Personality and Social Psychology Bulletin http://psp.sagepub.com Advantaged Group's Emotional Reactions to Intergroup Inequality: The Dynamics of Pride, Guilt, and Sympathy Nicole Syringa Harth, Thomas Kessler and Colin Wayne Leach Pers Soc Psychol Bull 2008; 34; 115 DOI: 10.1177/0146167207309193 The online version of this article can be found at: http://psp.sagepub.com/cgi/content/abstract/34/1/115 Published by: http://www.sagepublications.com On behalf of: Society for Personality and Social Psychology, Inc Additional services and information for Personality and Social Psychology Bulletin can be found at: Email Alerts: http://psp.sagepub.com/cgi/alerts Subscriptions: http://psp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations (this article cites 28 articles hosted on the SAGE Journals Online and HighWire Press platforms): http://psp.sagepub.com/cgi/content/refs/34/1/115 Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Advantaged Group’s Emotional Reactions to Intergroup Inequality: The Dynamics of Pride, Guilt, and Sympathy Nicole Syringa Harth Thomas Kessler Friedrich Schiller University Colin Wayne Leach University of Sussex Three studies establish intergroup inequality to investigate how it is emotionally experienced by the advantaged Studies and examine psychology students’ emotional experience of their unequal job situation with worse-off pedagogy students When inequality is ingroup focused and legitimate, participants experience more pride However, when inequality is ingroup focused and illegitimate, participants experience more guilt Sympathy is increased when inequality is outgroup focused and illegitimate These emotions have particular effects on behavioral tendencies In Study group-based pride predicts greater ingroup favoritism in a resource distribution task, whereas group-based sympathy predicts less ingroup favoritism Study replicates these findings in the context of students’ willingness to let young immigrants take part in a university sport Pride predicts less willingness to let immigrants take part whereas sympathy predicts greater willingness Guilt is a weak predictor of behavioral tendencies in all studies This shows the specificity of emotions experienced about intergroup inequality Keywords: I relative advantage; group-based emotions; pride; guilt; sympathy; behavioral tendencies magine yourself as a person living in an industrialized country Now imagine another person, your age, your gender, living in a developing country How you feel with this picture in your head? Prideful? Guilty? Sympathetic? Given that human societies are structured as systems of group-based hierarchies (Sidanius & Pratto, 1999), some of us belong to groups that have more and some of us belong to groups that have less Research has documented how members of groups that have less may react to their relative deprivation (e.g., Kessler & Mummendey, 2001; for a review, see Simon & Klandermans, 2001) Collective protest, demonstrations, riots—in short, the motivation to challenge inequality—is a consequence of relative deprivation when individuals feel dissatisfied or angry about it Thus, the specific emotions that individuals feel about intergroup inequality seem to have important implications for what they are willing to about it Although research on relative deprivation has made this point for some time, there is now a growing body of evidence that the emotions individuals feel about their group membership play an important role in explaining their willingness to act on behalf of their group or against other groups (e.g., Leach, Iyer, & Pedersen, 2006; Mackie, Devos, & Smith, 2000; for discussions, see E R Smith, 1993; H J Smith & Kessler, 2004) Authors’ Note: This research was supported by a doctoral fellowship to the first author in the International Research Training Group (GRK 622) titled “Conflict and Cooperation Between Social Groups” funded by the German Research Foundation (Deutsche Forschungsgemeinschaft) We thank Christopher Cohrs, Mirjam Dolderer, Ilka Gleibs, Natascha de Hoog, Timo Stich, Vincent Yzerbyt, and two anonymous reviewers for their helpful comments and discussions on an earlier draft Correspondence should be addressed to Nicole Syringa Harth, FriedrichSchiller-Universität, International Graduate College, Wildstrasse 1, D-07743 Jena, Germany; e-mail: n.harth@uni-jena.de PSPB, Vol 34 No 1, January 2008 115-129 DOI: 10.1177/0146167207309193 © 2008 by the Society for Personality and Social Psychology, Inc 115 Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 116 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN Although understanding of the antecedents and consequences of relative deprivation has grown recently, there is sparse knowledge about how people experience belonging to a group that has more than other groups However, just as in the case of relative deprivation, there is good reason to think that how members of groups with more perceive their position will affect how they feel and what they are likely to (Leach, Snider, & Iyer, 2002) For example, recent research has shown that the perception that one’s ingroup has illegitimate advantage is correlated with the unpleasant feeling of group-based guilt about this inequality (Iyer, Leach, & Crosby, 2003, Study 1; Leach et al., 2006; Powell, Branscombe, & Schmitt, 2005; Swim & Miller, 1999) Additionally, quasi-experimental research has shown that group-based guilt is increased when members of real-world groups are made to focus their attention on their ingroup’s preexisting illegitimate advantage (Iyer et al., 2003, Study 2; Powell et al., 2005) This suggests that self-focus and illegitimacy are appraisals of intergroup inequality that lead to group-based guilt However, to our knowledge, no prior research has examined these two dimensions experimentally by creating an intergroup inequality and manipulating its illegitimacy as well as group members’ focus of attention In addition, no prior experimental research has contrasted a self-focus on illegitimate intergroup inequality with other conditions This is important because other conditions of intergroup inequality may elicit different emotions (Leach et al., 2002) For example, Iyer et al (2003, Study 2) found that encouraging European Americans to focus on African Americans’ illegitimate disadvantage increased these European Americans’ sympathy and decreased their guilt Thus, in this article we examine the degree to which pride, guilt, and sympathy are distinct emotions about intergroup inequality We extend previous research by establishing an intergroup inequality and manipulating legitimacy and focus To investigate the potentially distinct behavioral tendencies triggered by pride, guilt, and sympathy, we examine actual ingroup-favoring behavior (Study 2) and the willingness to share resources with an outgroup (Study 3) Focus and Legitimacy: Differentiating Emotions About Intergroup Inequality Individuals become aware of intergroup inequality by comparing one group with another in terms of resources, level of success, or other attributes (Tajfel & Turner, 1986) A downward comparison that establishes one’s ingroup as better off than an outgroup establishes the ingroup’s relative advantage and the outgroup’s relative disadvantage (Guimond & Dambrun, 2002; Leach et al., 2002) Recent theorizing suggests that individuals may experience a variety of emotions about this kind of intergroup comparison and that each emotion will have specific implications for intergroup behavior Combining self-categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) and appraisal theories of emotions (e.g., C A Smith & Ellsworth, 1985), E R Smith (1993) proposed that individuals’ appraisals of their ingroup’s relation to an outgroup determines which emotions they experience These emotions trigger specific behavioral tendencies (e.g., Frijda, Kuipers, & ter Schure, 1989) Based in this general perspective, Leach et al (2002) offered a conceptual model that identified four appraisal dimensions as differentiating between group-based emotions about intergroup inequality In this article, we highlight two of these appraisal dimensions as we think that they are central to the differentiation of emotions among those who benefit from intergroup inequality: focus and legitimacy When combined, focus and legitimacy suggest the conditions under which members of groups that benefit from intergroup inequality will feel pride, guilt, and sympathy Intergroup relations are complex Thus, people usually not focus on intergroup relations in their entirety Appraisal theories of emotion (e.g., C A Smith & Ellsworth, 1985) suggest that individuals first appraise whether a situation affects the self or others This distinction allows individuals to focus their attention on the relevant party—the self or others As a result, focus of attention may guide subsequent appraisals and emotions In the context of an intergroup comparison where the ingroup is better off, a self-focus highlights the ingroup’s relative advantage over another group As detailed next, depending on the legitimacy of the intergroup inequality, this self-focus may result in feelings of guilt or pride (Leach et al., 2002) An other-focus in the case of an intergroup group comparison where the ingroup is better off highlights the outgroup’s relative disadvantage (Leach et al., 2002) Other-focus should lead