Higher Moral Obligations of Tolerance Toward Other Minorities

Abstract

In four experiments, we tested whether members of stigmatized groups are expected to be more tolerant toward other minorities than members of non-stigmatized groups and assessed the consequences of disconfirming those expectancies. Experiments 1 and 2 showed that majority group members expected members of a stigmatized group to be more tolerant toward immigrants, particularly when the stigmatized minority was perceived as having overcome the negative consequences of its victimization. When this tolerance expectation was disconfirmed, stigmatized group members were judged more immoral than members of a non-stigmatized group that held the same intolerant attitudes. Experiments 3 and 4 showed that these effects were driven by the belief that stigmatized groups should derive benefits from their suffering. These findings suggest that stigmatized groups are judged according to stricter moral standards than non-stigmatized groups because majority group members need to make meaning of the undeserved suffering experienced by victims of social stigma.

Keywords

intergroup relations social stigma just world victimhood moral obligations tolerance

Being a member of a stigmatized group can have multiple negative consequences. Members of stigmatized groups are often negatively stereotyped (Eberhardt, Goff, Purdie, & Davies, 2004), recipients of negative emotional responses (Cottrell & Neuberg, 2005), and experience pervasive discrimination in crucial life domains, including health, education, employment, and social interactions (see Gouvier, Sytsma-Jordan, & Mayville, 2003; Major & O’Brien, 2005; Schmitt & Branscombe, 2002).

In the present article, we identify a more subtle consequence of stigmatized group membership: Non-stigmatized group members may hold elevated expectations concerning the moral behavior that stigmatized group members should exhibit. Consistent with recent research on the meaning derived from victimization by third parties (Warner & Branscombe, 2012; Warner, Branscombe, Garczynski, & Solomon, 2011), we propose that stigmatized group members are expected to be more tolerant toward other disadvantaged groups relative to non-stigmatized group members. We argue that this is, in effect, an additional burden on the shoulders of the stigmatized. Indeed, such expectations can be seen as a subtle form of discrimination, with stigmatized groups facing greater social pressure to behave morally than the non-stigmatized. We provide evidence of the higher moral expectations placed on stigmatized groups and examine the underlying processes that explain the negative judgment consequences that occur when such expectations are violated.

The Higher Moral Obligations (HMO) Hypothesis

Recent research has revealed that victims are often judged more harshly than non-victims when they are perceived as having acted immorally. This counterintuitive consequence of having a victimization history has been explained by observers’ need to make meaning of undeserved suffering and protect observers’ belief in a just world. For example, Warner and Branscombe (2012, Experiment 5) found that members of a victimized group were perceived as more obligated to help other unrelated victimized groups when participants believed that the original perpetrators of the harm had not been punished (high justice threat condition) than when participants believed that the perpetrators had been punished (low justice threat condition). These findings are consistent with work by Anderson, Kay, and Fitzsimons (2010), who found that bestowing additional meaning on the lives of individuals who have suffered is a way of dealing with the threat to just world beliefs that victimization can elicit. Anderson et al. showed that participants whose justice motive was temporarily heightened perceived greater meaning (i.e., a more fulfilling existence) in the life of someone who had experienced a tragedy compared with someone who had not experienced a tragedy.

These results suggest that thinking about the meaning of victimization can lead to the perception that victims have obligations to help others and to not do harm. Victims incur these obligations because of the meaning of victimization that observers generate when they focus on victims—that of becoming better people, particularly in terms of their morality (Branscombe, Warner, Klar, & Fernández, 2013). Believing that victims are better off than they would have been without suffering gives the suffering a purpose. Warner and Branscombe (2011) found that to the extent that people believe that victims derive benefits from their suffering, the more they are expected to refrain from harming others. Thus, reinterpreting the victim’s tragic experiences so that benefits are derived can be viewed as a strategy for dealing with the threat to justice that observers experience when confronted with injustice (Lerner, 1980).

Despite the consistent evidence that has been found supporting the moral obligations hypothesis, existing studies have not tested the proposed underlying process (i.e., victims obtaining psychological benefits as a result of their suffering) by directly manipulating it. In line with recommendations for investigating mediation hypotheses (Spencer, Zanna, & Fong, 2005), we sought to not only obtain evidence that benefit finding mediates the HMO placed on stigmatized groups, but to also test the underlying process via its experimental manipulation. Another important aim of the present research was to extend findings concerning HMO beyond victims of genocide or individual wrongdoing to the realm of social stigmatization.

The Present Research

Our first goal for the current research was to test whether the HMO hypothesis applies to stigmatized groups. Second, we sought to assess the judgment consequences when stigmatized group members violate observers’ moral expectations. Finally, we wanted to deepen our understanding of the process underlying the HMO effect. In Experiment 1, we tested whether members of a group with a physical stigma (people with dwarfism) are expected to be more tolerant toward another stigmatized group (immigrants) relative to members of a non-stigmatized group. In Experiment 2, we assessed whether having overcome prior victimization is critical for the elevated moral expectations observed in Experiment 1. In Experiment 3, we investigated whether the HMO placed on a different stigmatized group (gay people) is mediated by expectations that they should have derived benefits from the discrimination they have suffered. In Experiment 4, we directly manipulate participants’ beliefs about the psychological strength that a stigmatized minority (gay people) derived from having experienced discrimination to assess the consequences when observers’ moral expectations are violated.

Experiment 1

In Experiment 1, we tested whether members of a highly stigmatized group (people with dwarfism), relative to members of a non-stigmatized group (young people), are expected to be more tolerant toward another stigmatized minority (immigrants). After assessing participants’ expectations, we manipulated the actual attitudes that the target group (stigmatized vs. non-stigmatized) holds toward immigrants. We predicted that participants would expect the stigmatized group to be more tolerant toward immigrants than the non-stigmatized group. Participants’ expectancies should be confirmed more strongly when the stigmatized group displays a positive attitude toward another victimized minority compared with when the non-stigmatized group does so. We also expected that a negative attitude on the part of the stigmatized group would be judged as more immoral than a negative attitude expressed by the non-stigmatized group. Although we expected that overall stronger negative emotions would be evoked when attitudes toward immigrants are negative (rather than positive), we predicted that the process underlying these negative reactions would depend on whether the group expressing intolerance was stigmatized or not. In line with the HMO hypothesis, we predicted that negative emotional reactions toward those displaying a negative attitude would be driven by harsher moral judgments in the case of the stigmatized group, but not in the case of the non-stigmatized group.

Method

Participants and design

Ninety-four Spanish undergraduates participated in exchange for course credit. After dismissing those who were immigrants or had a physical disability, 89 (73 women and 16 men, M age = 32.7, SD = 7.8) remained for the analysis. Participants were randomly assigned in a 2 (target group: stigmatized vs. non-stigmatized) × 2 (attitude toward immigrants: positive vs. negative) between-subjects design.

