegepaste data-analyse: Technieken voor niet-experimenteel onderzoek in de sociale wetenschappen. Hofstede, G. (1984). Cultural dimensions in management and planning. Asia Pacific journal of management, 1(2), 81-99. Hogan, B. (2010). The presentation of self in the age of social media: Distinguishing performances and exhibitions online. Bulletin of Science, Technology & Society, 0270467610385893. HU, X., & HA, L. (2015). Which form of word-of-mouth is more important to online shoppers? A comparative study of WOM use between general population and college students. Journal of Communication and Media Research, 7(2), 15-35. Hudders, L. (2012). Why the devil wears Prada: Consumers’ purchase motives for luxuries. Journal of Brand Management, 19(7), 609-622. Husnain, M., Qureshi, I., Fatima, T., & Akhtar, W. (2016). The Impact of Electronic Wordof-Mouth on Online Impulse Buying Behavior: The Moderating role of Big 5 Personality Traits. J Account Mark, 5(190), 2. Instagram (2016), about us and FAQ. Retrieved Mai 4, 2017, from https://www.instagram.com/about/us Internet Live Stats (2016), Retrieved May 28, 2017, from

85

http://www.internetlivestats.com/one-second/ Jones, E. E., & Pittman, T. S. (1982). Toward a general theory of strategic self-presentation. Psychological perspectives on the self, 1, 231-262. Kalwani, M. U., & Silk, A. J. (1982). On the reliability and predictive validity of purchase intention measures. Marketing Science, 1(3), 243-286. Kapidzic, S. (2013). Narcissism as a predictor of motivations behind Facebook profile picture selection. Cyberpsychology, Behavior, and Social Networking, 16(1), 14-19. Kilduff, M., & Day, D. V. (1994). Do chameleons get ahead? The effects of self-monitoring on managerial careers. Academy of Management Journal, 37(4), 1047-1060.

Kim, D., Jang, S., & Adler, H. (2015). What drives café customers to spread eWOM? Examining self-relevant value, quality value, and opinion leadership. International Journal of Contemporary Hospitality Management, 27(2), 261-282. Kozinets, R. V., De Valck, K., Wojnicki, A. C., & Wilner, S. J. (2010). Networked narratives: Understanding word-of-mouth marketing in online communities. Journal of marketing, 74(2), 71-89. Krämer, N. C., & Winter, S. (2008). Impression management 2.0: The relationship of selfesteem, extraversion, self-efficacy, and self-presentation within social networking sites. Journal of media psychology, 20(3), 106-116. Kristof-Brown, A., Barrick, M. R., & Franke, M. (2002). Applicant impression management: Dispositional influences and consequences for recruiter perceptions of fit and similarity. Journal of Management, 28(1), 27-46. Kulviwat, S., Bruner, G. C., & Al-Shuridah, O. (2009). The role of social influence on adoption of high tech innovations: The moderating effect of public/private consumption. Journal of Business Research, 62(7), 706-712. Kwon, H. H., Trail, G., & James, J. D. (2007). The mediating role of perceived value: Team identification and purchase intention of team-licensed apparel. Journal of Sport Management, 21(4), 540-554. Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review and two-component model. Psychological bulletin, 107(1), 34.

Lee, D., Kim, H. S., & Kim, J. K. (2012). The role of self-construal in consumers’ electronic

86

word of mouth (eWOM) in social networking sites: A social cognitive approach. Computers in Human Behavior, 28(3), 1054-1062. Lee, Y. H., & Ching Lim, E. A. (2010). When good cheer goes unrequited: how emotional receptivity affects evaluation of expressed emotion. Journal of Marketing Research, 47(6), 1151-1161. Lee, S. H., Noh, S. E., & Kim, H. W. (2013). A mixed methods approach to electronic wordof-mouth in the open-market context. International Journal of Information Management, 33(4), 687-696. Lennox, R. D., & Wolfe, R. N. (1984). Revision of the self-monitoring scale. Malär, L., Krohmer, H., Hoyer, W. D., & Nyffenegger, B. (2011). Emotional brand attachment and brand personality: The relative importance of the actual and the ideal self. Journal of Marketing, 75(4), 35-52. Matta, V., & Frost, R. (2011). Motivations of Electronic Word-of-Mouth Communications by Reviewers: A Proposed Study. Mehdizadeh, S. (2010). Self-presentation 2.0: Narcissism and self-esteem on Facebook. Cyberpsychology, behavior, and social networking, 13(4), 357-364.

