Computers in Human Behavior 76 (2017) 19e25

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Full length article Socioeconomic status and : The mediating role of belief in a just world

** * Yubo Hou a, , 1, Tonglin Jiang a, c, 1, Qi Wang b, a Peking University, China b Cornell University, USA c The University of Hong Kong, Hong Kong article info abstract

Article history: This study examined individual factors contributing to online shaming, a recent phenomenon where Received 8 March 2017 people engage in social policing by shaming transgressions using the technology. It focused on Received in revised form the effects of socioeconomic status (SES) and the belief in a just world (BJW). A sample of 245 city 9 June 2017 employees in Nanjing, China participated in the study. Participants read an online post about a real-life Accepted 1 July 2017 transgression and reported their willingness to engage in online shaming. Their subjective SES and BJW Available online 4 July 2017 were assessed. Participants with higher SES exhibited a greater tendency to engage in online shaming than those with lower SES, and participants with stronger BJW were more likely to engage in online Keywords: fi Online shaming shaming. BJW further mediated the relationship between SES and online shaming. The ndings have Socioeconomic status important practical implications in the Internet era. Belief in a just world © 2017 Elsevier Ltd. All rights reserved. Internet

1. Socioeconomic status and online shaming: the mediating orders often occur as a result (Cheung, 2014; Ronson, 2016). Given role of belief in a just world the personal and social impact of online shaming, it is important to ask who are the anonymous “Internet police” and why they The fast development of Web-based technology has greatly conduct online shaming. Our study intended to address these changed people's lives. In particular, Internet has become a plat- questions by examining the effects of subjective socioeconomic form for people to share information, voice viewpoint, and partic- status and belief in a just world. ipate in civic discourse (Shah, Cho, Eveland, & Kwak, 2005). On the other hand, online shaming, where people engage in social policing by shaming transgressions using the Internet technology, has 1.1. The dark side of online shaming quickly proliferated (Cheung, 2014; Wall & Williams, 2007). fi “ Internet has become a battlefield for people to call for punishment Shaming is generally de ned as a process by which citizens for those who have violated social norms. Across many countries, publicly and self-consciously draw attention to the bad dispositions the past decade has witnessed increasing online shaming, where or actions of an offender, as a way of punishing him or her for ” & numerous angry and abusive comments swarm into the targets' having those dispositions or engaging in those actions (Kahan fl personal webpages for the purpose of humiliation and social Posner, 1999). It re ects vigilantism in which people exercise so- condemnation. Serious disruptions of personal lives and social cial control when an established order is under threat from a transgression, a potential transgression, or an imputed trans- gression against institutionalized norms (Johnston, 1996). Thus, * Corresponding author. Department of Human Development, Cornell University, people publicly others to defend their social norms (Cheung, Ithaca, NY 14853, USA. 2014). According to the rational choice theory, perceived external ** Corresponding author. School of Psychological and Cognitive Sciences and sanctions often serve as a component in the cost evaluation of of- Beijing Key Laboratory of Behavior and Mental Health, Peking University, Beijing fenses and are therefore effective in deterring the offenses (Nagin & 100871, China. Paternoster, 1993). Previous research has shown that expected E-mail addresses: [email protected] (Y. Hou), [email protected] fi (Q. Wang). shame states can signi cantly inhibit the offending motivations 1 These authors contributed equally to this work. (Bachman, Paternoster, & Ward, 1992; Grasmick & Bursik; Jr, 1990; http://dx.doi.org/10.1016/j.chb.2017.07.003 0747-5632/© 2017 Elsevier Ltd. All rights reserved. 20 Y. Hou et al. / Computers in Human Behavior 76 (2017) 19e25

