The Effects of Online Uncivil Comments on Vicarious shame and Coping Strategies: Focusing on the Power of Social Identity and Social Recommendation

Abstract

Based on an online experiment, this research examined how uncivil expressions made by participants from the same political partisan group (in-group) influenced the emotional and behavioral intentions of other in-group members, especially when the incivility was supported by social recommendations such as "recommendations." As predicted, results showed that a higher level of vicarious shame was felt when participants perceived higher levels of incivility. However, no significant effects of social recommendations were found regarding levels of vicarious shame. That is, the level of shame was not significantly different between participants who were exposed to an in-group uncivil comment that received recommendations and participants who were exposed to in-group uncivil comment without recommendations. Findings further found two types of coping strategies -situation-reparation and situation-avoidance - among participants exposed to in-group uncivil comments. Yet no significant effects were found regarding coping strategies in response to the presence of social recommendations. Participants' feelings of shame were positively correlated with both types of coping strategies, supporting findings of previous studies. Implications of this study are further discussed.

1. Introduction

Online incivility, often understood as impolite expressions that societies generally criticize[1], has become a serious social problem in recent years. A majority of Americans, for example, consider online incivility to be a serious problem in the U.S.. Advancements of internet-based technologies, such as social media, are considered to have accelerated the dynamics and infectious influence of incivility, making it ever more complex and uncontrollable[2].

To better understand online incivility, much research has been conducted[3, 4, 5] that focuses on understanding uncivil comments directed toward out-groups whose opinions and social identities differ from those of an individual’s associated social groups[3,5]. Not well-understood, however, is how individuals react when uncivil comments are posted by others who share the same opinions or group identity(in-group). Specifically, not yet known is the emotional reactions as well as behavioral intentions, in group uncivil comments evoke among other in-group members during online discussions. Therefore, current study explored how uncivil comments made by in-group members might lead to emotional and behavioral intentions on the part of other in-group members.

In doing so, this study furhter examined the moderating effects of social recommendations(e.g., “like”) made by others in support of in-group incivility: in other words, how individuals were affected when others supported in-group uncivil comments. By considering the effect of the social recommendations, this study attempts to further understand how the incivility interacts with the power of audience to affect people’s emotion and behavior intentions. To address these questions, an experiment was conducted and the findings are discussed below.

2. Theoretical Discussion

Social identity is defined as ‘‘part of an individual’s self-concept which derives from his knowledge of his membership of a social group (or groups) together with the value and emotional significance attached to that membership’’ [6, p. 63]. The strong connection between self and group membership leads to the belief that a good social image of the in-group is necessary to preserve a positive group image[7]. As a result, individuals are heavily invested in maintaining a good social image of their group and, as a result, even disapprove when other in-group members engage in behavior that conflicts with the social norms and expectations[8].

Research has found that individuals feel ashamed upon witnessing undesirable behaviors exhibited by other in-group members, a type of shame based on “group identity” [9,10]. Referred to as vicarious shame, this type of shame is found to occur when individuals believe that the deviant behavior of another group member threatens an aspect of their shared identity[10]. When individuals experience vicarious shame due to disreputable behaviors by in-group members, they will often use different behavioral strategies to cope with their own feelings of shame[9, 10, 11, 12].

As documented, vicarious shame is associated primarily with self-defensive coping strategies, such as avoiding or covering up shame-inducing situations or distancing [9, 10, 13]. For example, a person may attempt to leave the site where misdeeds are exhibited. Vicarious shame may also be related to pro-social coping strategies, such as making reparations or repairs for damage caused by other in-group members[11, 12, 14]. For instance, a person may apologize to victims for the misdeeds that another in-group member committed in an effort to resolve the situation.

Incivility, considered to be outside general social norms [1], is a type of behavior that most people do not like to be associated with. While there is no agreed upon definition of what constitutes incivility, in online environments, incivility is commonly understood as a manner of expression that members of society consider unacceptable (i.e., cursing, pejorative expressions, etc.)[1]. Since incivility deviates from social norms, it is reasonable to expect that feelings of shame and coping strategies are likely to arise among those who are exposed to uncivil comments made by in-group members.

As an emotion, shame depends on how individuals consider they are perceived by others. In that sense, the presence of others can be expected to heighten the tendency to feel shame. Studies have shown that embarrassment following a norm transgression is much more pronounced when an audience is present, consisting of a large number of strangers or new acquaintances[14,15]. In short, the presence of others(an audience) is regarded as a key element for understanding the likelihood that someone will feel vicarious shame or engage in coping strategies.

Prior to the advent of online media, the power of “others” was experienced through the presentation of opinion climates employed by traditional media. In online communication settings, social endorsements, such as “recommend, ” help users to perceive each other’s opinions [16]. To express agreement and/or disagreement, other users simply click an icon, so that others are immediately aware of the level of opinion climate. As a result, users often scan online social environments to know how others feel about certain ideas[16].

