Getting rid of the bad ones: The relationship between group identification, deviant derogation, and identity maintenance☆
Introduction
The tendency to favor ingroup members over outgroup members is a pervasive but not necessarily inevitable feature of intergroup relations (e.g., Brewer, 1979, Mullen et al., 1992). In some situations ingroup members display the opposite and favor outgroup members over ingroup members. Research on the ‘black sheep’ effect (Marques & Yzerbyt, 1988) has showed that desirable ingroup members are favored over similar outgroup members, whereas undesirable outgroup members are favored over similar ingroup members. It has been suggested that devaluation might serve to exclude undesirable members from the representation of the ingroup and protect the group’s identity (e.g., Marques & Paez, 1994). Although this prediction has received considerable circumstantial support from research comparing evaluations of desirable and undesirable ingroup and outgroup members, the link between devaluation and identity maintenance has not yet been established empirically. The current research aimed to address this limitation. In two experiments we investigated the impact of the presentation of desirable and undesirable group members on perceptions of the group among participants who differed in their level of identification with the ingroup.
According to social identity theory (Tajfel & Turner, 1986), people’s identities are derived largely from their memberships in different social categories or groups. When a group membership is salient, individuals evaluate themselves less in terms of their unique and idiosyncratic attributes and more in terms of the comparative properties of the group. To this extent, people are thought to have a basic motivation to see their own groups as superior to outgroups on relevant dimensions. Excluding undesirable members from the ingroup thus serves the important function of maintaining a positive and distinctive social identity (Marques & Paez, 1994).
Support for these ideas comes from several studies showing that desirable ingroup members are evaluated more favorably than identical outgroup members, whereas undesirable ingroup members are evaluated less favorably than identical outgroup members—the ‘black sheep’ effect (see Marques & Paez, 1994, for a review). Marques and Paez (1994, p. 38) suggested that devaluation might serve to exclude undesirable members from the representation of the ingroup (see also Eidelman, Silvia, & Beirnet, 2006). In this view, the undesirable target is portrayed in such an extremely negative light that he/she can no longer be seen as a typical ingroup member. A presumed consequence of this psychological exclusion is that the overall perception of the ingroup should be enhanced. This follows from research showing contrast and assimilation effects on group-based judgments following exposure to salient group members. If information about salient group members is included in the representation that is formed when judging the group, the resulting perception of the group will assimilate to that information (e.g., Bless & Schwarz, 1998). Conversely, excluded information produces no assimilation and may be used as a standard against which the group as a whole is judged. Several studies have showed that excluding deviants from the representation of the group has a contrast effect on subsequent group stereotype judgments (e.g., Kunda & Oleson, 1997). Excluding an undesirable member from the ingroup should therefore enhance the perception of the ingroup’s positive stereotypical attributes.
Consistent with this reasoning, Hutchison and Abrams (2003) found that high identifiers expressed a more positive ingroup stereotype after, compared to before, reading about an undesirable ingroup member. High identifiers who read about an undesirable ingroup member also expressed a more positive ingroup stereotype than those presented with a desirable ingroup member, whereas low identifiers’ perceptions of the ingroup were unaffected by the target group members. This suggests that high identifiers may have excluded the undesirable member from their representation of the ingroup. This interpretation is supported further by target evaluation ratings, which showed that high identifiers were more negative than low identifiers towards the undesirable ingroup member (see also Castano, Paladino, Coull, & Yzerbyt, 2002).
Although these studies suggest that high identifiers may respond to the presence of an undesirable ingroup member with a motivation to protect their group’s identity, evidence for the inferred psychological exclusion process remains circumstantial and indirect. Moreover, the findings from which exclusion is inferred are open to alternative explanation. Group members may derogate an undesirable target extremely but still consider him/her to be a member of the ingroup. Indeed, high identifiers in Hutchison and Abrams’s (2003) study may have expressed more dislike for the deviant not necessarily in an attempt to exclude him/her from their representation of the ingroup, but rather because they were more threatened by the undesirable actions of a fellow ingroup member. This threat may also have led high identifiers to enhance their perception of the ingroup’s positive stereotypical attributes by way of compensation. To this extent, the inferred link between devaluation and psychological exclusion remains to be established empirically. The current research aimed to address this limitation.
In two studies we assessed the impact of the presentation of an undesirable group member on group stereotype judgments among participants with varying degrees of identification with the ingroup. If devaluation serves to exclude undesirable members from the representation of the ingroup, as traditional explanations of the black sheep effect have argued (e.g., Marques & Paez, 1994), we would expect the level of identification with the ingroup to impact on stereotype change following presentation of an undesirable ingroup member through evaluations of the undesirable ingroup member. If on the other hand devaluation and stereotype enhancement are unrelated to psychological exclusion, then there is no reason to expect target evaluation ratings to mediate the effect of identification on subsequent ratings of the ingroup. In Study 2, we also examined how an undesirable outgroup member might impact on perceptions of the outgroup stereotype. We reasoned that highly identified participants would be more motivated to exclude an undesirable target from the representation they formed when judging the ingroup than from their representation of the outgroup. This difference, we reasoned, should be reflected in ingroup and outgroup stereotype judgments following presentation of an undesirable ingroup or outgroup member, respectively.