to the emotion of sympathy where the inequality is illegitimate (Iyer et al., 2003, Study 2) Thus, framing the same intergroup inequality either as ingroup advantage or outgroup disadvantage should lead to different emotions because this framing focuses attention on either the ingroup or the outgroup However, it should also be clear that the legitimacy of the intergroup inequality is an important determinant of the specific emotions felt about intergroup inequality Legitimacy is known within social identity theory (Tajfel & Turner, 1986) as a sociostructural variable that indicates whether the intergroup relation should change According to this view, an illegitimate intergroup inequality makes both better- and worse-off groups aware of alternatives to the existing relation Other approaches to illegitimacy are similar but tend to emphasize the fairness and deservingness of intergroup Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS inequality (for a review, see Sidanius & Pratto, 1999) Thus, where a group’s effort or ability determines their position in an intergroup relation, this position can be said to be legitimate For example, a group that has received excellent education and training but is disadvantaged in job opportunities (compared to a less welleducated group) suffers an illegitimate disadvantage However, if the disadvantaged group made little effort and was poorly educated, it would be legitimate for its members to have worse job opportunities Legitimacy of this sort is an important basis of emotion, whether at the individual (e.g., C A Smith & Ellsworth, 1985) or group (e.g., Kessler & Mummendey, 2001) level Together, the combination of focus and legitimacy differentiate the three group-based emotions of guilt, sympathy, and pride about intergroup inequality Pride Evidence for group-based pride, and its association with appraisal, is limited (for a review, see Leach et al., 2002) For example, Cialdini’s (1976) notion of “basking in reflected glory” implies that individuals feel pride in their ingroup’s success over an outgroup However, none of this work has examined focus or legitimacy appraisal or has directly assessed pride Although individuals evaluate themselves more positively when their ingroup benefits from a legitimate advantage over an outgroup (for a meta-analytic review, see Bettencourt, Dorr, Charlton, & Hume, 2001), it is unclear whether this is self-focused or experienced in terms of pride For example, in a study that told psychology students that their job prospects were better than those of other students, Guimond and Dambrun (2002, Study 2) found this relative advantage to have no effect on positive affect or feelings of satisfaction, although it did lead to more positive evaluation of the ingroup and prejudice toward outgroups (see also H J Smith & Tyler, 1997) Nevertheless, theory and research at the individual level suggest that a self-focused framing of legitimate ingroup advantage should stimulate group-based pride For example, research at the individual level has shown that pride is associated with the achievement of good outcomes for the individual self (e.g., Rodriguez Mosquera, Manstead, & Fischer, 2000; C A Smith & Ellsworth, 1985) that are thought to be deserved Thus, at the group level, focusing on an ingroup’s achievement of a legitimate advantage over an outgroup should lead to group-based pride (Leach et al., 2002) 117 over an outgroup is associated with the feeling of groupbased guilt (Iyer et al., 2003, Study 1; Leach et al., 2006; Powell et al., 2005; Schmitt, Behner, Montada, Müller, & Müller-Fohrbrodt, 2000; Swim & Miller, 1999) When this feeling of guilt is based purely on one’s existence within a group that has illegitimate advantages, it can be referred to as existential guilt (e.g., Hoffman, 1976; Schmitt et al., 2000) Existential guilt is experienced without individuals or their ingroup having responsibility for illegitimate acts against an outgroup Although previous research has examined the role of legitimacy or self-focus on group-based guilt, no prior research has examined their combination In addition, as far as we are aware, all prior research has used quasi-experimental designs where the features of real intergroup inequalities were made more or less salient For example, Miron, Branscombe, and Schmitt (2006, Study 2) manipulated the legitimacy of gender inequality in pay by providing men with research evidence that women were either equal or lower in the abilities required for well-paid jobs When gender inequality was framed as illegitimate, men reported feeling more guilty about it Although this study is one of the only to manipulate the legitimacy of intergroup inequality, it did not address the self-focus also shown to be central to groupbased guilt In addition, like most other studies of guilt, Miron et al did not contrast the conditions that lead to guilt against those that might lead to other emotions Sympathy At the interpersonal level, it has been shown that a focus on others’ suffering motivates prosocial emotions and helping behavior (Batson et al., 2003) Sympathy seems to be the prevailing response to others’ misfortunes (Weiner, 1995) In one of the few studies at the intergroup level, Iyer et al (2003) showed that European Americans’ belief that African Americans suffered discrimination was associated with sympathy as well as support for greater equality However, like most previous studies of emotion about intergroup inequality, participants were well aware of the outgroup’s disadvantage before the study Few studies have manipulated the focus of attention in a more controlled setting, where the basis of the inequality can be specified In addition, less is known about how legitimacy appraisals affect sympathy at the intergroup level Generally, sympathy seems to be stronger when others’ disadvantage is perceived to be illegitimate (e.g., Montada & Schneider, 1989) Overview Existential Guilt Several recent studies show that the perception that one’s real-world ingroup has an illegitimate advantage Three studies were conducted to compare the groupbased emotions of pride, existential guilt, and sympathy in their distinctiveness for reacting to social inequality Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 118 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN Based in the theoretical model of Leach et al (2002) and the existing literature, we predicted that self-focus on a legitimate ingroup advantage would lead to greater feelings of group-based pride in comparison to illegitimate advantage or other-focus We also predicted that self-focus on an illegitimate ingroup advantage would lead to greater feelings of existential guilt whereas other-focus on an outgroup illegitimate disadvantage would lead to greater group-based sympathy The Emotion–Behavior Link Each emotional experiences of intergroup inequality may trigger different behavioral tendencies (Mackie et al., 2000) that either preserve or alter the status relation We expected group-based pride to be associated with ingroup-favoring behavioral tendencies because positive evaluation of the ingroup enhances ingroup favoritism (e.g., Verkuyten & Hagendoorn, 2002) Moreover, research on nationalism indicates that nationalistic pride predicts xenophobia (Cohrs et al., 2004) However, the relationship between guilt and behavioral tendencies is more ambiguous Guilt seems to explain support of abstract goals of compensation, but it does not appear to motivate concrete forms of political action, such as organizing demonstrations (Leach et al., 2006) Therefore, we expected existential guilt to be a relatively weak predictor for the willingness to act to reduce the intergroup inequality In contrast, feelings of sympathy have been found to be a powerful source of helping behavior (Eisenberg, 2003) Thus, in the context of intergroup inequality, we expected sympathy to be associated with behavioral tendencies to help the disadvantaged PILOT STUDY Method Participants and Procedure A pilot study was carried out to ensure that the material would be appropriate to manipulate the appraisals of focus and legitimacy Eighty-one students of the University of Jena took part in this experiment, which consisted of four conditions: ingroup focus (IGF) legitimate, IGF illegitimate, outgroup focus (OGF) legitimate, and OGF illegitimate Written scenarios in the form of fake newspaper articles were used to establish intergroup inequality and to manipulate focus and legitimacy Each version of the article described the job situation for social scientists in Germany Psychology students were told that their job opportunities were better than those of social pedagogy students In the IGF condition the intergroup inequality was framed as an ingroup advantage by stating that psychologists have better job opportunities and an average income of 130% more than social pedagogues (cf Iyer et al., 2003) In the OGF condition the intergroup inequality was framed as an outgroup disadvantage It was said that social pedagogues have worse job opportunities and an average income 70% less than psychologists The manipulation of legitimacy of the inequality referred to quality of education In the legitimate conditions the education of psychologists was described as excellent and the education of social pedagogues as poor, and vice versa in the illegitimate condition Measures