Nagin & Paternoster, 1993) and compel individuals to improve individuals use the Internet to express their viewpoints, they behaviors so as to avoid shame in the future (Lindsay-Hartz, De intend to reinforce social norms and promote justice, and they are Rivera, & Mascolo, 1995). empowered to act against social infractions through public The use of shaming as a way to deter deviant behaviors has been shaming (Cheung, 2014). observed since the first day of the cyber world (Goldman, 2015; Another motive of online shaming is to exercise , Skoric, Chua, Liew, Wong, & Yeo, 2010). Online shaming can be with the intention of bringing forth group solidarity and deterring considered to be a type of vigilantism where people exert external deviance (Johnston, 1996). Research has shown that informal social sanctions via Internet (Cheung, 2014; Wall & Williams, 2007). control motivates online shaming. For example, Rafferty (2011) People who conduct online shaming wish to reinforce social norms found that the purpose for people to use cyber sanctioning was and keep society operating in an orderly fashion (Wehmhoener, to exert control in penalizing transgressions so as to prevent similar 2010). Indeed, online exposure of perpetrators has boosted behaviors in the future. The motivation for social control distin- changes in institutional transparency, facilitated prosecution of guishes online shaming from other forms of online aggression such corruption and abuse of power by politicians or local authorities, as cyber-, although they can all produce psychological and contributed to political reform and democracy (e.g., Aman & harms to the targets (Smallridge, Wagner, & Crowl, 2016). Jayroe, 2013). However, online shaming, and shaming more Notably, individual participation in crime prevention activities is generally, has its dark side and can in fact disrupt social control and often voluntary (Rosenbaum, 1987). As a type of vigilantism, online make things more anarchic (Laidlaw, 2016; Solove, 2007). shaming shares the feature of autonomy, where individuals Furthermore, with the Internet-empowered rapid and broad in- participate autonomously and spontaneously, free from state sup- formation transmission, the negative effect of online shaming can port or other organizational input (Johnston,1996; Smallridge et al., be particularly pervasive. 2016). They voluntarily comment on or repost about the trans- Specifically, the use of the Internet to call for others to punish gression in order to exercise social control and restore justice. social infractions can be far-reaching, without geographical or However, individuals may be variably motivated to engage in on- locational constraints, and take place with great immediacy. Online line shaming. shaming often involves outburst of anger and aggression, exposing and sharing the transgression post, using or threatening the use of 1.3. Socioeconomic status and online shaming force, and so forth (Laidlaw, 2016; Wehmhoener, 2010). Further- more, online shaming often comes with the exposure of personal Socioeconomic status (SES) refers to individuals' relative social identifiable information of the offenders for the purpose of hu- rank generated by wealth inequality (Greitemeyer & Sagioglou, miliation, social condemnation and punishment (Cheung, 2014). 2016). SES is not simply how much one earns; it is also how one Empowered by the Internet, these online aggressive behaviors can perceives him- or herself in comparison with others, with the two easily escalate into a form of online mob trial and further extend to indexes being moderately correlated (Adler, Epel, Castellazzo, & offline (Cheung, 2014; Solove, 2007). As a result, people Ickovics, 2000; Johnson & Krueger, 2006; Kraus, Piff, & Keltner, who initially violate social norms may later turn into the victims of 2009). Research has shown that subjective SES is more important online shaming. than objective indicators of SES for psychological functioning. Even Online anonymity that keeps people safe from identification and after accounting for people's objective social standing, higher penalty for their aggressive actions further contributes to the subjective SES predicts better psychological well-being and greater negative effect of online shaming. Anonymity often brings about physical health (Adler et al., 2000; Boyce, Brown, & Moore, 2010; deindividuation (Zimbardo, 1969), which may reduce people's Cohen et al., 2008; Kraus & Stephens, 2012). ability to regulate their behavior and make them engage in abusive Importantly, SES not only determines the material resources one actions. Deindividuation may also make people react based on their has access to, but also shapes individuals' psychological patterns. In current emotional states and decrease their awareness of others' lower SES environments where resource is scarce, individuals often thoughts and feelings (McKenna & Bargh, 2000; Zimbardo, 1969). do not lend themselves to personal influence, choice, or control and Research has observed greater hostility and aggressive behaviors in tend not to believe in their own efficacy or their abilities in social computer-mediated interactions than face-to-face interactions influence (Gurin, Gurin, & Morrison, 1978; Kraus & Stephens, 2012; (Dubrovsky, Kiesler, & Sethna, 1991; Siegel, Dubrovsky, Kiesler, & Lachman & Weaver, 1998). In contrast, the abundant material re- McGuire, 1986). Moreover, the perceived online anonymity may sources often enable people in higher SES groups to experience make people feel free from traditionally constraining pressures of themselves with distinct agency, which makes them more willing social norms, conscience, morality, and ethics to behave in a to influence the world according to their personal preferences normative manner. This, in turn, makes online aggression particu- (Kraus & Stephens, 2012; Kraus, Piff, Mendoza-Denton, Rheinsch- larly pervasive and contributes to a mob-like mentality to correct midt, & Keltner, 2012; Kraus et al., 2009). Criminology research has the actions of the wayward and avenge those who have done wrong shown that people with higher SES generally exhibit stronger in- (Wehmhoener, 2010). tentions to intervene and exert informal social control and that they feel more responsible for deterring the offense, compared with 1.2. Motives for online shaming those with lower SES (Greenberg, Rohe, & Williams, 1982; Rosenbaum, 1987; Taylor, Gottfredson, & Brower, 1981). People who engage in online shaming usually do not consider it In general, voluntary social participation is greater among in- to be shaming. In fact, they often see it as a contribution to society dividuals who occupy more dominant social positions and roles, (Skoric et al., 2010). The two main motives for justice, the use of either ascribed or achieved (Smith, 1994). For example, people who punishment, and behavior control are applicable to online shaming devote their time to volunteer works are mainly from higher SES (Oswald, Hupfeld, Klug, & Gabriel, 2002; Tyler & Boeckmann, 1997; groups, with a relatively high income, good