Procedure

The study was presented as assessing opinions about the results of prior research concerning the attitudes of different social groups in Spain toward immigrants. Participants were told that they would be receiving some of the results from that research (which were actually fictitious), concerning one particular social group. Participants assigned to the stigmatized group condition were presented with results concerning the attitudes of people with dwarfism. Participants assigned to the non-stigmatized group condition were presented with the results concerning young people aged 17 to 23 years. Before seeing the results, participants first indicated their expectations about the attitudes that the specific target group would show toward immigrants. Subsequently, participants in the negative attitude condition learned that the target group’s attitude toward immigrants was negative (“70% of the [target] group think that illegal immigrants should be expelled out of Spain”). In contrast, participants in the positive attitude condition learned that the target group’s attitude toward immigrants was positive (“70% of the [target] group think that Spain should provide help and facilitate the integration of illegal immigrants in our society”). Once all measures were completed, participants were fully debriefed and thanked.

Measures

Participants responded to all items on a scale ranging from 0 (totally disagree) to 6 (totally agree).

Expected attitude

Prior to learning the target group’s attitude toward immigrants, participants were asked to indicate to what extent they expected the attitude of the target group to be “Tolerant,” “Racist” (reverse-scored), “Generous,” “Biased” (reverse-scored), “Egalitarian,” and “Supportive,” alpha = .89.

Confirming expectancies

After learning the target group’s ostensible attitude, participants were asked the extent to which the survey’s results had confirmed their previous expectancies: “To what extent do you agree the expectancies you had about [target group’s] attitude toward immigrants were confirmed.”

Negative emotions

Participants reported how they felt after having learned the target group’s attitude toward immigrants on the following emotions: “Disappointed,” “Sad,” “In a good mood” (reverse-scored), “Cheerful” (reverse-scored), “Angry,” “Happy” (reverse-scored), and “Uncomfortable,” alpha = .95.

Perceived morality of target group’s attitude

Using a scale adapted from Truxillo, Bauer, and Sanchez (2001), participants were asked to indicate the extent to which the target group’s attitude toward immigrants was “Respectful,” “Immoral” (reverse-scored), “Disloyal” (reverse-scored), and “Objective,” alpha = .90.

Manipulation checks

Target group’s attitude toward immigrants

A single item (“The [target] group has a negative attitude toward immigrants”) was used to check the attitude manipulation.

Perception of the target group as a disadvantaged minority

Two items assessed whether the target group was perceived to be a disadvantaged minority (e.g., “The [target] group suffers discrimination in our society”), r(88) = .72, p < .001.

Perception of immigrants as a disadvantaged minority

One item tested our assumption that participants would perceive immigrants as a disadvantaged minority (“Immigrants suffer discrimination in our society”).

Results

Manipulation checks

We conducted a 2 (target group: stigmatized vs. non-stigmatized) × 2 (attitude toward immigrants: positive vs. negative) ANOVA on each measure. The main effect of the attitude manipulation indicated that participants believed that a more negative attitude was expressed in the negative compared with the positive attitude condition, M = 4.28, SD = 1.42 versus M = 1.86, SD = 1.37, F(1, 85) = 67.02, p < .001, η² = .44. Neither the main effect of target group nor the interaction was significant, Fs < 1.19, ps > .28.

A significant main effect of target group indicated that participants perceived the target group as more disadvantaged in the stigmatized compared with the non-stigmatized condition, M = 3.21, SD = 1.32 versus M = 0.62, SD = 0.91, F(1, 85) = 120.37, p < .001, η² = .59. Neither the main effect of attitude nor the interaction was significant, Fs < 1.78, ps > .19.

The measure assessing the extent to which immigrants suffer discrimination was higher than the midpoint of the scale (3), M = 4.39, SD = 1.36, t(88) = 9.65, p < .001, indicating that participants perceived immigrants as a disadvantaged minority. The ANOVA revealed no significant effects for this measure, Fs < 0.88, ps > .35.

Main analyses

Table 1 presents the descriptive statistics for each measure by condition.

Table 1.

Means and Standard Deviations of the Effects of Attitude Valence and Target Group Manipulations on the Dependent Measures, Experiment 1.

	Expected attitude		Confirming expectancies		Perceived morality		Negative emotions
	Stigmatized	Non-stigmatized	Stigmatized	Non-stigmatized	Stigmatized	Non-stigmatized	Stigmatized	Non-stigmatized
Negative attitude	3.31 (0.75)	2.95 (0.75)	2.18 (1.44)	2.58 (1.72)	1.66 (0.76)	2.21 (0.79)	4.05 (0.87)	3.98 (0.89)
Positive attitude	3.57 (1.03)	2.86 (1.11)	3.47 (1.26)	2.38 (1.66)	4.68 (0.71)	4.24 (1.35)	1.27 (0.81)	0.98 (0.67)

Expected attitude

As predicted, there was a main effect of target group on expected attitude, F(1, 85) = 7.53, p < .01, η² = .08. Participants expected members of the stigmatized group to have more positive attitudes toward immigrants than members of the non-stigmatized group, M = 3.43, SD = 0.89 versus M = 2.91, SD = 0.94. Neither the attitude manipulation nor the interaction was significant, Fs < 0.77, ps > .38.

Confirming expectancies

Neither the main effect of target nor the attitude was significant, Fs < 2.70, ps > .10. As predicted, the target × attitude interaction was significant, F(1, 85) = 5.18, p < .05, η² = .06. When the target group was stigmatized, a positive attitude confirmed participants’ expectancies to a greater degree than a negative attitude, M = 3.47, SD = 1.26 versus M = 2.18, SD = 1.44, t(39) = 3.04, p < .01. When the target group was non-stigmatized, there was no significant difference in expectancy confirmation across conditions, M = 2.38, SD = 1.66 versus M = 2.58, SD = 1.72, for the positive and the negative attitudes, respectively, t(46) = 0.43, p = .67.

Perceived morality of the target group’s attitude

The main effect of attitude was significant, F(1, 85) = 152.82, p < .001, η² = .65, with the positive attitude (M = 4.44, SD = 1.12) being judged more moral than the negative attitude (M = 1.95, SD = 0.82). The main effect of target group was not significant, F(1, 85) = 0.07, p = .79. As predicted, the target × attitude interaction was significant, F(1, 85) = 5.95, p < .05, η² = .07. Participants rated the negative attitude of the stigmatized group as particularly low in morality (M = 1.66), significantly lower than the negative attitude of the non-stigmatized group (M = 2.21), t(43) = −2.38, p < .05. The morality of the positive attitude did not significantly differ by target group, M = 4.68 versus M = 4.24, for the stigmatized and the non-stigmatized group, respectively, p = .20.

Negative emotions

Only the main effect of the attitude manipulation was significant, F(1, 85) = 279.01, p < .001, η² = .77. Negative attitudes elicited more negative emotions than positive attitudes, M = 4.02, SD = 0.87 versus M = 1.11, SD = 0.74. Neither the main effect of the target group nor the interaction was significant, Fs < 1.07, ps > .30.

Underlying process

We hypothesized that participants expected the stigmatized group to be more tolerant toward immigrants because victimized groups are judged according to a higher moral standard of conduct than non-victimized groups. We therefore predicted that the effect of expressing negative attitudes toward immigrants on perceivers’ emotions would be driven by the perceived morality of the target group’s attitude only for the stigmatized group. To test this hypothesis, we used Hayes’s (2012) PROCESS macro for SPSS, specifying Model 59. Table 2 shows the relevant output of the macro. The significant IV × Moderator and Mediator × Moderator interactions on the mediator and dependent variable models, respectively, together with the signs of the coefficients, indicate that, as predicted, the indirect effect of an intolerant attitude on negative emotions through perceived morality was contingent on the moderator, target group. Consistent with our hypothesis, while the indirect effect in the case of the stigmatized group was significant (b = −.80, 95% confidence interval [CI] = [−1.324, −0.310], n boots = 5,000), it was not significant for the non-stigmatized group (b = −.14, 95% CI = [−0.463, 0.099], n boots = 5,000). These results confirm that, as expected, the negative emotional reaction evoked by expressing negative attitudes toward immigrants was mediated by harsher moral judgments only in the case of the stigmatized group.