A 3-point scale was used to check the focus manipulation (1 = social pedagogues, = both groups, = psychologists) Furthermore, participants had to indicate the perceived fairness of the inequality between psychologists and social pedagogues with three closely related items anchored by a 7-point scale that varied from (very unfair) to (very fair) Results A (focus) × (legitimacy) ANOVA on the focus manipulation check revealed a very large main effect of focus, MOGF = 1.03, MIGF = 2.88, F(1, 77) = 1126.53, p < 001, η²=.936, a very small main effect of legitimacy, MLegitimate = 2.05, MIllegitimate = 1.93, F(1, 77) = 6.79, p < 05, η² = 081, and a nonsignificant interaction, F(1, 77) = 2.89, p = 09 A (focus) × (legitimacy) ANOVA with perceived fairness as the dependent variable revealed that participants in the legitimate condition found the situation to be more fair (M = 4.48) than participants in the illegitimate condition (M = 3.08), F(1, 78) = 30.00, p < 001, η² = 279 Also, participants in the IGF conditions perceived the situation as more fair (M = 4.50) than participants in the OGF conditions (M = 3.05), F(1, 78) = 31.99, p < 001, η² = 290 The interaction between focus and legitimacy was not significant, F < Discussion We appeared to successfully manipulate participants’ focus of attention When an inequality was framed as an ingroup advantage, participants focused more on their ingroup than on the outgroup Although the legitimacy manipulation led participants to focus slightly more on the outgroup in the illegitimate condition, participants tended to focus on both groups irrespective of legitimacy Our manipulation of the legitimacy of the intergroup inequality also appeared successful Participants perceived Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS as more fair the inequality made legitimate by an “excellent” education compared to the inequality made illegitimate by a “poor” education The perceived fairness of the intergroup inequality was also affected by our manipulation of focus Consistent with previous research (e.g., Iyer et al., 2003, Study 2; for a review, see Mikula, 1993), participants appeared to justify the inequality when it was IGF Perceiving intergroup inequality as more fair when IGF seems likely to be a defensive reaction to belonging to an ingroup that is undeservedly advantaged (e.g., Leach et al., 2006; for a review, see Leach et al., 2002) Indeed, people in advantaged positions may benefit from justifying inequality (Sidanius & Pratto, 1999) The implication of this apparent justification is that our manipulations of focus and legitimacy may less easily provoke guilt If an IGF leads participants to view the inequality as more fair, this will undermine the effect of illegitimacy on guilt in the IGF condition Thus, our manipulations may generate a conservative test of the basis of group-based guilt Of course, if IGF leads participants to justify inequality, our manipulations will more easily provoke pride The following studies address this issue STUDY We investigated the group-based emotions of pride, existential guilt, and sympathy among members of an ingroup that is better off than an outgroup To expand on previous studies that tested the separate effects of focus and legitimacy, we investigated the joint effect of these two variables Thus, we used the design validated in the pilot study Method Design and Participants This experiment reproduced exactly the four conditions examined in the pilot study: IGF legitimate, IGF illegitimate, OGF legitimate, and OGF illegitimate Undergraduate psychology students from the Friedrich-Schiller University, Jena (N = 48, 85% females, M age = 21 years, range = 19–28 years) were randomly assigned to one of the four conditions We chose the × design for the sake of completeness, but we are mainly interested in the conditions IGF legitimate, IGF illegitimate, and OGF illegitimate, as these are the conditions for group-based pride, guilt, and sympathy For the OGF legitimate condition, we had no clear prediction and thus investigated this condition in a more exploratory manner A focus on legitimately disadvantaged others may trigger negative feelings, such as disdain (Leach et al., 2002) 119 Procedure The questionnaires were administered after psychology lectures First, participants were asked to read one of the fake newspaper articles pretested in the pilot study They were then asked to complete the measures detailed next Having completed the questionnaires, participants were debriefed and thanked for their participation with a chocolate bar Measures Intergroup inequality and identification We checked our establishment of an intergroup inequality based on a single item “Regarding the job situation, psychologists are relatively advantaged compared to social pedagogues” using a 7-point scale (1 = absolutely not, = absolutely) To check whether participants identified with the ingroup, four items measured identification with psychology students (e.g., “I identify with psychology students,” α = 83) using a 7-point Likert-type scale These items were taken from Doosje, Ellemers, and Spears (1995) Emotions Immediately after reading the fake newspaper article, participants were asked to indicate their emotional state (“When thinking about the described situation I feel ”) using a 9-point scale (1 = not at all, = very intense) Three items measured pride (proud, successful, happy), three items measured guilt (guilty, have a bad conscience, ashamed1), and two items measured sympathy (sympathy, compassion) To explore the effects of the OGF legitimate condition, three items measured disdain (contemn, disdain, disgusted) The order of the emotion items was varied randomly within the experimental conditions Results Preliminary Analyses Recognition of inequality Participants’ perception of intergroup inequality was significantly above the scale midpoint (M = 5.21), t(47) = 10.16, p < 001 In fact, all individuals perceived their ingroup as better off than social pedagogues Hence, the intergroup inequality we created was recognized A (focus) × (legitimacy) ANOVA revealed that the manipulations did not influence psychology students’ perception of the intergroup inequality (F < 1) Identification Identification with psychologists was well above the scale midpoint (M = 5.22), t(47) = 7.10, p < 001 Moreover, there were no significant differences Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 120 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN TABLE 1: Means and Standard Deviations for Study IGF Legitimate Illegitimate Group-based pride Existential guilt Group-based sympathy Group-based disdain IGF Illegitimate OGF Legitimate OGF Illegitimate M SD M SD M SD M SD 5.14 2.25 3.54 2.06 1.59 1.16 1.83 1.16 4.72 2.81 4.08 2.36 1.52 1.69 1.90 1.84 3.08 2.18 5.66 1.28 1.22 1.43 1.56 0.31 4.29 2.44 6.17 2.08 0.94 1.10 1.61 1.31 NOTE: IGF = ingroup focus; OGF = outgroup focus between the experimental conditions in identification (Fs < 2.19), indicating that randomization of participants was successful and that the manipulations had no significant effect on identification Factor analyses of group-based emotions The 11 emotion items were submitted to a principal-axis factor analysis with oblimin rotation A four-factor solution was obtained accounting for 60.31% of the common variance The first factor was indicated by the disdain items, the second factor was indicated by the sympathy items, the third factor was indicated by the pride items, and the fourth factor was indicated by the guilt items We aggregated the items of each factor into four emotion scales, which showed good internal consistencies: disdain (α = 82), sympathy (α = 83), pride (α = 70), and guilt (α = 76) Sympathy correlated with guilt (r = 26, p = 07), but not with disdain (r = –.12, p = 41) or pride (r = 07, p = 64) Disdain correlated with pride (r = 276, p = 06) and guilt (r = 374, p = 009), and pride and guilt were also positively correlated (r = 34, p = 02) Testing Hypotheses: Contrast Analyses To investigate our hypotheses about the combined effect of focus and legitimacy on group-based pride, existential guilt, and sympathy, we conducted three planned contrast analyses Contrast analyses are the appropriate method to test the predictions outlined in the introduction Given that our focus manipulation affected perceived fairness in the pilot study, using planned contrasts rather than omnibus interaction tests also better isolates the effects of our manipulations There were three facets to our analyses First, following Rosnow and Rosenthal (1996), we treated our × design as a × design to test our three main hypotheses For each of the three theoretical predictions, a contrast was created that described the hypothesized rank order of means regarding one group-based emotion (A > B = C = D) This is represented in the focal contrast with the coefficients 3-1-1-1 For example, feelings of pride should be greatest when the focus is on the ingroup and the inequality is legitimate (compared to the three other conditions) Thus, the condition in which we expected one of the three group-based emotions to be most intense was weighted with +3 and was compared to the other three conditions which were weighted with –1 Second, to check whether there is systematic variance other than that predicted, orthogonal contrasts were computed in addition to the focal contrast (Abelson & Prentice, 1997) Orthogonal contrasts are important because they reveal whether there is residual variance that is not explained by the focal contrast If the hypothesis represented in the focal contrast is correct, the focal contrast should be significant, and ideally, the orthogonal contrast should not be significant Given that there were four experimental conditions, we had df to compute two orthogonal contrasts (001-1 and 0-211) Third, to account for the intercorrelations between the emotions, and thus general emotionality, we included the nonfocal emotions as covariates in the contrast analyses Means and standard deviations for the emotion measures are reported in Table Pride The 3-1-1-1 contrast for group-based pride had a significant effect, F(1, 40) = 11.61, p = 002, η² = 225 As expected, participants in the IGF legitimate condition reported more pride than participants in the other three conditions However, the second orthogonal contrast (0-211) was significant, F(1, 40) = 6.15, p = 018, indicating that the variance was not fully explained by the focal contrast Existential guilt The -13-1-1 contrast computed for existential guilt had a significant effect, F(1, 40) = 5.09, p = 03, η² = 113 As hypothesized, participants in the IGF illegitimate condition reported greater existential guilt than participants in the other three conditions Neither of the orthogonal contrasts was significant Sympathy The -1-1-13 contrast computed for groupbased sympathy had a significant effect, F(1, 40) = 4.29, p = 045, η² = 097 As expected, feelings of sympathy were higher in the OGF illegitimate condition None of the orthogonal contrasts were significant Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS Disdain A (focus) × (legitimacy) ANOVA was conducted on disdain to investigate how participants would experience an OGF legitimate inequality The ANOVA showed that neither focus, F(1, 44) = 2.04, p = 16, nor legitimacy, F(1, 44) = 2.26, p = 14, nor their interaction (F < 1) had an effect on disdain for the outgroup Discussion Participants identified strongly with the ingroup of psychology students and acknowledged their experimentally established advantage over the social pedagogy students Considering identification as a precondition, the stage was set for group-based emotions.2 More specifically, this study supports our hypotheses regarding distinct emotional experiences of intergroup inequality Pride was greatest when participants were focused on a legitimate ingroup advantage whereas existential guilt was greatest when participants were focused on an illegitimate ingroup advantage Sympathy was most elicited when participants were made to focus on the outgroup’s illegitimate disadvantage It should be noted, however, that one orthogonal contrast for pride was also significant According to Abelson and Prentice (1997), this indicates that additional explanation is needed to fully account for the pattern of results However, this need for additional explanation in no way undermines the empirical support for our hypothesized explanation of pride Disdain did not seem to be a typical emotional response to an OGF legitimate inequality This may have been due to the relative obscurity of these emotion words In any case, in the following studies we concentrate on our three main hypotheses and leave out the OGF legitimate condition STUDY The first aim of Study was to replicate the effects for group-based pride, existential guilt, and sympathy In contrast to Study 1, we used a one-factorial design with three conditions to investigate the joint effect of focus and legitimacy Thus, we excluded the OGF legitimate condition that yielded little in Study The second aim was to investigate the link between emotion and behavioral tendencies To assess behavioral tendencies toward the ingroup and outgroup, a resource distribution task was used Although we expected overall ingroup bias in the distribution of resources to the two groups, we expected pride, existential guilt, and sympathy to differentially predict resource distribution As argued in the introduction, group-based pride, in contrast to guilt and sympathy, should predict behavioral tendencies that favor the ingroup In contrast, existential guilt and sympathy should predict less ingroup favoritism in the distribution of resources 121 Method Participants, Design, and Procedure Psychology students of the Universities of Dresden and Jena were recruited during lectures and took part in the study immediately after the lectures (N = 61, 90% female, M age =21 years, range = 18–30 years) Participants were randomly assigned to one of the three key conditions examined in Study 1: IGF legitimate, IGF illegitimate, and OGF illegitimate The paradigm and the fake newspaper articles were the same as in Study Measures Manipulation checks Recognition of inequality and group identification (α = 74) were measured in the same way as in Study Emotions Guilt (α = 76) and sympathy (α = 70) were measured exactly as in Study For pride, however, we substituted the item superior for happy As such, the pride scale now more narrowly focuses on the experience of pride associated with success and superiority (i.e., proud, successful, superior; α = 75) The eight emotion items were submitted to a principal-axis factor analysis with oblimin rotation A three-factor solution was obtained, which accounted for 57.20% of the common variance The first factor was indicated by the guilt items, the second factor was indicated by the pride items, and the third factor was indicated by the sympathy items Guilt correlated with sympathy (r = 286, p = 03) but not with pride (r = –.114, p = 38); pride and sympathy did not correlate (r = –.133, p = 30) Behavioral tendencies: resource distribution To assess participants’ behavioral tendency to distribute resources to the ingroup and outgroup, we instructed them to imagine they could influence the financial compensation of social scientists on the job market Their task was to distribute 100 monetary units between psychologists and social pedagogues They recorded their distribution in spaces provided in the questionnaire Results Preliminary Analyses Recognition of intergroup inequality Participants’ acknowledgment of group advantage was significantly above the scale midpoint (M = 5.57), t(60) = 13.10, p < 001 Moreover, a one-factorial ANOVA revealed that the acknowledgment was not significantly influenced by the manipulation, F(2, 58) = 1.84, p = 17 Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 122 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN TABLE 2: Means and Standard Deviations for Studies and IGF Legitimate Study Group-based pride Existential guilt Group-based sympathy Study Group-based pride Existential guilt Group-based sympathy IGF Illegitimate OGF Illegitimate M SD M SD M SD 5.58 1.92 3.40 1.85 1.11 1.34 4.50 2.65 3.58 1.32 1.72 1.61 3.78 2.28 5.43 1.51 1.22 1.68 3.78 2.64 5.50 1.29 1.31 1.41 3.15 4.13 5.88 1.75 1.92 1.26 2.89 3.42 6.43 1.64 1.38 1.43 NOTE: IGF = ingroup focus; OGF = outgroup focus Identification Identification with the ingroup of psychology students was significantly above the scale midpoint (M = 4.94), t(60) = 6.20, p < 001 Importantly, the manipulation did not affect identification with the ingroup, F(2, 58) = 1.16, p = 32 Group-Based Emotions To test our hypotheses about the combined effect of focus and legitimacy on the three group-based emotions, we conducted planned contrast analyses with the contrast coefficients 2-1-1 Given that we had three experimental groups, there was only df to compute one orthogonal contrast Again, the two nonfocal emotions were included as covariates in the analyses Means and standard deviations are reported in Table Pride The focal contrast for pride (2-1-1) was significant, F(1, 56) = 10.72, p = 002, η² = 161 As expected, participants in the IGF legitimate condition reported greater pride than in the other two conditions Existential guilt The focal contrast for guilt (-12-1) was significant, F(1, 56) = 4.52, p = 038, η² = 075 In line with our prediction, participants in the IGF illegitimate condition indicated higher guilt than participants in the other two conditions Sympathy The focal contrast for sympathy (-1-12) was significant, F(1, 56) = 23.15, p < 001, η² = 292 As expected, participants in the OGF illegitimate condition indicated stronger sympathy than participants in the other two conditions In all three cases, the orthogonal contrast analyses were not significant, indicating that there was no systematic residual variance left unexplained (all Fs < 2.9) Hence, the predicted patterns were the most reasonable and parsimonious This is an improvement over the findings for pride in Study Behavioral Tendencies: Resource Distribution Overall, participants gave more monetary units to the ingroup (M = 55.97) than to the outgroup (M = 44.03), t(60) = 6.27, p < 001 Thus, participants showed ingroup favoritism in resource distribution As participants were asked to distribute 100 units between the ingroups and outgroups, the amount distributed to one group complemented perfectly that given to the other group Thus, we used distribution to the ingroup as the dependent measure A contrast analysis with the coefficients 2-1-1 and the mean of the ingroup allocation as dependent variable should test whether participants in the IGF legitimate condition showed