From the researchers’ perspective, social recommendations offer a wealth of opportunities to examine the power of online bystanders who participate in internet-based discussions. In other words, the presence of social endorsements “recommendations” may influence the extent to which online users feel shame when in-group incivility is supported by other in-group members or, in the alternative, engage in coping strategies. Therefore, this study raised the following research questions and hypotheses:

H1: There will be a positive relationship between individuals’ level of vicarious shame and their perceptions of the level of incivility.

H2: The level of shame will differ between individuals who are exposed to an in-group uncivil comment that receives recommendations compared to situations where individuals are exposed to an in-group uncivil comment without recommendations.

RQ1: Will two different kinds of coping strategies occur when individuals are exposed to in-group uncivil comments? 

RQ2: Will the level of intentions to engage in coping strategies differ between individuals who are exposed to an in-group uncivil comment that receives recommendations compared with situations where individuals are exposed to an in-group uncivil comment without recommendations? 

H3: Individuals’ feelings of shame will be positively correlated with their intentions to use coping strategies.

3. Method

Originally, data were collected using a 2(group identity: in-group vs. out-group) x 2(social recommendation: yes vs. no) mixed-design online experiment where three news articles on gun control, abortion, and climate change were presented participants to read. For the current study, only two groups’ ((1) in-group uncivil comment with recommendations and (2) in-group uncivil comment without recommendations) responses on abortion were examined.

An online link to the study was sent to participants via online survey software by Qualtrics. Participants were first asked to indicate their political partisanship. Based on their responses, participants were randomly assigned to one of four conditions and asked to read a news article written in a neutral tone on the topic of abortion. The abortion issue was chosen for this study as a subject is known to receive numerous uncivil comments online[17] due to the highly polarized nature of the topic [18].

(Figure 1) The Experimental Stimuli

3.1 Participant

A total of 191 Participants over 18 years of age were recruited during a month-long period in 2015 via Mechanical Turk (MTurk), a crowd-sourcing system widely used by social scientists for experimental studies that allows researchers to post surveys for volunteer participants to complete for monetary compensation[5,19]. Table 1 shows the participants’ socio-demographic characteristics and two participants’ responses were missing.

(Table 1) Participants’ Demographics

3.2 Group Assignment

As explained above, participants’ political identity was used to group respondents into four different conditions of the experiment. For this, a partisanship measurement was created by combining items asking respondents’ their partisanship(1=Democrat, 2=Republican, 3=Independent, 4=Others, 5=None) and partisan leanings for those who identified themselves as either independent or none(While you may not identify yourself as a partisan, do you think you lean toward one of the two parties?: 1=Leaning toward the Democratic Party to 7=Leaning toward the Republican Party with 4=Not leaning toward the Democratic or Republican party). The partisan-leaning question was recoded to indicate either Democrat(1-3), Republican(5-7) and the participants who chose option 4 were eliminated from the analysis. Participants were also asked to answer the partisanship of the first manipulated commenter. Then, the answer was matched with participants’ partisanship to indicate either (a) the same as that of the participant (in-group) or b) the groups that differed from the participant (out-group). People who failed to identify the commenter’s partisanship correctly or deny identifying were removed.

3.3 Manipulation

Out of four comments, the initial uncivil comment was manipulated to appear with recommendations(189-201) or no recommendations. The three other civil comments that followed the first uncivil comment received 5 to 60 recommendations. The specific number of recommendations was determined after monitoring the typical number of recommendations made in top-rated online news in response to news articles published over a three-day period in the New York Times, Washington Post, and Chicago Tribune. Participants who did not answer the manipulation check question asking about the volume of recommendations on the first comment were removed from the analysis. Results of the t-test indicated the number of recommendations was successfully manipulated(t=-28.422, df=189, p<.001).

3.4 Measures

Incivility Perception: Three questions were created the study:The comment expressed by Democrats/Republicans (D/R) were inappropriate; The comment expressed by D/R were offensive; The comment expressed by D/R were uncivil. All responses (Strongly disagree(1) to strongly agree(7)) were added to indicate the perception of incivility (M=13.17, SD=5.52, α=.942).

Shame: Eight questions from a study by Rees and colleagues[20] were modified for use in the current study: I feel ashamed because the uncivil expressions made by D/R create a bad image about D/R in the eyes of others; I feel ashamed when I realize that other people might think negatively about D/R because of the uncivil comment; I feel ashamed when I think of how D/R are viewed negatively by other online news readers based on his/her nasty comments; I feel ashamed because the uncivil comment made by D/R may have damaged reputation of our party; The incivility of a D/R toward others whose opinions differ makes me feel somewhat ashamed about what it means to be a D/R; I feel ashamed to be a D/R because of the way a person in our group expressed our general opinions; I feel ashamed for any offense that we in our group may have caused others by posting uncivil expressions; I feel ashamed because uncivil expressions by D/R with regard to the other side are unethical. All responses to eight questions(Strongly disagree(1) to strongly agree(7)) were added to create the shame measurement(M=26.52, SD=14.10, α=.973).

Coping Strategies: Drawing upon research by Johns and colleagues[9], the current study modified nine items to examine coping strategies. Measurements for two coping strategies were created after a principal component analysis (PCA) was conducted. Factors are explained in the result.