Mill, J. (1984). High and low self‐monitoring individuals: Their decoding skills and empathic expression. Journal of Personality, 52(4), 372-388. Muntinga, D. G., Moorman, M., & Smit, E. G. (2011). Introducing COBRAs: Exploring motivations for brand-related social media use. International Journal of advertising, 30(1), 13-46. O'Keeffe, G. S., & Clarke-Pearson, K. (2011). The impact of social media on children, adolescents, and families. Pediatrics, 127(4), 800-804. Ong, E. Y., Ang, R. P., Ho, J. C., Lim, J. C., Goh, D. H., Lee, C. S., & Chua, A. Y. (2011). Narcissism, extraversion and adolescents’ self-presentation on Facebook. Personality and individual differences, 50(2), 180-185. Ratner, R. K., & Kahn, B. E. (2002). The impact of private versus public consumption on variety-seeking behavior. Journal of Consumer Research, 29(2), 246-257. Reynolds-McIlnay, R., & Taran, Z. (2010). Ten of your friends like this: Brand related wordof-mouth on Facebook. Marketing Management Association, 37-42.

87

Richins, M. L. (1983). Negative word-of-mouth by dissatisfied consumers: A pilot study. The journal of marketing, 68-78. Rosenfeld, P., Giacalone, R. A., & Riordan, C. A. (1995). Impression management in organizations: Theory, measurement, practice. Van Nostrand Reinhold. Rudman, L. A. (1998). Self-promotion as a risk factor for women: the costs and benefits of counterstereotypical impression management. Journal of personality and social psychology, 74(3), 629. Schlenker, B. R. (1980). Impression management. Brooks/Cole Publishing Company. Sernovitz, A., Godin, S., & Kawasaki, G. (2006). Word of mouth marketing: How smart companies get people talking. Kaplan Pub. Singh, V., Kumra, S., & Vinnicombe, S. (2002). Gender and impression management: Playing the promotion game. Journal of Business Ethics, 37(1), 77-89.

Snyder, M. (1979). Self-monitoring processes. Advances in experimental social psychology, 12, 85-128. Soetarto, B., Yap, K. B., & Sweeney, J. C. (2009). Electronic Word-of-Mouth: An Exploration into the Why, What, and How. In Australia and New Zealand Marketing Academy Conference (Vol. 17, pp. 4-7). Stevens, C. K., & Kristof, A. L. (1995). Making the right impression: A field study of applicant impression management during job interviews. Journal of applied psychology, 80(5), 587. Stokes, D., & Lomax, W. (2002). Taking control of word of mouth marketing: the case of an entrepreneurial hotelier. Journal of small business and enterprise development, 9(4), 349-357. Sweeney, J. C., Soutar, G. N., & Mazzarol, T. (2005, December). The difference between positive and negative word-of-mouth—emotion as a differentiator. In Proceedings of the ANZMAC 2005 Conference: Broadening the Boundaries (pp. 331-337). Taylor, D. G. (2010). “I speak, therefore I am”: Identity and self-construction as motivation to engage in electronic word of mouth. University of North Texas. Tetlock, P. E., & Manstead, A. S. (1985). Impression management versus intrapsychic explanations in social psychology: A useful dichotomy?. Psychological Review, 92(1), 59. Thomas, L. C. (2012). Think visual. Journal of Web Librarianship, 6(4), 321-324.