more ingroup-favoring behavior in the resource distribution task than participants in the other two conditions In fact, participants in the IGF legitimate condition distributed more resources to their ingroup (M = 60.0) than did participants in the IGF illegitimate (M = 54.4) or OGF illegitimate (M = 53.6) conditions, F(1, 58) = 9.89, p = 003, η² = 146 The difference between the IGF illegitimate and OGF illegitimate conditions was not significant (F < 1) To recap, we expected pride to predict greater ingroup-favoring behavior in the resource distribution task whereas we expected guilt and sympathy to lead to less ingroup-favoring behavior A multiple regression analysis was conducted The three group-based emotions were simultaneously entered as predictors of ingroup allocation As expected, higher pride about intergroup inequality predicted greater allocation of resources to a participant’s ingroup (β = 45, p < 001) Guilt did not predict allocation behavior in the distribution task (β = –.13, p > 10) Sympathy, however, led to marginally less ingroup-favoring allocations (β = –.22, p = 065) Mediation Analysis: Group-Based Pride In line with our prediction, IGF legitimate advantage led to the experience of pride as well as to Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS ingroup-favoring behavior in the resource distribution task Thus, we used multiple regression analyses to test whether the relation between IGF legitimate advantage and ingroup-favoring behavior was mediated by pride As existential guilt and sympathy did not predict the behavioral tendency, these emotions did not meet the requirements of mediation and thus were not analyzed as such (see Baron & Kenny, 1986) However, we did account for guilt and sympathy in the mediation models by including them as covariates First, a regression analysis with the IGF legitimate condition, in contrast to IGF illegitimate and OGF illegitimate conditions, as predictor (2-1-1) and the ingroup allocation of the resource distribution task as a criterion showed that both variables were correlated (β = 38, p < 002) Second, IGF legitimate (2-1-1) led to greater pride (β = 40, p = 003) and less sympathy (β = –.21, p = 120), and did not reliably affect guilt (β = –.11, p = 450) Third, the allocation measure was regressed on experimental condition, pride, guilt, and sympathy simultaneously Only the relationship between pride and the behavioral measure was significant (β = 39, p = 002); the relationship between the experimental condition and the allocation behavior became nonsignificant (β = 17, p = 170) Neither guilt (β = 001, p > 10) nor sympathy (β = –.18, p = 130) independently predicted the resource distribution The bootstrap3 confidence interval indicated the indirect effect of pride was significant (.036 to 3.58) This provides evidence for pride as a mediator between legitimate IGF advantage and ingroup-favoring behavior The confidence intervals of guilt (–.208 to 229) and sympathy (–.101 to 854) included zero; thus, their indirect effects were not significant Figure depicts this mediation Discussion The results of Study replicated those of Study Again, an IGF on legitimate advantage led to the greatest pride, whereas an IGF on illegitimate advantage led to the greatest guilt In contrast, an OGF on illegitimate disadvantage led to the greatest sympathy The results also shed light on the relation between group-based emotions and behavioral tendencies Most notably, the more participants expressed pride, the stronger was their ingroup favoritism in the distribution of resources In fact, pride mediated the effect of IGF and legitimacy on this behavioral tendency In contrast, existential guilt did not predict how participants acted in the resource distribution task Although it was not a mediator, sympathy had the opposite effect of pride, leading to less ingroup favoritism The relatively minor role of sympathy is likely due to the fact that this emotion should better explain help giving than ingroup favoritism This issue was addressed in the following study 123 Mediation model of the association between ingroup (IG) focus on legitimate advantage and the resource distribution mediated by group-based emotions, Study *p < 05 **p < 01 Figure STUDY This study differed from both previous studies in three respects First, we aimed at testing whether the same results would occur in a different intergroup relation Thus, students of the University of Jena were told they are relatively advantaged compared to young ethnic German immigrants4 living in Jena with regard to sport opportunities Although this advantage was perceived to be just as great as in the previous studies, the sport domain is likely less important than jobs and thus is likely to be a less strong basis of emotion As such, Study may provide a more subtle test of our hypotheses Second, a different kind of manipulation was used Instead of manipulating focus through the wording of a text that also included information about legitimacy, focus was realized through direct instruction This method should improve the orthogonality of the focus and legitimacy manipulations Third, we examined more concrete behavioral tendencies that allowed participants to help the outgroup or to deny them equal opportunities Method Participants and Design Jena University students (N = 84, 76% female, M age = 21.30 years, range = 18–36 years), who participated under circumstances similar to that of the first two studies, were randomly allocated to the three experimental conditions: IGF legitimate, IGF illegitimate, and OGF illegitimate Procedure and Material A cover story was developed describing the situation between students living in Jena and young ethnic German immigrants living in Jena Participants read about a project supported by a federal ministry called Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 124 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN “integration via sports” that aimed at integrating migrants in the society with the help of public sport clubs Participants were informed that this project really existed but had never been realized in the university context The study was disguised as an opinion survey about whether the university’s sport courses would be suitable for the “integration via sports” program Thus, all participants were informed about the situation of young ethnic German immigrants in Jena compared to students of Jena regarding opportunities for leisure activities The manipulation of focus was realized through direct instruction Participants in the IGF and OGF conditions got the same text but were instructed to focus either on the situation of the ingroup (students) or on the situation of the outgroup (ethnic German immigrants) To ensure that participants followed the focus instruction they were asked to complete a sentence to sum up the situation of the group they were instructed to focus on The sentence in the IGF condition started with “The students of the University of Jena ” and the sentence in the OGF condition started with “The young ethnic German immigrants in Jena ” As in the preceding studies, the legitimacy of the intergroup inequality was manipulated by providing information about the reason for the inequality In the legitimate condition, ethnic German immigrants were described as having problems with the German language, not being willing to engage in public sport clubs, and not taking care of the facilities in their vicinity By contrast, the student ingroup was described as very engaged and committed to the sport clubs, and as taking good care of the sport facilities In the illegitimate condition, the ethnic German immigrants were described as poor, not being in a position to pay the monthly contribution for sport clubs, and living in districts where hardly any space for sports is available The student ingroup was described as coming from families that are not poor and as having a wide variety of sports for very little money (11 euros per semester) with all facilities in the immediate vicinity To summarize, we had two kinds of manipulation texts, a legitimate version and an illegitimate version Participants were then instructed to either focus on the ingroup (IGF legitimate and IGF illegitimate) or on the outgroup (OGF illegitimate) After participants read the text, they completed the questionnaire and were debriefed and thanked for their participation with sweets In a pretest (N = 76, 55% female, M age = 21.50 years, range = 18–28 years), we checked whether our materials were an appropriate manipulation of legitimacy Participants were randomly assigned to the legitimate or illegitimate condition One item asked participants to rank whether they “think that the situation between students and ethnic German immigrants regarding sport opportunities is fair or unfair” on a 7point scale (1 = unfair, = fair) Participants in the legitimate condition reported the sport opportunities between both groups to be more fair (M = 4.40) than did those in the illegitimacy condition (M = 3.80), F(1, 75) = 7.44, p = 008, η² = 090 Although our manipulation of the legitimacy of the ingroup’s advantage in opportunities for sport was successful, it is worth noting that it had a smaller effect than our manipulation of the legitimacy of job opportunities This is likely because it is more difficult to portray an inequality in the opportunity for sport