functional health, and Wenzel & Thielmann, 2006). Online shaming may help to rees- good education (Auslander & Litwin, 1988; Edwards & White, 1980; tablish social norms disturbed by the offender, and further deter Hodgkinson, Weitzman, Noga, & Gorski, 1992; Smith, 1994; Wilson the offender as well as others from committing the offense again. & Musick, 1997). Given that people with higher SES have a higher Thus, one motive for online shaming is to ensure conformity to level of perceived control and autonomy, they may be more in- social norms through self-regulation (Wehmhoener, 2010). When clined to engage in the voluntary activity of online shaming than Y. Hou et al. / Computers in Human Behavior 76 (2017) 19e25 21 those with lower SES. age ¼ 29.52 years; SD ¼ 8.35) from the city of Nanjing, Jiangsu Province, China. They provided informed consent and were each 1.4. The effect of belief in a just world paid RMB10 for their participation. The participants were from a variety of professions, including laborers (7.4%), office clerks (18%), Belief in a just world (BJW) refers to the mental state in which service professionals (22.1%), entrepreneurs (2.5%), technicians people need to believe in a just world in order to maintain their (24.2%), civil servants (17.2%), and self-employed (4.5%), and a few subjective well-being and navigate in the complex social world who were temporally out of job (4.1%). Approximately half of the (Lerner & Miller, 1978; Lerner, 1980, pp. 9e30). As a ‘‘positive illu- participants had a monthly income lower than the average income sion’’ towards the world (Lipkus, Dalbert, & Siegler, 1996; Taylor & in Nanjing (i.e., RMB 3457; http://www.njtj.gov.cn/47448/47488/) Brown, 1988), BJW is a double-edged sword. On the one hand, it can and the other half had an above-average income. Out of the 243 protect individuals from the harsh realities of the real world and participants who reported information about their education, 77% thus promote mental health and subjective well-being (Dalbert, had a college education or beyond, 20% completed high school or 1999, 2009; Furnham, 2003; Lerner, 1980, pp. 9e30). On the professional school training, and 3% finished elementary school or other hand, to preserve their just-world beliefs, individuals tend to middle school. The participants reported surfing on the Internet for blame and derogate those who disrupt the social order (Hafer, an average of 2e3 h a day, browsing the News, watching videos, or Begue, Choma, & Dempsey, 2005). chatting with others. Lipkus et al. (1996) propose that there are two dimensions of BJW: the personal belief in just world (PBJW) and the general belief 2.2. Measures in just world (GBJW). PBJW refers to the belief that one's life events are fair, whereas GBJW refers to the belief that the world is 2.2.1. Online shaming generally fair. BJW is closely related to pro-social behaviors, with Research has shown that online shaming is expressed in a va- PBJW and GBJW playing different roles (Dalbert, 1999, 2009; Sutton riety of ways, including expressing anger, making abusive/negative & Winnard, 2007). When GBJW is threatened, people tend to take comments, reposting the transgression post, and following the actions such as to humiliate or derogate others to protect their just updates of the event (Laidlaw, 2016; Wehmhoener, 2010). Abusive beliefs, which may make them antagonistic towards victims. comments are difficult to measure and quantify given people's Studies have shown that GBJW predicts indifference and harsh presentational concerns. Therefore, in our study, online shaming attitudes towards social miseries. For example, people with stron- was captured by expression of anger, repost the transgression post ger GBJW show harsher attitudes towards refugees (Khera, Harvey, or like others' hostile comments, and willingness to follow the & Callan, 2014) and more prejudice toward the elderly and the poor updates. & (Sutton Douglas, 2005), and they exhibit less altruistic behavior We used a scenario test to capture the participants' tendency of  & (Begue, Charmoillaux, Cochet, Cury, De Suremain, 2008). online shaming in a controlled experiment. We adopted an online Unlike GBJW, PBJW is positively associated with various indexes post of a real-life transgression and relevant hostile comments from of psychological adjustment, such as greater life satisfaction Weibo, a social media site in China equivalent to . In this & (Dalbert, 1999), lower depression (Dzuka Dalbert, 2002), and transgression event, a BMW driver illegally parked her car and  & greater purpose in life (Begue Bastounis, 2003). As a valuable verbally abused the policeman who came to solve the traffic acci- psychological capital, PBJW not only protects individuals from dent caused by her. After reading the post and comments, partici- sufferings but also motivates them to behave benevolently. For pants were asked to report on 7-point Likert Scales their anger,  example, Begue et al. (2008) found that the more people believed in their tendency to repost the transgression post or like others' a just world for themselves, the more they gave to a street beggar. hostile comments, and their willingness to keep abreast of the In summary, past research has provided evidence that strong event. Participants also answered questions about the reasons general just-world belief is positively associated with punitive and behind their responses. harsh attitudes towards others, whereas strong personal just-world belief promotes voluntary prosocial behaviors. Thus, we expected that both forms of just-world beliefs would motivate online 2.2.2. SES shaming, either through the intention of punishment or benevo- The most widely used index of social class rank is the MacArthur lence, for the purpose of restoring justice and social order. Scale of Subjective Socioeconomic Status (Adler et al., 2000). In this Furthermore, we expected just-world beliefs to mediate the effect measure, participants rank themselves in society or in their local of SES in predicting online shaming. Higher SES individuals, who community on a ladder with 10 rungs representing ascending are more concerned about individual rights and social justice levels of income, education, and occupation (Adler et al., 2000; (Greenberg et al., 1982; Kraus et al., 2012; Smith, 1994), would Goodman et al., 2001). Researchers in China have found that a so- exhibit stronger just-world beliefs, which would in turn make them cial hierarchy of 5 levels is more appropriate for Chinese society (Lu, fi more inclined to engage in online shaming. PBJW and GBJW would 2004). We therefore made modi cation to the MacArthur Ladder, further manifest for different reasons: As online shaming is closely related to voluntary social responsibility (Skoric et al., 2010), PBJW would mediate the effect of SES on online shaming. On the other hand, given that online shaming is characterized by social condemnation and aggression (Cheung, 2014; Ronson, 2016; Wehmhoener, 2010), GBJW would also mediate the effect of SES on online shaming (see Fig. 1).