Table 2.

Outcome of the PROCESS Macro (Model 59) Used to Test Whether the Indirect Effect of the Attitude Valence Manipulation on Negative Emotions via Perceived Morality Was Contingent on the Target Group Manipulation, Experiment 1.

Variable	B	p
Mediator variable model
Constant	3.20	.00
Attitude manipulation (IV)	1.27	.00
Target group manipulation (Moderator)	0.03	.79
Attitude × Target group	−0.25	.02
Dependent variable model
Constant	3.64	.00
Attitude manipulation (IV)	−0.99	.00
Target group manipulation (Moderator)	−0.71	.03
Attitude × Target group	−0.37	.02
Perceived morality (Mediator)	−0.33	.00
Perceived morality × Target group	0.19	.05

Note. IV = Independent Variable.

Discussion

Study 1 supported the hypothesis that perceivers apply more demanding moral standards to a stigmatized group (people with dwarfism) than a non-stigmatized group when it comes to judging their behavior toward another stigmatized minority (immigrants). Participants had expectations of greater tolerance on the part of the stigmatized group, and those expectations were disconfirmed when the stigmatized group showed negative attitudes toward another disadvantaged minority. Moreover, the negative attitude of the stigmatized group was seen as more immoral compared with when the same attitude was expressed by the non-stigmatized group. Participants reported a similar level of negative affect across the stigmatized and the non-stigmatized conditions. That is, although an intolerant attitude held by a stigmatized group was perceived as particularly immoral—in fact, more immoral than the same negative attitude held by a non-stigmatized group—the emotional reaction toward those expressing the intolerance was equally negative despite the stigmatized condition of the target group. However, the process underlying this negative emotional reaction was contingent on the stigmatized condition of the target group: The negative effect of intolerant attitudes toward immigrants on participants’ emotional responses was mediated by the extent to which participants perceived the attitude of the stigmatized group as immoral. No such mediation by immorality was found in the non-stigmatized condition. Thus, the process responsible for observers’ negative emotions depends on whether the intolerant group is stigmatized or not. When the group that holds negative attitudes toward another disadvantaged minority is stigmatized, observers feel negative emotions because they perceive the stigmatized group as immoral.

Experiment 2

According to the HMO hypothesis (Branscombe et al., 2013; Warner & Branscombe, 2012), the moral expectations placed on victims stem from observers’ desire to protect their belief in a just world, which is threatened by undeserved suffering. Therefore, expecting stigmatized groups to have improved as a consequence of their suffering provides meaning to their victimhood. If that is so, stigmatized group members should be held to a particularly high moral standard of conduct if they have had the opportunity to overcome their suffering and have therefore been able to derive benefits (i.e., by becoming better, stronger people) from the social stigmatization they have experienced.

To test this hypothesis, we employed the same attitudes toward immigrants manipulation used in Experiment 1. In all conditions, the target group was people with dwarfism, but we varied whether this stigmatized group has overcome its early discrimination experience or not. We predicted that participants would expect the overcome stigmatized group to be more tolerant than the non-overcome stigmatized group. Furthermore, when this expectation is disconfirmed (i.e., when the group has overcome its suffering but their attitudes toward immigrants are still negative), participants should report having their expectancies more strongly violated, judge the intolerant attitude as particularly immoral, and feel negative emotions most strongly. Thus, we predicted an Overcome × Attitude interaction on expectancy violation, perceived morality, and negative emotions. Because we argue that the HMO hypothesis entails higher moral demands for victims, we also predicted that the expected interaction on negative emotions would be explained by the extent to which participants negatively evaluate the target group’s attitude in terms of morality. That is, we predicted that the expected Overcome × Attitude interaction effect on negative emotions would be mediated by perceived morality.

Method

Participants and design

Ninety-nine Spanish undergraduates participated for course credit. After dismissing those who were immigrants or had a physical disability, 91 (78 women and 13 men, M age = 30.6, SD = 7.0) remained for the analysis. Participants were randomly assigned in a 2 (overcome victimization: overcome vs. non-overcome) × 2 (attitude toward a stigmatized minority: negative vs. positive) between-subjects design.

Procedure

Participants in all conditions were first presented with a summary of the results of ostensibly prior research indicating that people with dwarfism encounter high levels of discrimination and social exclusion during their childhood and adolescence. Participants were given fictitious data about the group’s early suffering, for example, “82.4% of people with dwarfism have faced serious difficulties in their relations with peers during childhood.”

The overcome versus non-overcome manipulation

Participants then received information concerning the consequences of this past discrimination for people with dwarfism as adults. Participants in the overcome condition learned that the majority of adults with dwarfism have overcome all negative consequences of past discrimination, while participants in the non-overcome condition learned that the majority of adults with dwarfism have not overcome this past discrimination. We provided some fictitious data to support this conclusion, which was supposedly obtained from a representative sample of adults with dwarfism, for example, “the majority of adults with dwarfism (73.4%) state that they are highly satisfied with their lives” (overcome condition) versus “the majority of adults with dwarfism (73.4%) state that they are highly unsatisfied with their lives” (non-overcome condition). As in Experiment 1, participants at this point indicated their expectations concerning this group’s attitude toward another stigmatized minority.

Manipulation of the group’s attitude

Participants were provided with the results of fictitious research concerning the attitudes of adults with dwarfism toward immigrants, which were described as positive or negative. After completing the dependent measures, participants were fully debriefed and thanked.

Measures

As in Experiment 1, participants completed measures assessing the expected attitude of the target group, the degree to which their expectancies were confirmed, perceived morality of the target group’s attitude, and negative emotions (all αs greater than .85). All scales ranged from 0 (totally disagree) to 6 (totally agree) and were identical to those used in Experiment 1, except for confirming expectancies, which in Experiment 2 consisted of two items: “The results of the research confirmed my expectancies about how the attitude of people with dwarfism toward immigrants was going to be”; and “The attitude of people with dwarfism toward immigrants has surprised me” (reverse-scored), r(90) = .76, p < .001.

Manipulation checks

The same measures as in Experiment 1 were used to check the attitude manipulation and to confirm that participants considered immigrants a disadvantaged minority. To assess whether the overcome manipulation was effective, we asked participants to indicate the extent to which they agreed with the following item: “The majority of people with dwarfism are able to overcome the social problems they had with peers during childhood and adolescence.” Finally, we included one item assessing whether participants perceived people with dwarfism as having suffered discrimination in the past (“People with dwarfism suffer discrimination during their childhood and adolescence”).