as highly legitimate or illegitimate As a consequence, Study is likely to provide a conservative test of our hypotheses Measures Recognition of intergroup inequality A single item stated “In regard to leisure activities, students are relatively advantaged compared to young ethnic German immigrants.” Participants ranked their response on a 7point scale (1 = not at all, = absolutely) Identification Identification with students of Jena University was assessed by adapting the four items of the scale from Studies and (“I identify with other students ”) on a 7-point scale (1 = absolutely not, = absolutely; α = 81) Focus manipulation check Participants were asked to indicate whether they followed the instruction and focused on the situation of either the ingroup or the outgroup (1 = outgroup, = both groups, = ingroup) Emotions A principal-axis factor analysis with oblimin rotation was performed on the emotion items We obtained a three-factor solution, accounting for 51% of the common variance The first factor represented the emotion guilt (guilty, ashamed, have a bad conscience; α = 76); the second, pride (proud, successful, superior; α = 65); and the third, sympathy (sympathy, compassion; α = 62) Guilt and pride were negatively correlated (r = –.259, p = 02), guilt and sympathy did not correlate significantly (r = 147, p = 18), and sympathy and pride were not correlated (r = –.078, p = 48) Behavioral tendencies In this study we aimed at measuring behavioral tendencies to help the outgroup and provide them equal opportunities Thus, we assessed (a) whether participants would be “willing to let the ethnic German immigrants participate in university sport ” (1 = not at all, = yes, absolutely), (b) “how much money they would pay in the case of ethnic Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS German immigrants being allowed to participate” (1 = less than 8€, = more than 14€), and (b) “if they would sign a petition in favor of the [integration] project” (yes–no) The items were z transformed and combined into an index of behavioral tendencies (α = 70 A positive score indicates support of the integration project; a negative score indicates opposition Results Preliminary Analyses Recognition of inequality Students’ acknowledgment of being advantaged in leisure activities was significantly above the scale midpoint (M = 5.71), t(82) = 13.78, p < 001 A one-way ANOVA showed that the experimental manipulations had no effect on this judgment, F(2, 80) = 1.71, p = 19 Identification Students highly identified with their university ingroup; the mean was well above the scale midpoint (M = 5.76), t(81) = 16.24, p < 001 A oneway ANOVA showed that the level of identification did not differ between the experimental conditions (F < 1) Focus manipulation check Participants appeared to follow the instruction A one-way ANOVA showed that those in the OGF condition focused more on the outgroup (M = 1.48) than participants in the IGF condition (M = 1.95), F(1, 79) = 9.08, p < 01 Group-Based Emotions To test our hypotheses regarding the effects of our manipulations on pride, guilt, and sympathy, we conducted the same planned contrast analyses as in Study (with the contrast coefficients 2-1-1) Means and standard deviations can be found in Table Orthogonal contrasts were used to test whether there was any additional systematic variance besides the expected pattern As seen in Table 2, participants in the IGF legitimate condition showed greater pride than participants in the other two conditions However, this effect was not significant, F(1, 79) = 1.73, p = 19, η² = 021 As predicted, the IGF illegitimate condition led to greater guilt than the two other conditions, F(1, 79) = 9.18, p = 003, η² = 104 And, as expected, participants in the OGF illegitimate condition showed greater sympathy than participants in the other two conditions, F(1, 79) = 5.05, p = 027, η² = 060 The planned contrasts accounted fully for the variance in the three emotions, as the orthogonal contrast analyses were not significant (Fs < 1.2) 125 Behavioral Tendencies We also performed planned contrast analyses on participants’ behavioral tendencies, with the coefficients -2-11 These coefficients test the hypothesis that participants in the IGF legitimate condition oppose help to the outgroup, whereas participants in the other conditions support such help As expected, the IGF legitimate condition had a significant effect on behavioral tendencies to help the outgroup, F(1, 81) = 9.42, p = 003, η² = 104 Participants in the IGF legitimate condition opposed help to the outgroup (M = –.36) However, in both the IGF illegitimate condition (M = 11) and the OGF illegitimate condition (M = 25), participants supported help to the outgroup The level of support in these two conditions of IGF illegitimate advantage did not differ from each other (F < 1) To investigate which emotions predict the behavioral tendency to help the outgroup, we conducted a multiple regression analysis that included all three emotions as predictors As expected, pride about ingroup advantage predicted less support for help to the outgroup (β = –.28, p = 007) Although existential guilt did not predict behavioral tendencies (β = 08, p = 420), sympathy predicted more support for help to the outgroup (β = 34, p = 001) Thus, both pride and sympathy could mediate the effects of our manipulations on the behavioral tendencies Mediation Analysis The IGF legitimate contrast The results correspond to two of the four steps in a mediation analysis (Baron & Kenny, 1986) They show that the manipulation of IGF legitimate advantage only weakly increased pride but significantly decreased the behavioral tendency to help the outgroup When all three emotions and the manipulation were simultaneously regressed on this behavioral tendency, both pride (β = –.25, p = 014) and sympathy (β = 32, p = 003) predicted support for help to the outgroup, but guilt did not (β = 03, p = 758) The experimental manipulation no longer had a significant independent effect on the behavioral tendency to help (β = –.18, p = 09) This suggests full mediation by the group-based emotions More specifically, the bootstrap analyses showed that the indirect effect through sympathy (–.112 to –.002), was significant (Figure 2), as its confidence interval did not include zero Pride predicted less support for help to the outgroup but was not a reliable mediator (–.076 to 005) Consistent with the results, the indirect effect through guilt was not significant (–.045 to 029) Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 126 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN Discussion Mediation model of the association between legitimate ingroup (IG) advantage and support of the integration project mediated by group-based pride and sympathy, Study *p < 05 **p < 01 Figure Mediation model of the association between illegitimate outgroup (OG) disadvantage and support of the integration project mediated by group-based pride and sympathy, Study *p < 05 **p < 01 Figure The OGF illegitimate contrast The results show that the manipulation of OGF illegitimate disadvantage led to increased sympathy and more support for help to the outgroup When the OGF illegitimate manipulation and all three emotions were simultaneously regressed on this behavioral tendency (Figure 3), both pride (β = –.27, p = 012) and sympathy (β = 32, p = 003) predicted support for help to the outgroup, but guilt did not (β = 09, p = 39) Mediation by sympathy was suggested by the fact that the bootstrap analyses showed that the indirect effect through sympathy was significant (.008 to 101), as the confidence interval did not include zero The indirect effects through pride (–.004 to 076) and guilt were not significant (–.013 to 023) The IGF illegitimate contrast Finally, a mediation analysis with the contrast coded IGF illegitimate condition as a predictor was conducted to see whether guilt or one of the other two emotions would contribute something in this specific condition Neither a specific nor a total indirect effect was significantly different from zero The results of Study replicated those of Studies and in another context and with different social groups Independent of the experimental condition, students acknowledged their advantaged position compared with ethnic German immigrants and identified highly with their student ingroup When they received information about illegitimate ingroup advantage over an outgroup, participants reported greater existential guilt When these students received information that young immigrants were illegitimately disadvantaged relative to the student’s ingroup, they reported greater group-based sympathy Even though pride was greater in response to a legitimate ingroup advantage, this effect was relatively small and did not reach significance This is likely because the intergroup inequality in sports opportunities examined in this study was less severe than the unequal job opportunities examined in the preceding studies In addition, the legitimacy of this inequality was based in general attitudes and behavior rather than in the quality of education examined in Studies and As emotion research shows that pride is often based in legitimate advantage in achievement contexts (e.g., Rodriguez Mosquera et al., 2000), the kind of advantage established here is less likely to trigger pride Students had several