2. Methods

2.1. Participants

The participants included 245 employees (111 males; mean Fig. 1. Proposed two-mediator model. 22 Y. Hou et al. / Computers in Human Behavior 76 (2017) 19e25 asking participants to choose their SES from the bottom level 1 to the top level 5.

2.2.3. Belief in a just world The Scale of Belief in a Just World (Dalbert, 1999) consists of 13 items, with 7 items measuring PBJW (e.g., I believe that I usually get what I deserve) and 6 items measuring GBJW (e.g., I think basically the world is a just place). Participants rated the items on 6-point scales from 1 (strongly disagree) to 6 (strongly agree). The coeffi- cient of internal consistency of PBJW and GBJW was 0.88 and 0.80, respectively.

Fig. 2. The mediating roles of PBJW and GBJW in the relationship between SES and * *** 3. Results anger expression. Unstandardized coefficients are displayed. p < 0.05, p < 0.001.

Descriptive statistics and inter-correlations of all measures are shown in Table 1. The participants' anger, willingness to repost or entered as mediators. There was a significant total indirect effect of like the hostile comments, and willingness to keep abreast of the SES on anger, B ¼ 0.18, SE ¼ 0.06, 95% CI [0.0866, 0.3351]. PBJW event were all positively correlated. In other words, individuals significantly mediated the effect of SES on anger, B ¼ 0.13, SE ¼ 0.07, who felt angrier about an online event of social transgression would 95% CI [0.0012, 0.2858]. However, the mediating effect of GBJW was also be more likely to repost or like the hostile comments and to not significant, B ¼ 0.06, SE ¼ 0.05, 95% CI [-0.0290, 0.1970] (see follow the updates on the event. Fig. 2). Furthermore, participants with higher SES tended to repost or Next, we tested the mediation model for the tendency to repost like more the hostile comments and showed more willingness to or like the hostile comments. We regressed repost on PBJW and keep abreast of the event, although they reported feeling no more GBJW, respectively, and found that both PBJW, B ¼ 0.05, SE ¼ 0.01, anger than participants with lower SES. PBJW and GBJW were both t ¼ 3.23, p < 0.01, and GBJW, B ¼ 0.08, SE ¼ 0.01, t ¼ 5.22, p < 0.001, positively correlated with the three measures of online shaming, significantly predicted repost. We further regressed repost on SES indicating that people with stronger beliefs in a just world were and found that SES significantly predicted repost, B ¼ 0.26, more likely to engage in online shaming. In addition, higher SES SE ¼ 0.13, t ¼ 2.05, p ¼ 0.04. Further multiple mediation analysis individuals exhibited stronger PBJW and GBJW. (model 4 in PROCESS) showed a significant total indirect effect of In the following analyses, we tested the mediating roles of PBJW SES on repost, B ¼ 0.12, SE ¼ 0.06, 95% CI [0.0142, 0.2599]. Whereas and GBJW in the relationship between SES and each of the three GBJW significantly mediated the effect of SES on repost, B ¼ 0.18, online shaming variables (i.e. anger, repost/like and follow-up), SE ¼ 0.06, 95% CI [0.0755, 0.3324], the mediating effect of PBJW was respectively. To perform the mediation analyses, we first not significant, B ¼0.06, SE ¼ 0.06, 95% CI [-0.2112, 0.0516] (see regressed PBJW and GBJW on SES, respectively. SES significantly Fig. 3). predicted both PBJW, B ¼ 2.65, SE ¼ 0.53, t ¼ 4.98, p < 0.001, and Finally, we tested the mediation model for the willingness to GBJW, B ¼ 2.03, SE ¼ 0.52, t ¼ 3.90, p < 0.001. follow the event updates. We regressed online follow-up on PBJW Then in the model for anger expression, we regressed anger on and GBJW, respectively, which showed that both PBJW, B ¼ 0.06, PBJW and GBJW, respectively. Results showed that both PBJW, SE ¼ 0.01, t ¼ 3.92, p < 0.001, and GBJW, B ¼ 0.08, SE ¼ 0.01, t ¼ 5.56, B ¼ 0.06, SE ¼ 0.01, t ¼ 4.40, p < 0.001, and GBJW, B ¼ 0.06, SE ¼ 0.01, p < 0.001, significantly predicted follow-up behaviors. We further t ¼ 4.13, p < 0.001, significantly predicted anger towards the online regressed follow-up on SES and found a significant effect, B ¼ 0.25, transgression. We further regressed anger on SES, which showed SE ¼ 0.13, t ¼ 1.98, p ¼ 0.05. Further multiple mediation analysis that SES did not significantly predict anger, B ¼0.02, SE ¼ 0.13, (model 4 in PROCESS) showed a significant total indirect effect of t ¼0.17, p ¼ 0.86. Recent technical literature on mediation has SES on follow-up, B ¼ 0.16, SE ¼ 0.06, 95% CI [0.0491, 0.2008]. GBJW suggested that there need not be a significant effect of the inde- significantly mediated the effect of SES on follow-up, B ¼ 0.17, pendent variable on the dependent variable to establish mediation SE ¼ 0.06, 95% CI [0.0741, 0.3129], but the mediating effect of PBJW (Zhao, Lynch, & Chen, 2010). Hence, although the effect of SES on was not significant, B ¼0.02, SE ¼ 0.06, 95% CI [-0.1473, 0.0916] anger expression was not significant, we went further to test the (see Fig. 4). mediation. A multiple mediation analysis (model 4 in PROCESS) was 4. Discussion conduced with bootstrapping methods to examine the proposed two-mediator model. PBJW and GBJW were simultaneously The emergence of Internet technology and the associated social

Table 1 Means, standard deviations, and bivariate correlations.