Results

Manipulation checks

The 2 (overcome victimization) × 2 (attitude toward immigrants) ANOVA on the target group’s attitude toward immigrants revealed that participants in the negative attitude condition perceived the group’s attitude as more negative (M = 0.98, SD = 0.84) than participants in the positive condition (M = 4.35, SD = 1.49), F(1, 87) = 211.54, p < .001, η² = .71. A main effect of the overcome manipulation also emerged with participants in the overcome condition perceiving the attitude of the target group as more negative than participants in the non-overcome condition, M = 2.98, SD = 2.11 versus M = 2.17, SD = 1.97, F(1, 87) = 13.84, p < .001, η² = .14, and the interaction was marginally significant, F(1, 87) = 3.06, p = .08, η² = .03. A follow-up analysis indicated that, when the attitude was negative, participants perceived it as more negative when the group had overcome its suffering than when the group had not overcome, M = 5.00, SD = 0.77 versus M = 3.73, SD = 1.75, t(41) = 3.06, p < .01. When the attitude was positive, this difference was marginally significant, M = 1.21, SD = 0.98 versus M = 0.75, SD = 0.61, t(46) = 1.95, p = .06. This is consistent with our HMO hypothesis: The negative attitude toward a disadvantaged minority by a stigmatized group that has overcome its past suffering tended to be perceived as particularly negative.

The 2 × 2 ANOVA on the overcome manipulation check revealed a significant main effect of the overcome factor, F(1, 87) = 18.57, p < .001, η² = .18. Participants in the overcome condition agreed that the group had overcome their past victimization more than participants in the non-overcome condition, M = 3.56, SD = 1.41 versus M = 2.24, SD = 1.48, p < .001. Neither the main effect of the attitude manipulation nor the interaction was significant, Fs < 0.06, ps > .82.

Both the indexes assessing whether people with dwarfism have suffered discrimination in the past and whether immigrants in Spain suffer discrimination were higher than the scale midpoint (3), M = 4.98, SD = 1.15, t(93) = 16.34, p < .001 and M = 4.51, SD = 1.19, t(93) = 12.10, p < .001, respectively. The 2 × 2 ANOVAs on these measures revealed no significant effects (Fs < 1.18, ps > .28).

Main analyses

Table 3 presents the descriptive statistics for all measures.

Table 3.

Means and Standard Deviations of the Effects of Attitude Valence and Overcome Manipulations on the Dependent Measures, Experiment 2.

	Expected attitude		Confirming expectancies		Perceived morality		Negative emotions
	Overcome	Non-overcome	Overcome	Non-overcome	Overcome	Non-overcome	Overcome	Non-overcome
Negative attitude	4.32 (0.82)	3.57 (1.14)	0.67 (0.84)	2.41 (1.65)	1.88 (0.77)	2.59 (1.33)	4.00 (0.80)	3.10 (1.04)
Positive attitude	4.04 (0.94)	3.20 (0.95)	3.98 (1.42)	3.00 (1.51)	4.83 (0.71)	4.69 (0.79)	1.31 (0.86)	1.33 (0.99)

Expected attitude

Consistent with our hypothesis, the overcome manipulation significantly affected participants’ expectancies for the target group’s attitude toward another stigmatized group, F(1, 87) = 15.27, p < .001, η² = .15. Participants in the overcome condition expected more positive attitudes than participants in the non-overcome condition, M = 4.17, SD = 0.88 versus M = 3.38, SD = 1.05. Because this measure preceded the attitude manipulation, the attitude expressed had no effect on this variable, F(1, 87) = 2.50, p = .12.

Confirming expectancies

Participants’ expectancies were confirmed more strongly when the attitude was positive than when it was negative, M = 3.49, SD = 1.53 versus M = 1.56, SD = 1.57, F(1, 87) = 44.27, p < .001, η² = .34. There was no main effect of the overcome manipulation, F(1, 87) = 1.69, p = .20, but, as predicted, the Overcome × Attitude interaction was significant, F(1, 87) = 21.52, p < .001, η² = .20. When the attitude expressed was positive, participants in the overcome condition reported that their expectancies were confirmed more so than in the non-overcome condition, M = 3.98 versus M = 3.00, t(46) = −2.31, p < .05. When the attitude was negative, the opposite pattern was found: Participants reported their expectancies to be disconfirmed in the overcome condition compared with the non-overcome condition, M = 0.67 versus M = 2.41, t(41) = 4.34, p < .001.

Perceived morality of the target group’s attitude

There was a significant main effect of attitude, F(1, 87) = 168.17, p < .001, η² = .66, with a positive attitude being judged more moral than the negative attitude, M = 4.76, SD = 0.75 versus M = 2.24, SD = 1.14. Although the main effect of overcome was not significant, F(1, 87) = 2.10, p = .15, the Overcome × Attitude interaction was F(1, 87) = 4.83, p < .05, η² = .05. Participants rated the negative attitude of the overcome stigmatized group as significantly less moral than the negative attitude of the non-overcome stigmatized group, M = 1.88 versus M = 2.59, t(41) = −2.13, p < .05. The positive attitude was seen as equally moral regardless of whether the stigmatized group had overcome or not, M = 4.83 versus M = 4.69, t(46) = 0.67, p = .50.

Negative emotions

The main effect of the attitude manipulation indicated that negative attitudes evoked more negative emotions than positive attitudes, M = 3.54, SD = 1.03 versus M = 1.32, SD = 0.92, F(1, 87) = 130.68, p < .001, η² = .60. There was also a significant main effect of the overcome manipulation, with the overcome stigmatized group evoking more negative emotions than the non-overcome, M = 2.57, SD = 1.59 versus M = 2.18, SD = 1.34, F(1, 87) = 5.09, p < .05, η² = .06. As predicted, these main effects were qualified by the Overcome × Attitude interaction, F(1, 87) = 5.65, p < .05, η² = .06. Participants reported more negative emotions when the stigmatized group had a negative attitude toward immigrants and had overcome their suffering than when the stigmatized group had a negative attitude but had not overcome their past, M = 4.00 versus M = 3.10, t(41) = −3.18, p < .01. However, there was no difference in negative emotions between the overcome and the non-overcome conditions when the attitude was positive, M = 1.31 versus M = 1.33, t(46) = 0.09, p = .93.

Underlying process

We argue that the Overcome × Attitude interaction effect on participants’ negative emotions is due to the overcome stigmatized group being judged according to a harsher moral standard. To test this hypothesis, we followed Hayes (2012) and specified Model 8 in the PROCESS macro. Figure 1 summarizes the results relevant to the test of mediated moderation. When controlling for the main effects of both manipulations, the Overcome × Attitude interaction predicted perceived morality, and greater morality in turn reduced negative emotions. Indicating mediated moderation, the direct effect of the Overcome × Attitude interaction on negative emotions when the mediator was included in the model became non-significant and the indirect effect was significant, b = −.14, 95% CI = [−0.280, −0.021], n boots = 5,000.

Figure 1.

Experiment 2 test for mediated moderation, indicating that the Overcome × Attitude interaction effect on negative emotions was mediated by perceived morality of the target group’s attitude.