behavioral options within the context of university sports They could either support the disadvantaged outgroup and act for equal opportunities, or they could oppose equal opportunities As we expected, pride in legitimate ingroup advantage predicted opposition to help and equal opportunities to the outgroup As in the preceding study, existential guilt did not predict behavioral tendencies Consistent with its role in helping and in support of equal opportunities, sympathy played the most important role in these behavioral tendencies Thus, sympathy served as a mediator that explained why illegitimate outgroup disadvantage led to greater support of help and equal opportunities to the ethnic German immigrants Indeed, students’ sympathy about the outgroup’s illegitimate disadvantage led them to support paying more to participate in sports GENERAL DISCUSSION In this research, we investigated how intergroup inequality is emotionally experienced among those who benefit from it We wanted to show that there are different ways the advantaged experience inequality, and we were particularly interested in the group-based emotions of pride, existential guilt, and sympathy The three studies experimentally established an intergroup inequality Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS and manipulated the way the ingroup and outgroup were compared to each other Thus, the intergroup inequality was framed as either advantaging the ingroup or disadvantaging an outgroup by focusing participants’ attention on the self or the other (see also Iyer et al., 2003) By manipulating the legitimacy of the intergroup inequality, we showed that together focus and legitimacy lead to the specific emotions of pride, existential guilt, and sympathy Similar paradigms were used to replicate and thus validate the expected effects over three studies Pride Together, the three studies provide evidence that pride is based on an IGF on legitimate ingroup advantage As discussed previously, the weak results in Study may be due to the special establishment of relative advantage, where a sense of achievement was missing Being advantaged concerning job opportunities is more likely to trigger pride than being advantaged in regard to sport opportunities This article provides what appears to be the first evidence of the basis of group-based pride Even though prior studies showed that individuals celebrate ingroup achievement (e.g., Cialdini., 1976) and evaluate their ingroup more positively than outgroups as a result of comparative success (e.g., Guimond & Dambrun, 2002), no work of which we are aware has explicitly examined feelings of group-based pride Although H J Smith and Tyler (1997) have tied a general feeling of pride in an ingroup to ingroup favoritism, they did not examine the basis of this pride Our studies found that IGF legitimate advantage enhanced ingroup favoritism in resource distribution through the emotion of pride In addition, pride led participants to oppose help to a disadvantaged outgroup and to oppose financial investment in providing equal opportunities to the outgroup (Study 3) Thus, pride about ingroup advantage appeared consistent with the conceptualization offered by Leach et al (2002) As pride appears to play an important role in perpetuating intergroup inequality, it seems deserving of more concerted attention in the study of prejudice and intergroup relations Existential Guilt As discussed in the pilot study, those made to focus on an ingroup advantage tended to justify it by perceiving it as more fair (see also Iyer et al., 2003, Study 2) This tendency to justify ingroup advantage may help explain why group-based guilt appears to be relatively rare (e.g., Leach et al., 2006; for a discussion, see Leach et al., 2002) In fact, in the present studies, the mean-level endorsement of guilt was relatively low Nevertheless, we provided the first evidence that individuals feel the most 127 guilt when made to focus on an ingroup advantage that is illegitimate Although the combined effect of focus and legitimacy on guilt has not been tested before, our results are consistent with research showing that group-based guilt is self-focused (Iyer et al., 2003) and is associated with illegitimacy (Miron et al., 2006) Also consistent with recent research by Leach et al (2006, Study 3), we found no association between guilt about ingroup advantage and concrete behavioral tendencies Following the argument by Leach et al., existential guilt seems to be limited in its explanation of efforts to change inequality between groups There may be two reasons for this First, as pointed out by Steele (1990), guilt, as an IGF emotion, seems to motivate positive behavior toward a disadvantaged outgroup only insofar as it provides redemption from the unpleasant guilt feeling The IGF may constrain relatively advantaged groups to be interested in equal opportunity support for the disadvantaged but to focus on their own well-being Providing compensation may be a form of coping behavior to restore one’s well-being Second, we measured existential guilt because we assumed that responsibility is not crucial to evoke feelings of guilt when benefiting from inequality However, it could be that a sense of responsibility is the missing link to explain behavior tendencies and that responsibility would also strengthen the intensity of guilt feelings If guilt is experienced about a wrongdoing, people may be more motivated to compensate for it Sympathy In all three studies, group-based sympathy was triggered by an illegitimate inequality between the advantaged and disadvantaged groups However, in contrast to guilt, sympathy was based on a focus on the outgroup’s disadvantage As far as we are aware, this is the first evidence for the appraisal basis of group-based sympathy Sympathy was negatively related to ingroup-favoring behavior intentions and predicted such behavior intentions that supported the interests of the disadvantaged group This latter finding is in line with that of Iyer et al (2003, Study 2), who showed sympathy about an outgroup disadvantage to predict support for affirmative action policies designed to provide equal opportunities to a disadvantaged ethnic minority Moreover, Study showed sympathy to mediate the link between an otherfocus on illegitimate outgroup disadvantage and behavioral tendencies to help young immigrants to Germany have an equal opportunity to participate in sports Limitations Our findings are based on participants’ use of specific emotion words to describe their experience as we Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution 128 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN did not use nonverbal measures.5 However, measurement of physiology might not be more appropriate to distinguish between the self-conscious, socially elaborated emotions examined here This is likely why selfreport is still one of the most common ways to measure emotional experiences Furthermore, we examined the willingness to act, but we did not actually observe behavior Nevertheless, we think our measures in Study are good proxies for real behavior Obviously, the present studies were not designed to examine the case of OGF and legitimate disadvantage Only Study included this condition, and there was little evidence of the disdain suggested by Leach et al (2002) However, it may be that the other two dimensions of the Leach et al model may be necessary to produce disdain Greater competition between ingroup and outgroup might also be necessary for disdain Future studies should address this issue and investigate more closely possible emotions triggered by OGF and legitimate disadvantage Knowing about the potential effect of identification for the emotions (Doosje, Branscombe, Spears, & Manstead, 1998), we checked whether our findings were moderated by identification with the ingroup No consistent or reliable moderation effects were found over the three studies Shrout and Bolger (2002) suggested that mediation can be demonstrated by showing that the indirect effect is significantly different from zero They recommend a bootstrap technique (see also Preacher & Hayes, 2004) that computes a confidence interval around the product term (a*b) If zero is not included in the interval, the indirect effect is significant Bootstrapping is superior to the Sobel test because Sobel assumes a*b is normally distributed However, especially in small samples, a more narrow and asymmetrical distribution is found The bootstrap interval converges to the actual distribution of the indirect effect Resettlers are from Eastern European countries, which are German in basic constitutional law on the basis of their German ancestry and maintenance of German cultural heritage We know from our pilot tests that measures of arousal and valence did not account for the three distinct emotions and did not predict behavioral tendencies Conclusion Abelson, R P., & Prentice, D A (1997) Contrast tests of interaction hypothesis Psychological Methods, 2, 315-328 Baron, R M., & Kenny, D A (1986) The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations Journal of Personality & Social Psychology, 51, 1173-1182 Batson, C D., Lishner, D A., Carpenter, A., Dolin, L., HarjusolaWebb, S., Stocks, E L., et al (2003) “ As You Would Have Them Do Unto You”: Does imagining yourself in the other’s place stimulate moral action? Personality and Social