88

Thoumrungroje, A. (2014). The influence of social media intensity and EWOM on conspicuous consumption. Procedia-Social and Behavioral Sciences, 148, 7-15. Turnley, W. H., & Bolino, M. C. (2001). Achieving desired images while avoiding undesired images: exploring the role of self-monitoring in impression management. Journal of Applied Psychology, 86(2), 351. Weiss, B., & Feldman, R. S. (2006). Looking good and lying to do it: Deception as an impression management strategy in job interviews. Journal of Applied Social Psychology, 36(4), 1070-1086. Williams, M., & Buttle, F. (2011). The eight pillars of WOM management: Lessons from a multiple case study. Australasian Marketing Journal (AMJ), 19(2), 85-92. Xiao-chai, G., & Bei-lei, C. H. E. N. (2009). Determinants of WOM Effects in Virtual Community: an Empirical Analysis. Journal of Beijing Institute of Technology (Social Sciences Edition), 2, 011. Yang, W., & Mattila, A. S. (2017). The Impact of Status Seeking on Consumers’ Word of Mouth and Product Preference—A Comparison Between Luxury Hospitality Services and Luxury Goods. Journal of Hospitality & Tourism Research, 41(1), 3-22. Yeh, Y. H., & Choi, S. M. (2011). MINI-lovers, maxi-mouths: An investigation of antecedents to eWOM intention among brand community members. Journal of Marketing Communications, 17(3), 145-162. Young, A. L., & Quan-Haase, A. (2009). Information revelation and internet privacy concerns on social network sites: a case study of facebook. In Proceedings of the fourth international conference on Communities and technologies (pp. 265-274). ACM. Zywica, J., & Danowski, J. (2008). The faces of Facebookers: Investigating social enhancement and social compensation hypotheses; predicting Facebook™ and offline popularity from sociability and self‐esteem, and mapping the meanings of popularity with semantic networks. Journal of Computer‐Mediated Communication, 14(1), 1-34.

89

MANAGEMENT SUMMARY

Instagram users are keeping themselves busy with impressing others by showing off purchased products. Customers who talk about products online is one of the most effective forms of marketing (i.e. eWOM). They have the power to influence (potential) fellow customers in marketing related goals as brand awareness, brand preference, purchase intention

90

and loyalty (Buttle, 1998; Cheung and Thadani, 2012; Cheung, Lee & Rabjohn, 2008; Gruen, Osmonbekov & Czaplewski, 2006). Especially luxury products seem to be suitable for impression management activities, which is an interesting fact for luxury brands. But what drives consumers to talk positively about a luxury brand via Instagram? This study adds value by finding out if self-monitoring and impression management are drivers of eWOM. High self-monitoring individuals are better in adjusting their behaviour according to the social situation compared to low self-monitoring people. Preferred behaviour in the context of social media is presenting an outgoing and extravert self (Hall and Pennington, 2013; Zywica and Danowski, 2008). An opportunity to present an outgoing self on Instagram is to engage in eWOM related activities as liking, sharing and posting. Hereby, it is expected that impression management acts as the underlying mechanism (moderator). Apparently, it is important to enhance and promote ‘the self’ by showing other consumers that you are a person of product expertise, good taste and great influence (Chu and Kim, 2011; HenningThurau et. al, 2004; Kim, Jang & Adler, 2015; Matta and Frost, 2011; Taylor, 2010; Xiaochai and Bei-lei, 2009). This study aims to find out if someone’s relatively high level of self-monitoring drives the need for impressing other people, which leads eventually to more actively eWOM about luxury products via Instagram. Hereby, the distinction is made between privately and publicly used luxury products. According to prior research in the offline field, people talk more about products that are used in public situations. Testing this difference in an online environment as Instagram brings relatively new insights to the light. Data was generated using an online questionnaire targeting women with an Instagram account. A total of 121 volunteers participated in the study. Data was analysed by using linear regression analyses and PROCESS for mediation. All types of Instagram related eWOM activities (like, share, post) were tested for both privately and publicly used luxury products.

91

Hypotheses were not confirmed, so self-monitoring has no direct effect on eWOM activities and this potential relationship is also not mediated by impression management. Exploratory analyses provided explanations and useful insights for luxury brand marketers. A significant and positive relationship was found between the mediator impression management and eWOM. This means that impression management is driven on its own or by a trait different from self-monitoring. As expected, effects came out stronger for products that are used in public situations (e.g. Gucci handbag). This difference should only be interpret as remarkable since it was not confirmed with signification. The results help marketers understand what triggers customers to talk about luxury products via Instagram. Customers are sensitive to opportunities whereby they can impress followers with for instance showing off luxury products. A suggested strategy is to approach influencing Instagram users and reward them to talk about your luxury products. It is important to keep the difference between privately and publicly used products in mind when setting out these strategies. It is expected that stimulating eWOM about a luxury car (public product) is more effective than triggering eWOM about a luxury coffee machine (private product).

Related Documents


More Documents from "Beatriz Cerqueira"

Kamp.pdf
June 2020 9
2.pdf
July 2020 9
4.pdf
July 2020 13
Popo.pdf
June 2020 12
17.pdf
June 2020 8