Discussion

Experiment 2 revealed that, when a stigmatized group was believed to have overcome its past suffering but was intolerant toward another stigmatized minority, participants judged their attitude as more immoral and reported feeling higher levels of negative emotions compared with when the stigmatized group was intolerant but had not overcome its past victimization. Overcoming past victimization was therefore a crucial moderator of the extent to which observers expected stigmatized groups to be tolerant toward another disadvantaged group. These results suggest that a process of making meaning of underserved suffering underlies the HMO placed on stigmatized group members. We argue that overcoming past victimization is perceived by observers as a form of personal growth. Because of this, those stigmatized group members who have overcome their prior suffering are seen by majority group members as having gained benefits, which in turn leads to expectations of more moral conduct by the overcome stigmatized minority. There is, however, an important limitation of Experiments 1 and 2 with regard to our argument: We did not directly measure nor manipulate observer benefit finding in either study. A second limitation concerns the generalizability of these results, as we have focused only on one stigmatized minority (people with dwarfism). With the aim of addressing these limitations, we conducted two additional studies. In Experiment 3, we measured the extent to which members of a different stigmatized minority—gay people—were expected to derive psychological benefits from their discrimination experiences and expectations of tolerance toward immigrants relative to non-stigmatized groups. In Experiment 4, we directly manipulated participants’ beliefs concerning the growth and strength derived as a consequence of gay people having experienced discrimination. We then tested whether participants’ tolerance expectancies were contingent on derived benefits and whether violating those expectancies has negative consequences for responses to the stigmatized group.

Experiment 3

Bestowing additional meaning on victimized individuals is a way of dealing with the threat to just world beliefs that victimization can elicit (Anderson et al., 2010). Expecting stigmatized group members to grow in terms of their psychological strength and become more moral persons capable of dealing with adversity than the non-stigmatized can be explained by majority group members’ need to make meaning of undeserved suffering (Warner & Branscombe, 2011). If victimized individuals are expected to gain in moral character as a consequence of their suffering, they will be expected to behave accordingly, that is, to be more tolerant and generous toward others. Therefore, expecting members of stigmatized minorities to obtain psychological benefits as a result of their suffering (e.g., becoming stronger and more moral people) should mediate perceivers’ higher expectations that stigmatized group members will be generous in their treatment of others who have suffered discrimination. To test this hypothesis, we measured the extent to which members of a stigmatized minority (gay people) are expected to gain psychological benefits from experiences associated with their group membership compared with two non-stigmatized minorities (civil servants and bank employees). We also measured the extent to which each group was expected to have a positive attitude toward immigrants and predicted greater expectations of tolerance for the stigmatized group than the non-stigmatized group. This expectation of tolerance for the stigmatized group should be driven by the perceived benefits derived from their suffering compared with non-stigmatized groups.

Method

Participants

In all, 105 Spanish undergraduates participated in exchange for course credit. After excluding those who belonged to one of the groups involved in the manipulations (i.e., gay people, civil servants, bank employees, or immigrants), 96 participants (62 women and 34 men, M age = 36.0, SD = 10.3) remained.

Procedure and materials

Similar to Experiments 1 and 2, participants were presented with the results of a study about the attitudes that different groups in society hold toward immigrants. Participants were told that they would read the results concerning only one of the many groups assessed. In reality, participants were randomly assigned to read about the attitudes of one of three groups: one stigmatized (gay people, n = 34) or one of two non-stigmatized (civil servants, n = 30, or bank employees, n = 32). We chose civil servants and bank employees as comparison groups because they are non-stigmatized and are regarded as out-group categories for our university student participants. Upon completion of the materials, participants were fully debriefed and thanked.

Unless stated otherwise, all items were rated on a scale ranging from 0 (totally disagree) to 6 (totally agree). Participants were asked to indicate the extent to which the experiences associated with being a member of [the target group] make members of this group: “Stronger, more resistant to pain and suffering,” “Especially motivated to improve themselves,” “People who appreciate and value life very much,” “More sensitive to the needs of others,” “People who have particularly high moral qualities,” and “Better persons in terms of their moral qualities” (α = .92). Once participants completed this benefits of group membership scale, they indicated the extent to which they expected the [the target group’s] attitude toward immigrants to be “Tolerant,” “Biased” (reverse-scored “Egalitarian,” “Sympathetic,” and “Generous” (α = .86). Finally, participants answered two items assessing the extent to which they considered the target group a stigmatized minority (e.g., “This group [target group] is a disadvantaged minority in Spain” and “Members of the [target group] face social injustice because of their group membership”), r(95) = .75, p < .001.

Results

Preliminary analysis

To ensure that the measures were distinguishable as intended, we conducted two factor analyses. First, we performed a principal components analysis with the six items assessing benefits of group membership. The results yielded, as expected, a single factor in which all six items had high loadings (all loadings > .74). Together with the high reliability of this scale (α = .92), this indicates that participants perceived, as intended, all six items as capturing character benefits. Second, to ensure that the benefits mediator was empirically distinguishable from the dependent variable, expectations of tolerance, we conducted a principal components analysis with varimax rotation on the 11 items composing both measures. As expected, two factors emerged, accounting for 69.5% of the variance. The six benefits items loaded highly on the first factor (all loadings > .77) and low on the second factor (all loadings < .37); the five tolerance expectations items loaded highly on the second factor (all loadings > .75) and low on the first factor (all loadings < .25). Although significant, the correlation between these two measures was low, r = .32, p < .001.

As intended, gay people (M = 3.35, SD = 1.32) were perceived as more stigmatized than civil servants (M = 1.00, SD = 1.17) and bank employees (M = 0.98, SD = 0.97), ts > 7.48, ps < .001. These latter two groups were perceived as equally non-stigmatized, t(59) = 0.06, p = .95. We therefore collapsed the civil servants and bank employees into a single non-stigmatized group and compared it with the stigmatized group, gay people.

Main analyses

Consistent with our hypothesis, gay people were expected to derive greater benefits from experiences associated with their group membership compared with non-stigmatized groups, M = 2.57, SD = 1.27 versus M = 1.57, SD = 1.01, F(1, 94) = 17.96, p < .001, η² = .16. Furthermore, gay people were expected to have more positive attitudes toward immigrants than non-stigmatized group members, M = 3.52, SD = 0.89 versus M = 2.87, SD = 0.76, F(1, 94) = 14.18, p < .001, η² = .13.

To test our main hypothesis that perceived benefits mediated the higher expectations of tolerance effect, we used the PROCESS macro for SPSS (Hayes, 2012, Model 4) with the target group manipulation as independent variable, benefits from suffering as mediator, and expectations of tolerance as dependent variable. Target group was coded with −1 for the stigmatized and 1 for the non-stigmatized conditions. The results revealed that target group had a direct effect on benefits (b = .50, SE = .12, p < .001) and expectations of tolerance (b = .36, SE = .09, p < .001). When the mediator (benefits) was added to the model, benefits predicted expectations of tolerance (b = .17, SE = .07, p < .05) and the direct effect of target group on expectations of tolerance decreased, b = .28, SE = .09, p < .01. Indicating mediation, the indirect effect of the target group manipulation on expectations of tolerance via benefits from suffering was significant, b = .09, boot SE = .05, 95% CI = [0.012, 0.203], n boots = 5,000.¹

Discussion

The results of Experiment 3 showed that gay people were indeed expected to be more tolerant toward another disadvantaged group, immigrants, compared with non-stigmatized group members. Moreover, to the extent that gay people were expected to derive psychological benefits from their experiences as a stigmatized minority, higher expectations were placed on them relative to the non-stigmatized. These results support our central claim that a process of making meaning of undeserved suffering underlies the higher moral expectations put on stigmatized minorities by majority group members. In Experiment 4, we manipulate, rather than measure, participants’ beliefs about the psychological benefits that a stigmatized group (i.e., gay people) has derived as a consequence of experiencing discrimination.