MSD1234567

Age 29.52 8.36 1 ** SES 2.17 0.77 0.18 1 *** PBJW 26.70 6.74 0.04 0.30 1 *** *** GBJW 23.48 6.46 0.01 0.24 0.74 1 *** *** Anger 4.46 1.51 0.04 0.01 0.27 0.26 1 * ** *** *** Repost/like 4.01 1.53 0.02 0.13 0.20 0.32 0.69 1 * *** *** *** *** Follow 4.22 1.54 0.03 0.13 0.25 0.34 0.57 0.66 1

Note. SES ¼ socioeconomic status. * ** *** p < 0.05. p < 0.01. p < 0.001. Y. Hou et al. / Computers in Human Behavior 76 (2017) 19e25 23

willing to take social responsibilities and participate in voluntary work than those with lower SES (Smith, 1994). Therefore, when higher SES individuals are confronted with social transgression events, they are more likely to take actions against the transgressor. Our finding is consistent with previous observations that in- dividuals who more frequently contribute on online shaming websites tend to be more socially responsible and have a higher level of openness (Skoric et al., 2010). In addition, online shaming calls for , with the expression of anger and contempt towards the offender. It is essentially a form of aggression. Research has shown that aggres- sion often results from the urge to take control of a situation that & Fig. 3. The mediating roles of PBJW and GBJW in the relationship between SES and elicits distress (Bushman, Baumeister, Phillips, 2001). People in * *** repost behaviors. Unstandardized coefficients are displayed. p < 0.05, p < 0.001. higher SES groups are characterized by a high demand for control and tend to feel that they deserve more than others (Piff, 2014). They are also more likely to endorse punishment over rehabilita- tion for criminal offenders (Kraus & Keltner, 2013). Online shaming takes place in service of gaining control over the transgressions. Hence, people with higher SES engaged more in this form of aggressive behavior than those with lower SES. Notably, our findings showed that SES was significantly associ- ated with the tendency to repost or like the hostile comments and the willingness to follow the event updates, but was not directly related to the expression of anger. This suggests that participants felt equally angry about the online transgression regardless of SES. On the other hand, SES had a significant indirect effect on anger via PBJW, which we discuss in the next section. In addition, when asked to explicitly express how angry they felt about the event, Fig. 4. The mediating roles of PBJW and GBJW in the relationship between SES and participants might restrain their reports due to a social desirability * follow-up on event updates. Unstandardized coefficients are displayed. p < 0.05, effect. This is reflected in the participants' reported reasons for their *** < p 0.001. online shaming behavior, which was mainly to defend against injustice rather than to vent their anger. Nevertheless, those who expressed strong angers about the transgression were also more media have dramatically changed people's lives. Online shaming likely to repost or like the hostile comments and to follow the can be regarded as a new form of social control exercised via updates on the event. Internet. Although it may be intended for reestablishing the dis- rupted social order, its aggressive and hostile tactics to defend so- 4.2. The mediating role of BJW cial norms can bring about detrimental effects (Skoric et al., 2010). By examining online shaming from the perspectives of socioeco- We found that both personal and general beliefs in a just world nomic status and belief in a just world, we aimed to delineate in- were positively associated with all three forms of online shaming. dividual factors that motivate people to engage in online shaming. This is consistent with the theorization that strong beliefs in a just world motivate individuals to maintain the existing social order 4.1. SES effect on online shaming and protect the fairness of the world (Hafer et al., 2005). Further- more, as predicted, PBJW and GBJW mediated the effect of SES on As expected, the SES level was positively correlated with online online shaming, whereby higher SES individuals, in line with their shaming, whereby those with higher SES were more inclined to interest in individual rights and social justice (Greenberg et al., conduct online shaming than those with lower SES. Intuitively, one 1982; Kraus et al., 2012; Smith, 1994), held stronger just-world may think that lower SES individuals would behave more aggres- beliefs, which in turn made them more inclined to engage in on- sively towards transgressions posted online. It is a common line shaming. Interestingly, PBJW and GBJW, given their different perception that the decentralized feature of Internet is conductive characteristics, manifested in different ways. Whereas PBJW for grassroots to make their voice heard and to participate in online mediated the effect of SES on anger expression, GBJW mediated the discourse production and dissemination (Bennett, 2003; Tabbi, effect of SES on the tendency to repost or like the hostile comments 1997; Tang & Yang, 2011). However, in reality, the Internet discur- and the willingness to keep abreast of the event. sive power often resides in those who come from middle and upper PBJW has been found to have positive implications for in- classes (Tang & Yang, 2011; Zhao & Fu, 2010), and Internet access is dividuals' psychological well-being as well as benevolent behavior positively