Psychology Bulletin, 29, 1190-1201 Bettencourt, B A., Dorr, N., Charlton, K., & Hume, D L (2001) Status differences and ingroup bias: A meta-analytic examination of the effects of status stability, status legitimacy, and group permeability Psychological Bulletin, 127, 520-542 Cialdini, R (1976) Basking in reflected glory: Three (football) field studies Journal of Personality and Social Psychology, 34, 366-375 Cohrs, C., Dimitrova, D., Kalchevska, T., Kleinke, S., Tomova, I., Vasileva, M., et al (2004) Ist patriotischer Nationalstolz wünschenswert? Zeitschrift für Sozialpsychologie, 35, 201-215 Doosje, B., Branscombe, N R., Spears, R., & Manstead, A S R (1998) Guilt by association: When one’s group has a negative history Journal of Personality and Social Psychology, 75, 872-886 Doosje, B., Ellemers, N., & Spears, R (1995) Perceived intragroup variability as a function of group status and identification Journal of Experimental Social Psychology, 3, 410-436 Eisenberg, N (2003) Prosocial behavior, empathy, and sympathy In M H Bornstein, L Davidson, C L M Keyes, & K A Moore (Eds.), Well-being: Positive development across the life course (pp 253-265) Mahwah, NJ: Lawrence Erlbaum Frijda, N H., Kuipers, P., & ter Schure, E (1989) Relations among emotion, appraisal, and emotional action readiness Journal of Personality and Social Psychology, 57, 212-228 Guimond, S., & Dambrun, M (2002) When prosperity breeds intergroup hostility: The effects of relative deprivation and relative gratification on prejudice Personality and Social Psychology Bulletin, 28, 900-912 Hoffman, M L (1976) Empathy, role-taking, guilt, and development of altruistic motives In T Lickone (Ed.), Moral development and behavior: Theory, research, and social issues (pp 281-313) New York: Academic Press Iyer, A., Leach, C W., & Crosby, F J (2003) White guilt and racial compensation: The benefits and limits of self focus Personality and Social Psychology Bulletin, 29, 117-129 The three studies suggest that group-based emotions are an important aspect of intergroup inequality Our findings may be relevant to media representations of disadvantaged groups or to charity campaigns For example, our results suggest that simply highlighting intergroup inequality may not be sufficient to motivate positive intergroup behavior In fact, it may be counterproductive to lead individuals to focus on their ingroup’s advantage alone, as this may encourage them to justify it If raising support for help and equal opportunity is one’s aim, it appears most effective to lead individuals to focus on the illegitimate disadvantage suffered by an outgroup As this condition most often evokes sympathy, it appears to be the most promising way to reduce intergroup inequality by reducing ingroup favoritism and increasing support for efforts at equality between the groups NOTES We are aware of the conceptual distinction between guilt and shame, depending on whether one’s own actions are evaluated negatively or whether they reflect on one’s character However, there are two reasons we did not differentiate between guilt and shame First, we are not concerned with wrongdoings but with the mere perception of illegitimate ingroup advantage triggering existential guilt Agency is not involved here Second, shame and guilt are semantically very close to each other (Smith & Ellsworth, 1985) When research focuses on this narrow semantic field, the differentiation between guilt and shame may be highly relevant The present research, however, focused on the broader semantic field of emotions by contrasting pride, guilt, and sympathy Thus, the small semantic distance between shame and guilt is less meaningful than the wider distance among pride, guilt, and sympathy Nonetheless, we analyzed our data on a single-item level to check whether different patterns of results would occur for the three items Basically, the results were the same and displayed the pattern for the guilt scale REFERENCES Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution Harth, Kessler, Leach / INEQUALITY AND GROUP-BASED EMOTIONS Kessler, T., & Mummendey, A (2001) Is there any scapegoat around? Determinants of intergroup conflict at different categorization levels Journal of Personality and Social Psychology, 81, 1090-1102 Leach, C W., Iyer, A., & Pedersen A (2006) Anger and guilt about ingroup advantage explain the willingness for political action Personality and Social Psychology Bulletin, 32, 1232-1245 Leach, C W., Snider, N., & Iyer, A (2002) “Poisoning the consciences of the fortunate”: The experience of relative advantage and support for social equality In I Walker & H J Smith (Eds.), Relative deprivation: Specification, development, and integration (pp 136-163) New York: Cambridge University Mackie, D M., Devos, T., & Smith, E R (2000) Group-based emotions: Explaining offensive action tendencies in an intergroup context Journal of Personality and Social Psychology, 79, 602-616 Mikula, G (1993) The experience of injustice In W Stroebe & M Hewstone (Eds.), European review of social psychology (Vol 4, pp 223-244) Chichester, UK: Wiley Miron, A M., Branscombe, N R., & Schmitt, M T (2006) Collective guilt as distress over illegitimate intergroup inequality Group Processes and Intergroup Relations, 9, 163-180 Montada, L., & Schneider, A (1989) Justice and emotional reactions to the disadvantaged Social Justice Research, 3, 313-344 Powell, A A., Branscombe, N R., & Schmitt, M T (2005) Inequality as ingroup privilege or outgroup disadvantage: The impact of group focus on collective guilt and interracial attitudes Personality and Social Psychology Bulletin, 31, 508-521 Preacher, K J., & Hayes, A F (2004) SPSS and SAS procedures for estimating indirect effects in simple mediation models Behavior Research Methods, Instruments, and Computers, 36, 317-331 Rodriguez Mosquera, P M., Manstead, A S R., & Fischer, A H (2000) The role of honor related values in the elicitation, experience, and communication of pride, shame, and anger: Spain and the Netherlands compared Personality and Social Psychology Bulletin, 26, 833-844 Rosnow, R L., & Rosenthal, R (1996) Contrasts and interactions redux: Five easy pieces Psychological Science, 7, 253-257 Schmitt, M., Behner, R., Montada, L., Müller, L., & MüllerFohrbrodt, G (2000) Gender, ethnicity, and education as privileges: Exploring the generalizability of the existential guilt reaction Social Justice Research, 13, 313-337 Shrout, P E., & Bolger, N (2002) Mediation in experimental and non-experimental studies: New procedures and recommendations Psychological Methods 7, 422-445 129 Sidanius, J., & Pratto, F (1999) Social dominance: An intergroup theory of social hierarchy and oppression New York: Cambridge University Press Simon, B., & Klandermans, B (2001) Politicized collective identity: A social psychological analysis American Psychologist, 56, 319331 Smith, C A., & Ellsworth, P C (1985) Patterns of cognitive appraisal in emotion Journal of Personality and Social Psychology, 48, 813-838 Smith, E R (1993) Social identity and social emotions: Toward new conceptualizations of prejudice In D M Mackie & D L Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp 297-315) San Diego, CA: Academic Press Smith, H J., & Kessler, T (2004) Group-based emotions and intergroup behavior: The case of relative deprivation In L Z Tiedens & C W Leach (Eds.), The social life of emotions (pp 292-313) New York: Cambridge University Press Smith, H J., & Tyler, T R (1997) Choosing the right pond: The impact of group membership on self-esteem and grouporiented behavior Journal of Experimental Social Psychology, 33, 146-170 Steele, S (1990) White guilt American Scholar, 59, 497-506 Swim, J K., & Miller, D L (1999) White guilt: Its antecedents and consequences for attitudes toward affirmative action Personality and Social Psychological Bulletin, 25, 500-514 Tajfel, H., & Turner, J C (1986) The social identity theory of intergroup behavior In W G Austin & S Worchel (Eds.), Psychology of intergroup relations (2nd ed., pp 7-24) Monterey, CA: Brooks/Cole Turner, J C., Hogg, M A., Oakes, P J., Reicher, S D., & Wetherell, M S (1987) Rediscovering the social group: A self-categorization theory Oxford, UK: Blackwell Verkuyten, M., & Hagendoorn, L (2002) Ingroup favoritism and self-esteem: The role of identity level and trait valence Group Processes & Intergroup Relations, 5, 285-297 Weiner, B (1995) Inferences of responsibility and social motivation In M P Zanna (Ed.), Advances in experimental social psychology (Vol 27, pp 1-47) San Diego: Academic Press Received January 26, 2007 Revision accepted June 12, 2007 Downloaded from http://psp.sagepub.com at University of Sussex Library on January 7, 2008 © 2008 Society for Personality and Social Psychology, Inc All rights reserved Not for commercial use or unauthorized distribution ... asked to complete a sentence to sum up the situation of the group they were instructed to focus on The sentence in the IGF condition started with ? ?The students of the University of Jena ” and the. .. self or the other (see also Iyer et al., 2003) By manipulating the legitimacy of the intergroup inequality, we showed that together focus and legitimacy lead to the specific emotions of pride,. . .Advantaged Group’s Emotional Reactions to Intergroup Inequality: The Dynamics of Pride, Guilt, and Sympathy Nicole Syringa Harth Thomas Kessler Friedrich