Experiment 4

We first manipulated participants’ beliefs about the consequence of gay people experiencing discrimination—that it results in greater psychological strength or weakness. Following this manipulation, participants’ expectations concerning gay people’s attitudes toward immigrants were assessed. Feedback from ostensible prior research with gay people was used to vary the attitudes toward immigrants held by gay people, which were described in either positive or negative terms. Participants’ moral judgments and emotional reactions to the target group’s attitude were assessed. We hypothesized that participants would expect gay people to have a more positive attitude when they were said to have been psychologically strengthened (vs. weakened) by their experiences with discrimination. We also expected that gay people with negative attitudes toward immigrants would be judged as particularly immoral and evoke high levels of negative emotions when participants thought that gay people were strengthened rather than weakened by discrimination. Moreover, we expected this two-way interaction on negative emotions to be mediated by perceived morality (i.e., we expected that perceived morality would drive the effect on negative emotions).

Method

Participants

In all, 109 Spanish undergraduates participated in exchange for course credit. After eliminating those who belonged to one of the two minorities involved in the manipulations (i.e., gay people or immigrants), 97 participants (76 women and 21 men, M age = 33.2, SD = 10.0) remained for the analysis.

Procedure

We informed participants that the goal of the study was to assess their opinions about the results of a study that had investigated aspects of gay people’s life experiences, including discrimination and the psychological consequences that discrimination has for them. All participants read that the results of the study showed that gay people in Spain suffer high levels of discrimination in crucial life domains including employment, housing, and health. Participants were then randomly assigned to either a low or high psychological consequences condition. Participants learned that harm or benefits are derived by gay people experiencing discrimination:

The results of the study clearly showed that, in general, suffering discrimination has made homosexuals psychologically weaker [vs. stronger]. As a consequence of experiencing discrimination, the majority of homosexuals become less [vs. more] effective in dealing with adversity, and they develop a weaker [vs. stronger] character which impairs [vs. helps] them cope with the demands of everyday life.

Participants were then told that another finding of the study concerned the attitudes that gay people hold toward immigrants. Participants were first asked to indicate their expectations concerning the attitudes of gay people before we presented the attitude manipulation. After completing the expectations measure, participants were randomly assigned to either a positive or negative attitude condition, and then they completed the morality and negative emotions measures. Participants were then fully debriefed and thanked.

Measures

If not otherwise stated, all items were rated on a 0 (not at all) to 6 (extremely) scale.

Expected attitude

Prior to receiving the manipulation of gay people’s ostensible attitudes toward immigrants, participants were asked to indicate to what extent they expected that attitude to be “Tolerant,” “Biased” (reverse-scored), “Egalitarian,” “Sympathetic,” and “Generous,” alpha = .86.

Perceived morality of target group’s attitude

After learning the attitudes of gay people, participants rated the attitudes toward immigrants held by gay people as “Fair,” “Disloyal” (reverse-scored), “Respectful,” “Immoral” (reverse-scored), and “Indecent” (reverse-scored), alpha = .94.

Negative emotions

Participants indicated how they felt after having learned the target group’s attitudes toward immigrants on each of the following emotions: “Disappointed,” “In a good mood” (reverse-scored), “Sad,” “Cheerful” (reverse-scored), “Angry,” “Uncomfortable,” and “Annoyed,” alpha = .83.

Finally, participants responded to three manipulation checks, consisting of three items each. The first one assessed the benefits manipulation (e.g., “Homosexuals become particularly strong as individuals as a consequence of experiencing discrimination”), alpha = .95. The second assessed whether the manipulation of the group’s attitude was effective (e.g., “The group homosexuals have positive attitudes toward immigrants living in Spain”), alpha = .95. The third assessed the extent to which gay people were perceived as a stigmatized minority (e.g., “The group homosexuals is a disadvantaged minority in Spain”), alpha = .85.

Results

Preliminary analysis

The manipulations were effective. Participants in the high benefits condition agreed more that gay people are stronger as a consequence of experiencing discrimination compared with those in the low benefits condition, M = 4.14, SD = 1.13 versus M = 2.14, SD = 1.71, F(1, 90) = 43.54, p < .001, η² = .33. This measure did not vary by attitude condition, and the benefits × attitude interaction was not significant, Fs < 0.34, ps > .56. Participants in the positive attitude condition agreed more strongly with the view that gay people have positive attitudes toward immigrants compared with those in the negative attitude condition, M = 4.90, SD = 0.90 versus M = 1.03, SD = 1.13, F(3, 90) = 332.72, p < .001, η² = .79. Neither the benefits manipulation nor the interaction significantly affected this measure, Fs < 0.15, ps > .70. The extent to which gay people were perceived as a disadvantaged minority (M = 3.43, SD = 1.33) was significantly above the mean of the scale (i.e., 3), t(93) = 3.12, p < .01. Neither of the manipulated variables nor their interaction significantly affected this measure, Fs < 1.41, ps > .24.

Main analyses

As predicted, participants expected gay people to have more positive attitudes toward immigrants when they believed that gay people are psychologically strengthened as a consequence of experiencing discrimination (M = 4.28, SD = 0.87) compared with when they are weakened (M = 3.76, SD = 1.05), F(1, 92) = 6.77, p < .05, η² = .07.

As expected, the main effect of attitude on perceived morality was significant, F(1, 90) = 251.67, p < .001, η² = .74, with participants perceiving the positive attitude as more moral (M = 5.04, SD = 0.84) than the negative attitude (M = 1.95, SD = 1.09). The main effect of benefits on this variable was not significant, F(1, 90) = 0.78, p = .38, η² = .01, with no significant differences between the high (M = 3.46, SD = 1.97) and the low benefits condition (M = 3.73, SD = 1.67). Central to our hypothesis was the significant Benefits × Attitude interaction on perceived morality, F(1, 90) = 6.07, p < .05, η² = .06. When gay people had a negative attitude but had been strengthened by discrimination, participants rated the group’s attitude as particularly immoral (M = 1.63, SD = 0.89), significantly more immoral than when gay people had a negative attitude, but had been weakened by experiencing discrimination (M = 2.29, SD = 1.19), t(42) = 2.06, p < .05. When, however, gay people had a positive attitude, there was no significant difference in the extent to which the group’s attitude was perceived as moral; in this case, the low and high benefits conditions resulted in high perceived morality, M = 4.89, SD = 0.92 versus M = 5.20, SD = 0.73, t(48) = 1.39, p = .20.

An identical pattern of results was obtained for negative emotions. The main effect of the attitude manipulation was significant, F(1, 90) = 158.22, p < .001, η² = .64, with participants reporting less negative emotions when gay people had positive attitudes (M = 1.08, SD = 0.85) than when they had negative attitudes (M = 3.59, SD = 1.11). The main effect of benefits was not significant, F(1, 90) = 1.80, p = .18, η² = .02, with no difference between the low benefits (M = 2.09, SD = 1.47) and high benefits conditions (M = 2.43, SD = 1.70). However, the expected Benefits × Attitude interaction was significant, F(1, 90) = 5.10, p < .05, η² = .05. When participants believed that gay people became stronger as a consequence of experiencing discrimination but they displayed negative attitudes toward immigrants, negative emotions were evoked most strongly (M = 3.93, SD = 0.99), significantly more than when participants were told that experiencing discrimination makes gay people weaker and their attitudes toward immigrants were negative (M = 3.22, SD = 1.41), t(42) = 2.22, p < .05. When gay people had positive attitudes toward immigrants, there was no significant difference in negative emotions across the low and high benefits conditions, M = 1.17, SD = 1.00 versus M = 0.99, SD = 0.65, t(48) = 0.76, p = .45.