associated with income and education (Goldman, 2015). (Begue et al., 2008; Dalbert, 1999, 2009; Lipkus et al., 1996). GBJW, Indeed, people with lower SES not only have less access to oppor- in contrast, has been associated with antagonistic and harsh atti- tunities to express their viewpoints online, they also have a higher tudes towards those who disrupt the social order (Begue et al., level of tolerance for injustice and tend to accept the existing social 2008; Khera et al., 2014; Sutton & Douglas, 2005). Compared order in which they are at disadvantage (Jost, Banaji, & Nosek, with the other two measures of online shaming, feeling angry 2004; Kraus & Keltner, 2013). about a transgression may be less oriented towards actions of de- Furthermore, online shaming is an autonomous behavior with terring and punishing deviant behaviors. Thus, by expressing anger, the purpose of defending social norms and preventing future higher SES individuals with strong PBJW may gain a sense of con- deviance. People with higher SES tend to have higher income, trol and justice without derogating or harming others. In contrast, better functional health, and higher education. They are more by endorsing abusive comments and following the event updates, 24 Y. Hou et al. / Computers in Human Behavior 76 (2017) 19e25 higher SES individuals with strong GBJW may adopt punitive ac- 317e347. http://dx.doi.org/10.1111/dome.12024. tions to deter transgressions and protect their just beliefs. Auslander, G. K., & Litwin, H. (1988). Sociability and patterns of participation: Im- plications for social service policy. Nonprofit and Voluntary Sector Quarterly, As an emerging form of informal social control, online shaming 17(2), 25e37. is exercised for the purpose of justice and behavioral control. Bachman, R., Paternoster, R., & Ward, S. (1992). The rationality of sexual offending: Informal social control is often considered to be a way of reducing Testing a deterrence/rational choice conception of sexual assault. Law and So- ciety Review, 343e372. http://dx.doi.org/10.2307/3053901. crime and deviance but not being a deviant action itself. However, Begue, L., & Bastounis, M. (2003). Two spheres of belief in justice: Extensive support when it is exercised online, individuals often adopt antisocial and for the bidimensional model of belief in a just world. Journal of Personality, 71(3), 435e463. http://dx.doi.org/10.1111/1467-6494.7103007. abusive behaviors that are vastly different from how they usually  fi Begue, L., Charmoillaux, M., Cochet, J., Cury, C., & De Suremain, F. (2008). Altruistic act in the real world (Solove, 2007). Our ndings further suggest behavior and the bidimensional just world belief. The American Journal of Psy- that the dilemma of online shaming between “good intention” and chology,47e56. http://dx.doi.org/10.2307/20445443. “bad behavior” may indeed be underlined by different mechanisms Bennett, W. L. (2003). Power: The Internet and Global Activism. In N. Couldry, & J. Curran (Eds.), Contesting media Power: Alternative media in a concerning individuals' beliefs in excising benevolence versus networked world (pp. 17e38). Oxford: Rowman & Littlefield. punishment to maintain a just world. Boyce, C. J., Brown, G. D. A., & Moore, S. C. (2010). Money and happiness rank of income, not income, affects life satisfaction. Psychological Science, 21(4), 471e475. http://dx.doi.org/10.1177/0956797610362671. 4.3. Theoretical and practical implications Bushman, B. J., Baumeister, R. F., & Phillips, C. M. (2001). Do people aggress to improve their mood? Catharsis beliefs, affect regulation opportunity, and Although much work has been done on socioeconomic status, aggressive responding. Journal of Personality and Social Psychology, 81(1), 17. http://dx.doi.org/10.1037/0022-3514.81.1.17. many issues have not yet been fully understood in terms of how SES Cheung, A. S. Y. (2014). Revisiting privacy and dignity: Online shaming in the global shapes psychological experiences (Kraus & Stephens, 2012). Studies e-village. Laws, 3(2), 301e326. http://dx.doi.org/10.3390/laws3020301. on the relation between SES and aggression have not always yiel- Cohen, S., Alper, C. M., Doyle, W. J., Adler, N., Treanor, J. J., & Turner, R. B. (2008). & Objective and subjective socioeconomic status and susceptibility to the com- ded consistent results (Greitemeyer Sagioglou, 2016; Kraus, mon cold. Health Psychology, 27(2), 268. http://dx.doi.org/10.1037/0278- Horberg, Goetz, & Keltner, 2011). Our study is the first to examine 6133.27.2.268. the effect of SES on the aggressive behavior of online shaming. The Dalbert, C. (1999). The world is more just for me than Generally: About the personal e findings suggest that the motives behind online shaming are a key belief in a just world Scale's validity. Social Justice Research, 12(2), 79 98. http:// dx.doi.org/10.1023/A:1022091609047. factor to be taken into consideration in order to understand the Dalbert, C. (2009). Belief in a just world. Handbook of Individual Differences in Social impact of SES. Online shaming typically aims at defending social Behavior, 288e297. norms, although it often manifests in the way of aggression. People Dubrovsky, V. J., Kiesler, S., & Sethna, B. N. (1991). The equalization phenomenon: Status effects in computer-mediated and face-to-face decision-making groups. with higher SES exhibit stronger