Section snippets
Participants
Sixty-seven psychology students from the University of Kent participated as a course requirement. Fifty-five were female and 12 were male.
Materials and procedure
The study was conducted as part of a research methods lecture. After a brief introduction to the lecture, participants completed four items assessing their level of identification with the group ‘psychology students’ (from Doosje, Ellemers, & Spears, 1995): “I see myself as a psychology student,” “I identify with psychology students as a group,” “I am pleased
Study 2
The results from Study 1 suggest that devaluation might serve to exclude undesirable members from the representation of the ingroup. Study 2 aimed to replicate and extend these findings by also examining how the presentation of an undesirable outgroup member might impact on the perception of the outgroup stereotype. If devaluation indeed excludes undesirable members from the group, then we would expect high identifiers to be more motivated to exclude an undesirable target from the
Group stereotype ratings
We predicted that identification with the ingroup would correlate positively with ingroup stereotype ratings following presentation of an undesirable ingroup member. We tested this prediction using hierarchical regression analyses. In four steps we entered (1) the predictors group (effect-coded −1 = ingroup, 1 = outgroup), target (effect-coded −1 = desirable, 1 = undesirable), time of judgment (effect-coded −1 = pre-manipulation, 1 = post-manipulation) and the standardized group identification scores, (2)
General discussion
Our starting point was the assumption, following traditional explanations of the black sheep effect (e.g., Marques & Paez, 1994), that high identifiers would be motivated to exclude an undesirable member from their representation of the ingroup. We reasoned that excluding an undesirable member from the ingroup would have a contrast effect on subsequent perceptions of the ingroup stereotype. Consistent with this, in both studies we found that ingroup identification had an impact on the ingroup
References (19)
- et al.
Perceived intragroup variability as a function of group status and identification
Journal of Experimental Social Psychology
(1995) - et al.
Dissidence from within: Examining intragroup reactions to attitudinal opposition
Journal of Experimental Social Psychology
(2003) - et al.
The maintenance of entitativity: A subjective group dynamics approach
- et al.
Multiple regression: Testing and interpreting interactions
(1991) - et al.
The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations
Journal of Personality and Social Psychology
(1986) - et al.
Context effects in political judgement: Assimilation and contrast as a function of categorization process
European Journal of Social Psychology
(1998) Ingroup bias in the minimal intergroup situation: A cognitive-motivational analysis
Psychological Bulletin
(1979)- et al.
Protecting the ingroup stereotype: Ingroup identification and the management of deviant ingroup members
British Journal of Social Psychology
(2002) - et al.
Protecting the ingroup: Motivated allocation of cognitive resources in the presence of threatening ingroup members
Group Processes and Intergroup Relations
(2001)
Cited by (46)
When compatriot tourists behave badly: The impact of misbehavior appraisal and outgroup criticism construal
2022, Journal of Destination Marketing and ManagementCitation Excerpt :Low-severity tourist misbehavior will more likely lead to the activation of ingroup bias, which will result in lenient judgments towards the misbehaving compatriot tourist. In some situations, individuals display the opposite by derogating an ingroup member who engages in rule-breaking behavior (Hutchison, Abrams, Gutierrez, & Viki, 2008). An ingroup deviant, or the ‘black sheep’, can be viewed as an atypical ingroup member undermining the legitimacy of the positive social identity (Iyer et al., 2012).
Social media users’ crisis response: A lexical exploration of social media content in an international sport crisis
2021, Public Relations ReviewDoes the group membership shape evaluations on other drivers? The role of symbolic cues in traffic
2019, Transportation Research Part F: Traffic Psychology and BehaviourTime to go! Leaving the group in response to norm-deviations
2017, Journal of Experimental Social PsychologyKicking out the trolls – Antecedents of social exclusion intentions in Facebook groups
2017, Computers in Human BehaviorCitation Excerpt :The effect of deviant (vs. non-deviant) behavior on intentions to exclude the deviate from the SNS group is mediated by perceptions of identity subversion. Next to exclusion, derogation of the deviate is one of the most commonly investigated reactions to norm-deviations shown by in-group members (e.g., Eidelman & Biernat, 2003; Hutchison, Abrams, Gutierrez, & Viki, 2008; Kessler et al., 2010; Levine & Ruback, 1980; Marques, Abrams, & Serôdio, 2001; Pinto et al., 2010). On the one hand, this reaction may be construed as an attempt to (re-)socialize the deviate (e.g., Levine & Moreland, 1994; Nicholls & Rice, 2017).
Threat and reactions to violated expectations in groups: Adding control to the equation
2024, European Review of Social Psychology
- ☆
The research reported in this article was supported by funding from the Economic and Social Research Council (R42200034207). The authors are grateful to Jeff Stone and two anonymous reviewers for their comments on a previous version of the article.