Using the PROCESS macro (Hayes, 2012, Model 8), we tested whether, as hypothesized, morality mediated the Benefits × Attitude interaction on negative emotions. As shown in Figure 2, when controlling for the main effects of both manipulations, the Benefits × Attitude interaction term had a significant effect on perceived morality, which in turn reduced negative emotions. Indicating mediated moderation, the direct effect of the Benefits × Attitude interaction term on negative emotions became non-significant when the mediator was included in the model and the corresponding indirect effect was significant, b = .69, 95% CI = [0.132, 1.347], n boots = 5,000.

Figure 2.

Experiment 4 test for mediated moderation, indicating that the Benefits × Attitude interaction effect on negative emotions was mediated by perceived morality of the target group’s attitude.

Discussion

Experiment 4 demonstrates that when majority group members believe that discrimination has positive psychological consequences for gay people, they expect gay people to have more tolerant attitudes toward immigrants compared with when they believe that discrimination has negative psychological consequences. This result confirms that believing that victims of stigmatized groups derive psychological benefits as a consequence of experiencing discrimination leads perceivers to expect more moral behavior from them. Indeed, disconfirmation of this premise (i.e., when stigmatized groups are thought to obtain benefits as a consequence of experiencing discrimination, but they lack the expected moral virtue) results in particularly harsh moral judgments and negative emotional reactions by majority group members. Moreover, the high level of negative emotions reported by participants when the stigmatized group gained benefits but displayed intolerance was driven by participants’ judgment that the stigmatized group is immoral.

General Discussion

Together, these experiments provide evidence of the existence of a subtle consequence of being a member of a stigmatized group—being judged according to a demanding moral standard. Extending the findings beyond groups that have suffered genocide in the past (Branscombe et al., 2013; Warner & Branscombe, 2012), we found that stigmatized minorities are expected to be especially tolerant toward another disadvantaged group. When members of stigmatized minorities do not fulfill this expectation and instead display intolerance toward another disadvantaged group, then majority group members judge them especially harshly as compared with non-stigmatized groups holding the same negative attitudes. In line with the idea that the HMO placed on stigmatized groups result from observers’ need to believe that victims should become better persons, our results show that the moral demands on the stigmatized are greater if the group has overcome its past victimization relative to when it has not. We argue that the process of overcoming suffering is perceived as giving stigmatized group members’ the opportunity to derive psychological benefits and, as a result, they are expected to emerge as better persons. Moreover, when we manipulated participants’ beliefs so that they believed that members of a stigmatized group become stronger as a consequence of experiencing discrimination, stigmatized group members were held to a higher moral standard than non-stigmatized group members and were judged more harshly if they violated that expectancy by exhibiting negative attitudes. The fact that these results were obtained with two different types of stigmatized groups—a physical disability in the case of people with dwarfism and a sexual orientation minority in the case of gay people—whose socially devalued characteristic is perceived to differ on a number of dimensions (i.e., visibility, concealability, and perhaps even controllability)—speaks in favor of the generalizability of the process we investigated.

Making Meaning of Social Stigmatization as a Justification of Inequality

Expecting stigmatized group members to gain psychological benefits in the form of becoming stronger and developing more moral qualities because of the negative experiences associated with being a member of a disadvantaged group underlies the higher moral expectations placed on them. This process can be understood as a means of dealing with the threat that undeserved suffering can pose to observers’ belief in a just world (Branscombe et al., 2013). By bestowing meaning on the lives of those who have suffered, observers manage the threat that exposure to innocent victimhood can evoke (Anderson et al., 2010). Thus, expecting stigmatized groups to be better than the non-stigmatized can be seen as a form of coping with threat to justice on the part of majority group members. This form of coping is different than those already identified in the literature such as the well-known tendency to derogate and blame victims for their suffering (Aguiar, Vala, Correia, & Pereira, 2008; Correia, & Vala, 2003) or to dehumanize them (Leyens et al., 2001). The current work illustrates how observers who are confronted with innocent suffering do not derogate victims if they can be seen as overcoming their tragic experiences or deriving benefits from them (Lerner, 1980)—unless those victims appear to subsequently act immorally.

Expected Versus Actual Attitudes of Stigmatized Minorities Toward Each Other

While our research examined majority group expectations concerning the attitudes that stigmatized groups should exhibit toward other minority groups, other recent research has assessed the attitudes that members of different stigmatized minorities actually do hold toward other stigmatized groups (Craig, DeHart, Richeson, & Fiedorowicz, 2012; Craig & Richeson, 2012). This research has illustrated that stigmatized minorities can sometimes exhibit either positive or negative attitudes toward other low-status groups. For example, Craig et al. (2012) found that women hold negative attitudes toward other stigmatized minorities (ethnic minorities in the United States) when the discrimination women have experienced was made salient. A similar effect was found by Wohl and Branscombe (2008) with non-stigmatized groups (Jews and Americans) when they were reminded of their group’s previous suffering; members of those groups legitimized their current harm doing toward a new adversary and felt less guilt for doing so. But stigmatized minorities have also been found to display positive attitudes toward other stigmatized groups; Craig and Richeson (2012) found that members of ethnic minorities reacted more positively toward another ethnic minority when the shared nature of their suffering (based on similar forms of discrimination) was salient. An important issue that arises when comparing our findings with investigations addressing the attitudes that stigmatized minorities do actually hold toward each other is the underlying processes that shape those attitudes (both the actual attitudes displayed and those expected by majority group members). In this regard, it is important to note that the effects reported in the present research of harsher moral judgments when minorities fail to meet expectancies stem from the threat to justice experienced by majority group members, whereas Craig and Richeson’s positive attitudes among stigmatized minorities are due to increased perception of similarity among disadvantaged minorities who have suffered the same type of discrimination.

The HMO effect, where stigmatized groups are judged according to a harsher moral standard than the non-stigmatized, suggests a secondary victimization process on the part of observers that is motivated by observers’ need to manage the threat that the suffering by stigmatized groups poses for majority group members. In this sense, expecting more from victimized groups is a hidden burden for non-privileged groups in society, where they not only face discrimination but are also expected to be more moral in their treatment of others and are judged more severely when they violate these expectations—all of which is in the service of protecting privileged group members’ well-being. Chen and Tyler (2001) describe a related process operating in the context of admission processes to elite universities, where legitimizing myths serve to protect privileged group members’ psychological well-being when they take advantage of their privileged position to ensure their own admissions while excluding the non-privileged. In this case, advantaged groups protect their self-esteem and their perception of justice without having to renounce their unfair privileges. Likewise, the phenomenon of holding stigmatized groups to a higher moral standard of conduct occurs when majority group members believe that stigmatized group members have obtained something as a result of their suffering—that of being better persons. Our results show that victimized groups are expected to become better moral people if they have had the opportunity to overcome their suffering, which serves the interests of majority group members.