just-world beliefs and greater Human-computer Interaction, 6(2), 119e146. motivation to intervene and, consequently, show greater engage- Dzuka, J., & Dalbert, C. (2002). Mental Health and Personality of Slovak Unemployed ment in online shaming. Intervention programs that are designed Adolescents: The Impact of Belief in a Just World1. Journal of Applied Social Psychology, 32(4), 732e757. http://dx.doi.org/10.1111/j.1559- to mitigate the negative impact of online shaming should focus on 1816.2002.tb00240.x. this population. Edwards, J. N., & White, R. P. (1980). Predictors of social participation: Apparent or Our findings also have implications for the theory of belief in a real? Journal of Voluntary Action Research, 9(1e4), 60e73. Furnham, A. (2003). Belief in a just world: Research progress over the past decade. just world. Unlike previous studies showing effects of PBJW on a Personality and Individual Differences, 34(5), 795e817. http://dx.doi.org/10.1016/ variety of positive outcomes, we found that people with strong S0191-8869(02)00072-7. PBJW, just like those with strong GBJW, were more inclined to Goldman, L. M. (2015). Trending Now: The Use of Social Media Websites in Public Shaming Punishments. American Criminal Law Review, 52,415. engage in online shaming, although they may be less action- Goodman, E., Adler, N. E., Kawachi, I., Frazier, A. L., Huang, B., & Colditz, G. A. (2001). oriented than those with strong GBJW. GBJW, on the other hand, Adolescents' perceptions of social status: Development and evaluation of a new may motivate more harmful behaviors in cyberspace. The different indicator. Pediatrics, 108(2). e31ee31. fi motivations for people with strong PBJW and those with strong Grasmick, H. G., & Bursik, R. J., Jr. (1990). Conscience, signi cant others, and rational choice: Extending the deterrence model. Law and Society Review,837e861. GBJW to engage in online shaming require further investigation. http://dx.doi.org/10.2307/3053861. Shaming as a way of crime prevention has been widely dis- Greenberg, S. W., Rohe, W. M., & Williams, J. R. (1982). Safety in urban neighbor- cussed in criminology and sociology. However, there have been hoods: A comparison of physical characteristics and informal territorial control in high and low crime neighborhoods. Population and Environment, 5(3), very few psychological studies examining its causes, processes, and 141e165. http://dx.doi.org/10.1007/BF01257054. consequences. Furthermore, online shaming is a fast-growing Greitemeyer, T., & Sagioglou, C. (2016). Subjective socioeconomic status causes phenomenon with the emergence of the Internet technology and aggression: A test of the theory of social deprivation. Journal of Personality and Social Psychology, 111(2), 178e194. http://dx.doi.org/10.1037/pspi0000058. it warrants attention from researchers and practitioners. Our study Gurin, P., Gurin, G., & Morrison, B. M. (1978). Personal and ideological aspects of took the first step and revealed the influences of socioeconomic internal and external control. Social Psychology,275e296. http://dx.doi.org/ status and belief in a just world. Further studies are called for to 10.2307/3033581. Hafer, C. L., Begue, L., Choma, B. L., & Dempsey, J. L. (2005). Belief in a just world and examine online shaming and provide empirical basis for inter- commitment to long-term deserved outcomes. Social Justice Research, 18(4), vention programs to reduce or even eliminate its disruptive effects. 429e444. http://dx.doi.org/10.1007/s11211-005-8569-3. Hodgkinson, V. A., Weitzman, M. S., Noga, S. M., & Gorski, H. A. (1992). Giving and volunteering among American teenagers 12 to 17 years of age. Washington, DC: Acknowledgement Independent Sector. Johnson, W., & Krueger, R. F. (2006). How money buys happiness: Genetic and fi This research was supported by a grant from the Chinese Na- environmental processes linking nances and life satisfaction. Journal of Per- sonality and Social Psychology, 90(4), 680. http://dx.doi.org/10.1037/0022- tional Natural Science Foundation (31528014) to Qi Wang and Yubo 3514.90.4.680. Hou. Johnston, L. (1996). What is vigilantism? British Journal of Criminology, 36(2), 220e236. http://dx.doi.org/10.1093/oxfordjournals.bjc.a014083. Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system justification theory: References Accumulated evidence of conscious and unconscious bolstering of the status quo. Political Psychology, 25(6), 881e919. http://dx.doi.org/10.1111/j.1467- Adler, N. E., Epel, E. S., Castellazzo, G., & Ickovics, J. R. (2000). Relationship of sub- 9221.2004.00402.x. jective and objective social status with psychological and physiological func- Kahan, D. M., & Posner, E. A. (1999). Shaming white-collar criminals: A proposal for * tioning: Preliminary data in healthy, White women. Health Psychology, 19(6), reform of the federal sentencing guidelines . The Journal of Law and Economics, 586. http://dx.doi.org/10.1037//0278-6133.19.6.586. 42(S1), 365e392. Aman, M. M., & Jayroe, T. J. (2013). ICT, social media, and the Arab transition to Khera, M. L. K., Harvey, A. J., & Callan, M. J. (2014). Beliefs in a just world, subjective Democracy: From venting to acting. DOMES: Digest of Middle East Studies, 22(2), well-being and attitudes towards refugees among refugee workers. Social Y. Hou et al. / Computers in Human Behavior 76 (2017) 19e25 25