Control Variables: The control variables included demographics, comment reading(how often do you read comments on the site?; M=3.50, SD=.99), comment writing (how often do you write comments on the site?; M=2.07, SD=.94), and social recommendation use(how often do you use social recommendation systems?; M=2.63, SD=1.13). Response options were 1=never, 2=rarely, 3=sometimes, 4= quite often, 5=very often.

4. Result

A correlation analysis was conducted to examine the relationship between perception of incivility and the feeling of shame. Individuals felt higher levels of shame as they perceived higher levels of incivility expressed in comments by members of their own in-group (r=.507, p<.000). Thus, H1 was supported. Furthermore, results of the t-test indicated that the level of shame was not significantly different between respondents who were exposed to in-group uncivil comments that received recommendations and participants who were exposed to in-group uncivil comments without recommendations (t= -1.02, p =.307). In sum, results suggest that the presence of in-group’s incivility may be enough to provoke the feeling of shame and that people’s feeling of shame was not necessarily influenced by how much others support the shameful acts. Thus, H2 was not supported.

A PCA with a varimax rotation was conducted to scrutinize the nine items of coping strategies. Due to cross-loadings, an item(I am not sure if I agree with the majority of D/R on this matter) was eliminated and eight items were retained. Table 2 shows that the items reflected the two concepts: 1) situation-reparation(M= 18.78, SD= 8.01, α= .886) and 2) situation-avoidance(M=9.50 SD= 4.93, α= .833). Although two kinds of coping strategies were found, results suggest that people would not necessarily prefer to use ‘pro-social’ ways to repair the situation but rather use both pro-social and dissociation methods. Otherwise, individuals simply preferred to avoid such conflicting situation.

(Table 2) PCA on Coping Strategies

The results of t-tests indicated that the level of intentions to be engaged in coping strategies was not significantly different between respondents who were exposed to in-group uncivil comments with or without recommendations: situation-reparation(t=-.314, p=.754) and situation-avoidance (t=-1.43, p=.155). Even after controlling for demographics and other control variables, individuals’ feelings of shame were positively correlated with both types of intentions to use coping strategies: situation-reparation (r=.787, p<.000) and situation-avoidance (r=.465, p<.000). The results imply that either type of coping strategies will likely to appear when a certain level of shame is felt, or even both to occur. Thus, H3 was supported.

5. Discussion

The research found that the level of vicarious shame increased as the perception of incivility increased. The finding suggest that participants may perceive the uncivil expressions as a factor that can threaten their group’s image. Contrary to expectations, results showed that participants were not highly sensitive to the presence of social recommendations when they experienced vicarious shame. In other words, no statistically significant difference was found in the level of shame between those who were exposed to an in-group uncivil comment that received recommendations and participants who saw no recommendations. Arguably, this particular finding raises two possible explanations. On one hand, it may be that individuals’ emotions were not affected by other in-group members’ incivility to begin with so that the presence of recommendations did not affect their level of shame. On the other hand, the influence of others may have been ever present in users’ imaginations, even in a condition where an audience was not explicitly displayed. This research question is open to examination by future researchers.

Unlike previous study’s finding[9] that identified two coping strategies – pro-social and self-defensive – the current study’s factor analyses found two kinds of coping strategies that reflect different concepts: situation-reparation and situation-avoidance. Again, the presence of social recommendations did not influence participants’ intentions to be engaged in the two coping strategies. It seems that the first factor, situation-reparation, reflected the participants’ desire to make amends for another in-group member’s incivility, though not all pro-socially, while the second factor reflected the participants’ desire to avoid a shameful situation. How participants selected one strategy over the other is unknown. It is possible that certain appraisals, such as controllability of the situation and an individual’s level of responsibility, may have interacted with one of the coping strategies. Future studies may consider exploring how these conditions influence behavioral intentions.

The results clearly indicate that feelings of shame were highly correlated with participants’ intentions to adopt both types of coping strategies. Yet, participants’ feelings of shame had a stronger positive relationship with their intentions to repair rather than to avoid the situation. In sum, strong correlations between shame and both coping strategies imply people’s strong desire to deal with the feeling of shame, preferably by repairing the situation whether the method may be pro-social or not.

6. Conclusion

As with any experimental study, this research is weak on its external validity and generalizability. Especially, scholars like Sobieraj and Berry[21] and many others argue that levels and types of incivility in online comments are extremely diverse and need to be further explored and examined. Another limitation concerns the use of measurements, adopted from previous studies, to ascertain feelings of shame and coping strategies. Although modified for the current study, those measures were originally developed for offline communication settings. Future studies may consider developing measurements for shame and coping strategies that specifically apply to incivility in online communication environments in order to provide a more nuanced understanding of how people feel and react to online incivility.

Despite its limitations, this study helps to provide a preliminary results for those who pursue research on in-group incivility and its influence on other in-group members. In particular, the current study serves as one of the first to test the effects of audience (via recommendations) and provides a preliminary analysis of how an audience affects individuals’ emotional and behavioral intentional responses regarding in-group incivility online. As societies experience greater increases in online incivility, the findings of this research may be particularly relevant to efforts to promote civil discourse.