Limitations and Future Directions

In the present research, we measured and kept constant across all experiments, expected tolerance toward immigrants, while assessing how different stigmatized groups are held to high moral standards of conduct. Negative attitudes toward immigrants have been associated with a general tendency toward intolerance and authoritarian ideological attitudes (Cohrs & Stelzl, 2010). However, expected tolerance toward immigrants is only one specific case in which moral expectations may be expressed. In future studies, it would be useful to consider whether stigmatized minorities are expected to be tolerant toward other disadvantaged groups besides immigrants. Moreover, it would be useful to determine whether stigmatized minorities are expected to adhere to other types of social norms and not just those related to the treatment of disadvantaged groups. Likewise, future research might examine other consequential contexts where higher moral expectations might be placed more readily on stigmatized groups. For example, members of stigmatized groups who attain decision-making positions in the workplace might be expected to support policies that can be seen as reflecting concern for other disadvantaged groups (e.g., support for affirmative action policies). When they fail to make such expected “moral” choices, as indicated by our studies, harsher moral judgments may be made that have negative consequences for disadvantaged group members’ ability to climb further up the corporate ladder.

Conclusion

The present research extends previous theorizing concerning the moral expectations of victims and illustrates its applicability to two rather different types of stigmatized groups. Our findings suggest that being held to a higher moral standard is an extra burden placed on stigmatized group members that reflects a process of meaning making on the part of observers. By associating the suffering caused by social stigmatization with a process of psychological growth and moral gains, majority group members find meaning in the suffering of stigmatized groups. This meaning (in the form of deriving psychological benefits from the suffering of the stigmatized) facilitates majority group members’ perception that the world is just. Yet, when stigmatized groups fail to seemingly derive the expected moral benefits from their suffering and instead violate expectations by being intolerant toward others who are disadvantaged, observers feel justified in their negative responses toward the stigmatized.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received the following financial support for the research, authorship, and/or publication of this article: This research was supported in part by the ALPE-Achondroplasia Foundation, “la Caixa” Foundation, and the Regional Ministry of Social Welfare and Equality of the Principality of Asturias, Spain.

Notes

References

Aguiar

Vala

Correia

Pereira

(2008). Justice in our world and in that of others: Belief in a just world and reactions to victims. Social Justice Research, 21, 50-68. doi:10.1007/s11211-007-0059-3

Anderson

J. E.

Kay

A. C.

Fitzsimons

G. M.

(2010). In search of the silver lining: The justice motive fosters perceptions of benefits in the later lives of tragedy victims. Psychological Science, 21, 1599-1604. doi:10.1177/0956797610386620

Branscombe

N. R.

Warner

R. H.

Klar

Fernández

(2013). “They should know better”: Why third-party observers assign moral obligations to victimized groups. Manuscript submitted for publication.

Chen

E. S.

Tyler

T. R.

(2001). Cloaking power: Legitimizing myths and the psychology of the advantaged. In Lee-Chai

A. Y.

Bargh

J. A.

(Eds.), The use and abuse of power: Multiple perspectives on the causes of corruption (pp. 241-261). New York, NY: Psychology Press.

Cohrs

Stelzl

(2010). How ideological attitudes predict host society members’ attitudes toward immigrants: Exploring cross national differences. Journal of Social Issues, 66, 673-694. doi:10.1111/j.1540-4560.2010.01670.x

Correia

Vala

(2003). When will a victim be secondarily victimized? The effect of observer’s belief in a just world, victim’s innocence and persistence of suffering. Social Justice Research, 16, 379-400. doi:10.1023/A:1026313716185

Cottrell

C. A.

Neuberg

S. L.

(2005). Different emotional reactions to different groups: A sociofunctional threat-based approach to “prejudice.” Journal of Personality and Social Psychology, 88, 770-789. doi:10.1037/0022-3514.88.5.770

Craig

M. A.

DeHart

Richeson

J. A.

Fiedorowicz

(2012). Do unto others as others have done unto you? Perceiving sexism influences women’s evaluations of stigmatized racial groups. Personality and Social Psychology Bulletin, 38, 1107-1119. doi:10.1177/0146167212445210

Craig

M. A.

Richeson

J. A.

(2012). Coalition or derogation? How perceived discrimination influences intraminority intergroup relations. Journal of Personality and Social Psychology, 102, 759-777. doi:10.1037/a0026481

10.

Eberhardt

J. L.

Goff

Purdie

V. J.

Davies

P. G.

(2004). Seeing Black: Race, crime, and visual processing. Journal of Personality and Social Psychology, 87, 876-893. doi:10.1037/0022-3514.87.6.876

11.

Gouvier

Sytsma-Jordan

Mayville

(2003). Patterns of discrimination in hiring job applicants with disabilities: The role of disability type, job complexity, and public contact. Rehabilitation Psychology, 48, 175-181. doi:10.1037/0090-5550.48.3.175

12.

Hayes

A. F.

(2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from http://www.afhayes.com/public/process2012.pdf

13.

Lerner

M. J.

(1980). The belief in a just world: A fundamental delusion. New York, NY: Plenum.

14.

Leyens

Rodriguez-Perez

Rodriguez-Torres

Gaunt

Paladino

Vaes

Demoulin

(2001). Psychological essentialism and the differential attribution of uniquely human emotions to ingroups and outgroups. European Journal of Social Psychology, 31, 395-411. doi:10.1002/ejsp.50

15.

Major

O’Brien

L. T.

(2005). The social psychology of stigma. Annual Review of Psychology, 56, 393-421. doi:10.1146/annurev.psych.56.091103.070137

16.

Schmitt

M. T.

Branscombe

N. R.

(2002). The meaning and consequences of perceived discrimination in disadvantaged and privileged social groups. European Review of Social Psychology, 12, 167-199.

17.

Spencer

S. J.

Zanna

M. P.

Fong

G. T.

(2005). Establishing a causal chain: Why experiments are often more effective than mediational analyses in examining psychological processes. Journal of Personality and Social Psychology, 89, 845-851. doi:10.1037/0022-3514.89.6.845

18.

Truxillo

Bauer

Sanchez

(2001). Multiple dimensions of procedural justice: Longitudinal effects on selection system fairness and test-taking self-efficacy. International Journal of Selection and Assessment, 9, 336-349. doi:10.1111/1468-2389.00185

19.

Warner

R. H.

Branscombe

N. R.

(2011). Observers’ benefit finding for victims: Consequences for perceived moral obligations. European Journal of Social Psychology, 41, 241-253. doi:10.1002/ejsp.772

20.

Warner

R. H.

Branscombe

N. R.

(2012). Observer perceptions of moral obligations in groups with a history of victimization. Personality and Social Psychology Bulletin, 38, 882-894. doi:10.1177/0146167212439212

21.

Warner

R. H.

Branscombe

N. R.

Garczynski

A. M.

Solomon

E. D.

(2011). Judgments of sexual abuse victims. Basic and Applied Social Psychology, 33, 207-219. doi:10.1080/01973533.2011.589294

22.

Wohl

M. J. A.

Branscombe

N. R.

(2008). Remembering historical victimization: Collective guilt for current ingroup transgressions. Journal of Personality and Social Psychology, 94, 988-1006. doi:10.1037/0022-3514.94.6.988