Justice Research, 27(4), 432e443. http://dx.doi.org/10.1007/s11211-014-0220-8. Siegel, J., Dubrovsky, V., Kiesler, S., & McGuire, T. W. (1986). Group processes in Kraus, M. W., Horberg, E. J., Goetz, J. L., & Keltner, D. (2011). Social class rank, threat computer-mediated communication. Organizational Behavior and Human Deci- vigilance, and hostile reactivity. Personality and Social Psychology Bulletin, sion Processes, 37(2), 157e187. http://dx.doi.org/10.1016/0749-5978(86)90050- 0146167211410987. 6. Kraus, M. W., & Keltner, D. (2013). Social class rank, essentialism, and punitive Skoric, M. M., Chua, J. P. E., Liew, M. A., Wong, K. H., & Yeo, P. J. (2010). Online judgment. Journal of Personality and Social Psychology, 105(2), 247. http:// shaming in the Asian Context: Community empowerment or civic vigilantism? dx.doi.org/10.1037/a0032895. Surveillance & Society, 8(2), 181e199. Kraus, M. W., Piff, P. K., & Keltner, D. (2009). Social class, sense of control, and social Smallridge, J., Wagner, P., & Crowl, J. N. (2016). Understanding cyber-vigilantism: A explanation. Journal of Personality and Social Psychology, 97(6), 992. http:// conceptual framework. Journal of Theoretical & Philosophical Criminology, 8(1), dx.doi.org/10.1037/a0016357. 57. Kraus, M. W., Piff, P. K., Mendoza-Denton, R., Rheinschmidt, M. L., & Keltner, D. Smith, D. H. (1994). Determinants of voluntary association participation and vol- (2012). Social class, solipsism, and contextualism: How the rich are different unteering: A literature review. Nonprofit and Voluntary Sector Quarterly, 23(3), from the poor. Psychological Review, 119(3), 546. http://dx.doi.org/10.1037/ 243e263. a0028756. Solove, D. J. (2007). The future of reputation: Gossip, rumor, and privacy on the Kraus, M. W., & Stephens, N. M. (2012). A road map for an emerging psychology of Internet. Yale University Press. social class. Social and Personality Psychology Compass, 6(9), 642e656. http:// Sutton, R. M., & Douglas, K. M. (2005). Justice for all, or just for me? More evidence dx.doi.org/10.1111/j.1751-9004.2012.00453.x. of the importance of the self-other distinction in just-world beliefs. Personality Lachman, M. E., & Weaver, S. L. (1998). The sense of control as a moderator of social and Individual Differences, 39(3), 637e645. http://dx.doi.org/10.1016/ class differences in health and well-being. Journal of Personality and Social j.paid.2005.02.010. Psychology, 74(3), 763. http://dx.doi.org/10.1037/0022-3514.74.3.763. Sutton, R. M., & Winnard, E. J. (2007). Looking ahead through lenses of justice: The Laidlaw, E. B. (2016). Online shaming and the . http://dx.doi.org/ relevance of just-world beliefs to intentions and confidence in the future. British 10.20944/preprints201612.0112.v1. Journal of Social Psychology, 46(3), 649e666. http://dx.doi.org/10.1348/ Lerner, M. J. (1980). The belief in a just world. In the Belief in a just World. US: 014466606X166220. Springer. Tabbi, J. (1997). Reading, writing, hypertext: Democratic politics in the virtual Lerner, M. J., & Miller, D. T. (1978). Just world research and the attribution process: classroom. In D. Porter (Ed.), Internet culture (pp. 233e252). New York: Looking back and ahead. Psychological Bulletin, 85(5), 1030. http://dx.doi.org/ Routledge. 10.1037/0033-2909.85.5.1030. Tang, L., & Yang, P. (2011). Symbolic power and the internet: The power of a ‘horse’. Lindsay-Hartz, J., de Rivera, J., & Mascolo, M. (1995). Differentiating shame and guilt Media, Culture & Society, 33(5), 675e691. http://dx.doi.org/10.1177/ and their effects on motivation. In J. P. Tangney, & K. W. Fischer (Eds.), Self- 0163443711404462. conscious emotions: Shame, guilt, embarrassment, and pride (pp. 274e300). New Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social psychological York: Guilford Press. perspective on mental health. Psychological Bulletin, 103(2), 193. http:// Lipkus, I. M., Dalbert, C., & Siegler, I. C. (1996). The importance of distinguishing the dx.doi.org/10.1037/0033-2909.103.2.193. belief in a just world for self versus for others: Implications for psychological Taylor, R. B., Gottfredson, S. D., & Brower, S. N. (1981). Informal control in the urban well-being. Personality and Social Psychology Bulletin, 22(7), 666e677. residential environment (Final Report). Washington, DC: Community Crime Lu, X. Y. (2004). Research report on Chinese contemporary social class. Beijing: Prevention Division, National Institute of Justice, US Department of Justice. Research report on Chinese contemporary social class (in Chinese). Tyler, T. R., & Boeckmann, R. J. (1997). Three strikes and you are out, but why? The McKenna, K. Y. A., & Bargh, J. A. (2000). Plan 9 from cyberspace: The implications of psychology of public support for punishing rule breakers. Law and Society Re- the Internet for personality and social psychology. Personality and Social Psy- view,237e265. http://dx.doi.org/10.2307/3053926. chology Review, 4(1), 57e75. Wall, D. S., & Williams, M. (2007). Policing diversity in the digital age Maintaining Nagin, D. S., & Paternoster, R. (1993). Enduring individual differences and rational order in virtual communities. Criminology and Criminal Justice, 7(4), 391e415. choice theories of crime. Law and Society Review, 467e496. http://dx.doi.org/10.1177/1748895807082064. Oswald, M. E., Hupfeld, J., Klug, S. C., & Gabriel, U. (2002). Lay-perspectives on Wehmhoener, K. A. (2010). Social norm or social harm: An exploratory study of criminal deviance, goals of punishment, and punitivity. Social Justice Research, . Graduate Thesis. Retrieved from Graduate College at Digital 15(2), 85e98. http://dx.doi.org/10.1023/A:1019928721720. Repository (Accession No. 11572). Piff, P. K. (2014). Wealth and the inflated self class, entitlement, and narcissism. Wenzel, M., & Thielmann, I. (2006). Why we punish in the name of justice: Just Personality and Social Psychology Bulletin, 40(1), 34e43. http://dx.doi.org/ desert versus value restoration and the role of social identity. Social Justice 10.1177/0146167213501699. Research, 19(4), 450e470. http://dx.doi.org/10.1007/s11211-006-0028-2. Rafferty, R. (2011). Motivations Behind Cyber Bullying and Online Aggression: Cyber Wilson, J., & Musick, M. (1997). Who cares? Toward an integrated theory of Sanctions, Dominance, and Trolling Online (Electronic Thesis or Dissertation). volunteer work. American Sociological Review, 694e713. Retrieved from https://etd.ohiolink.edu/. Zhao, Y. Z., & Fu, B. Q. (2010). Analysis of the social strata of Internet discursive Ronson, J. (2016). So you've been publicly shamed. New York: Riverhead Books. power in China mainland. Journal of International Communication, 5,63e70. Rosenbaum, D. P. (1987). The theory and research behind neighborhood watch: Is it Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and a sound fear and crime reduction strategy? Crime & Delinquency, 33(1), truths about mediation analysis. Journal of Consumer Research, 37(2), 197e206. 103e134. Zimbardo, P. (1969). The human choice: Individuation, reason and order versus Shah, D. V., Cho, J., Eveland, W. P., & Kwak, N. (2005). Information and expression in deindividuation, impulse and chaos. In W. J. Arnold, & D. Levine (Eds.), Nebraska a digital age modeling Internet effects on civic participation. Communication Symposium on motivation (Vol. 16, pp. 237e307). Lincoln: University of Research, 32(5), 531e565. Nebraska Press.