اثرات تعدیلی جنسیت و تعداد دوستان در رابطه بین خودارائه گری و کلمات مصطلح مربوط به برند در فیس بوک

Abstract This study examines the influence of self-presentation on brand-related word-of-mouth (WOM) and the moderating roles of gender and number of Facebook friends on the relationship between self-presentation and brand-related WOM. Data were collected from Facebook users (N = 400) via an online survey. Hierarchical multiple regression analysis revealed that self-presentation is positively related to brand-related WOM. It was also found that gender moderates the relationship between self-presentation and brand-related WOM. Results show that men are more likely than women to post brand-related content on Facebook when they actively engage in self-presentation. The article concludes with a discussion of how the number of Facebook friends negatively moderates the relationship between self-presentation and brand-related WOM.

Introduction Social networking sites (SNSs) are regarded as a powerful tool for word-of-mouth (WOM) because consumers freely create and spread brand-related information through family, friends, colleagues, and other acquaintances (Vollmer & Precourt, 2008). Brand communities on SNSs such as Facebook encourage consumers to take part in social interactions by commenting, liking, or sharing content within their social networks. These interactions play a critical role in WOM because SNS users actively look for advertising content and often disseminate information about products and brands to their social networks (Taylor, Lewin, & Strutton, 2011). Nielsen (2012) recently reported that people spend more time on SNSs than any other website and consumer attitudes toward advertisements on social media are gradually improving (e.g., 26% of people are more likely to pay attention to advertising content that has been posted by their social connections). These findings indicate that the importance of brand-related WOM will increase because not only are SNSs the most popular online activity, but they also provide consumers with an easy way to share information about products and brands (Chu & Kim, 2011). Therefore, a large number of firms spend a high percentage of their marketing costs on social media to interact with customers in order to generate positive WOM (Dholakia & Durham, 2010). However, the effectiveness of SNS marketing remains unclear. Despite the importance of WOM on SNS, there is relatively little research on brand-related WOM. In particular, it has not yet been confirmed what motivates SNS users who passionately generate WOM. To answer this question, our study considers various variables that are regarded as important factors in SNS usage. In many studies about Internet usage and SNS, self-presentation is a key motivation for hosting a personal homepage (Krämer, 2008 and Zhao et al., 2008), and other research shows that there are gender differences in self-presentation on SNSs (Haferkamp, Eimler, Papadakis, & Kruck, 2012). Moreover, prior studies indicate that the number of friends also has a strong influence on outcomes of SNS use, such as subjective-wellbeing (Kim & Lee, 2011) and user perceptions of the host (Tong, Van Der Heide, Langwell, & Walther, 2008). Based on these studies, we predict that self-presentation, gender, and number of friends are closely associated with brand-related WOM. Therefore, we examine how these variables affect brand-related WOM and identify what inspires WOM. Furthermore, this study proposes that self-presentation is positively related to users’ WOM behavior, and the relationship between self-presentation and brand-related WOM is moderated by gender and number of Facebook friends. The results of this study have implications for target marketing and content strategy on SNSs.

4. Results Prior to testing our three hypotheses, we conducted a correlation analysis between the variables. The descriptive statistics and correlation matrix are provided in Table 1. Results show that the number of brand fan pages was positively correlated with Facebook activity (r = 0.30, p < 0.001), self-presentation (r = 0.21, p < 0.001), number of Facebook friends (r = 0.14, p < 0.01), and WOM on Facebook(r = 0.37, p < 0.001). Facebook activity was positively correlated with self-presentation (r = 0.53, p < 0.001), number of Facebook friends(r = 0.35, p < 0.001), and WOM on Facebook (r = 0.20, p < 0.001). Also, self-presentation was positively correlated with number of Facebook friends (r = 0.26, p < 0.001) and WOM on Facebook (r = 0.44, p < 0.001). However, none of the correlations between number of Facebook friends and WOM on Facebook reached significance. From this result, we estimated that there is no significant effect of number of friends on WOM on Facebook. Table 1. Descriptive statistics and correlations among variables. M SD 1 2 3 4 5 1. Word-of-mouth on Facebook 1.79 1.05 1 2. Number of brand fan pages 2.21 .95 .37† 1 3. Age 29.07 5.29 .04 .01 1 4. Facebook activity 3.13 .93 .20† .30† −.29† 1 5. Self-presentation 2.56 .96 .44† .21† −.16† .53† 1 6. Number of Facebook friends 1.45 .78 .05 .14⁎ −.33† .35† .26† ⁎ p < 0.01. † p < 0.001. Table options To test our hypotheses, we conducted hierarchical multiple regression (Table 2). Brand-related WOM was regressed on self-presentation, gender, number of Facebook friends, and two interaction terms (self-presentation × gender, self-presentation × number of Facebook friends) along with control variables (number of brand fan pages, age, Facebook activity). The moderating variables were centered to avoid multi-collinearity. Table 2. Results of hierarchical regression analyses. Step 1 Step 2 Step 3 B B B Number of brand fan pages .366‡ .338‡ .328‡ Age .015 .012 .015 Facebook activity .142⁎ −.091 −.096 Self-presentation .485‡ .477‡ Gendera .082 .084 Number of Facebook friends −.081 −.002 Self-presentation × gender .274† Self-presentation × number of Facebook friends −.209‡ R2 0.152 0.296 0.331 ΔR2 0.144 0.036 a Gender was coded 0 = female; 1 = male. ⁎ p < 0.05. † p < 0.01. ‡ p < 0.001. Table options The first step showed that the explanatory power of the control variables was significant. In addition, considering the number of brand fan pages was significant in all steps, the number of brand fan pages significantly affected brand-related WOM on Facebook, as expected. In the second step, the adjusted R2 increased by 0.144 from the first step. Furthermore, the positive relationship between self-presentation and brand-related WOM was significant (β = 0.485, p < 0.001). Thus, hypothesis 1 is supported. Additionally, age, Facebook activity, gender and the number of Facebook friends were not significant. Therefore, we concluded that active Facebook users are more likely to engage in brand-related WOM. Finally, in the third step, the change in adjusted R2 increased by 0.036. There was still a positive main effect to Facebook activity (β = 0.477, p < 0.001) on brand-related WOM and all of the two-way interaction effects were significant. In other words, gender and the number of Facebook friends moderate the relationships between self-presentation and brand-related WOM. To interpret both moderating roles, we plotted the interaction effects by using procedures based on the work of Aiken and West (1991). As shown in Fig. 1, men’s brand-related WOM was higher than women when self-presentation was high. In other words, men are more likely to engage in brand-related WOM than women when users more actively engage in self-presentation activity. Therefore, hypothesis 2 is supported. As shown in Fig. 2, brand-related WOM among users with a large number of Facebook friends was lower than that of users with a low number of Facebook friends when self-presentation was high. In other words, users with more friends led to the weaker the effect of self-presentation on brand-related WOM. Therefore, hypothesis 3 is also supported. Interpreting interaction for the effect of gender. Fig. 1. Interpreting interaction for the effect of gender. Figure options Interpreting interaction for the effect of number of friends. Fig. 2. Interpreting interaction for the effect of number of friends.