The different roles of relative ingroup prototypicality in the outgroup attitudes of majority and minority groups

Abstract

Based on the argument that the ingroup projection model may not be applicable to the minority group when addressing the effect of relative ingroup prototypicality (RIP) on outgroup attitudes, two studies investigated whether RIP and its effects on outgroup attitudes differ for the majority (Han) and an ethnic minority group (Tibetan). We measured RIP and outgroup attitudes in Study 1 (N = 164) and manipulated RIP in Study 2 (N = 145). The results indicated that the Hans presented high RIP, whereas the Tibetans presented low RIP. The effects of RIP on outgroup attitudes were moderated by group size: High RIP among Hans resulted in negative outgroup attitudes, whereas high RIP among Tibetans led to positive outgroup attitudes. These findings imply that improving the minority group’s RIP by making its culture prototypicality equal to that of the majority group would lead to positive outgroup attitudes.

Keywords

majority and minority groups outgroup attitudes outgroup projection relative ingroup prototypicality

In many multiethnic countries, cultural diversity is becoming salient. For example, China has 56 ethnicities and an abundant diversity of cultures. Tibetans, as members of one ethnic minority group in China, have their own distinctive culture that differs from that of the Hans (the ethnic majority group in China), including a unique religious faith (Tibetan Buddhism), a separate language (Tibetan), and a distinctive way of life. Although cultural diversity potentially leads to a more dynamic and interesting country, it is accompanied by the risk that people will categorize others in terms of their ethnicities and perceive groups competitively (Putnam, 2007). This social categorization process (e.g., “ingroup” vs. “outgroup”) serves as the psychological basis for intergroup conflicts (Tajfel & Turner, 1979). Although the common ingroup identity model (Gaertner & Dovidio, 2000) proposed that identification with a higher order category could lead to positive intergroup attitudes (e.g., Gaertner & Dovidio, 2009), the superordinate category can also provide “a common reference frame” regarding the norms and standards according to which both groups are compared (Kessler & Mummendey, 2009).

According to the ingroup projection model (Mummendey & Wenzel, 1999), intersubgroup comparison results in subgroups exhibiting ingroup projection and high relative ingroup prototypicality (RIP). That is, a subgroup is eager to declare its present ingroup as more prototypical of a given superordinate group than other subgroups (Wenzel, Mummendey, & Waldzus, 2007). However, the RIP could be restrained by social reality (Waldzus, Mummendey, Wenzel, & Boettcher, 2004). In this case, the minority group may have to represent low RIP and outgroup projection because of its relatively small group size. That is, it may declare that the majority group is more prototypical of the superordinate group than its own ingroup.

Many studies have found that a low RIP is associated with positive outgroup attitudes according to the ingroup projection model (e.g., Kessler et al., 2010). However, little research has examined whether this positive effect of low RIP is applicable to the minority group. Based on the arguments of Sindic and Reicher (2008), the minority group’s low RIP and outgroup projection may be not only a reflection of the social reality but also the consequences of the instrumental approach to achieving ingroup interests. The outgroup projection may allow minority groups to proclaim that their ingroup characteristics are not represented well within the superordinate group. Minority groups’ desire to achieve positive distinctive ingroup identification may not be satisfied because relative prototypicality is the equivalent of positive distinctiveness (Turner, 1987). When minority groups believe that the superordinate group mostly reflects the characteristics of the majority group, they may encounter distinctiveness threat and favor independence from the majority group (Reicher, 2004; Wohl, Giguère, Branscombe, & Mcvicar, 2011). As a result, low RIP among the minority group may be associated with negative outgroup attitudes, which may be in contrast to the effect of low RIP for the majority group proposed by the ingroup projection model. The purpose of the present study is to explore whether RIP and its effects on outgroup attitudes differ for majority and minority groups.

Majority- and Minority-Group Members’ Perceptions of Relative Ingroup Prototypicality

Majority-group members are likely to perceive that their own group is the most prototypical, that is, representing high RIP and ingroup projection, not only because of ingroup projection but also because this perception is consistent with the social reality (Ullrich, Christ, & Schlüter, 2006; Waldzus, Mummendey, Wenzel, & Weber, 2003; Wenzel et al., 2007; Wohl et al., 2011). For instance, West Germans (the majority) reported a greater sense of prototypicality as Germans than did East Germans (the minority; Waldzus et al., 2004).

Minority groups, however, may be less prone to claim high RIP because of their relatively small group size (Wenzel et al., 2007). Therefore, minority groups may claim low RIP and outgroup projection (e.g., Waldzus et al., 2004). For instance, African Americans presented lower RIP than Caucasian Americans did (Devos & Banaji, 2005). However, Ng Tseung-Wong and Verkuyten (2010) found that Creoles, members of the minority group in Mauritius (with Hindus representing the majority group), showed high RIP because the language commonly used by Mauritians was largely inherited from the Creoles. This study implied that the extent to which the minority culture places value on its representativeness of the superordinate group seems to affect the minority group’s RIP beyond group size. Therefore, when the minority culture is well represented by the superordinate group (e.g., the minority group has equal culture status with the majority group), the minority group may present high RIP.

Relative Ingroup Prototypicality and the Outgroup Attitudes of Majority and Minority Groups

For the majority group, high RIP helps ingroup members achieve a positive identity (Mummendey & Wenzel, 1999), legitimizes ingroup superiority status and entitlement to the largest share of resources (Peker, Crisp, & Hogg, 2010), promotes inequality among subgroups, and establishes the inferiority of other subgroups (Ullrich et al., 2006; Waldzus et al., 2003). Therefore, majority-group members’ high RIP leads to negative outgroup attitudes (Kessler et al., 2010; Waldzus, Mummendey, & Wenzel, 2005; Waldzus et al., 2004; Waldzus et al., 2003; Wenzel et al., 2007). By contrast, low RIP may reflect that the majority group positively values the characteristics of the minority group. Indeed, previous studies have found that only outgroup prototypicality is related to the majority group’s positive outgroup attitudes (Ufkes, Otten, van Der Zee, Giebels, & Dovidio, 2012; Ullrich, 2009; Wenzel et al., 2007). Therefore, for the majority group, the minority group’s prototypicality (i.e., outgroup prototypicality) may be related to positive outgroup attitudes.

In addition, although Waldzus and Mummendey (2002) found a causal effect of RIP on intergroup bias, the sizes of the ingroup (Germans) and outgroup (British) were similar within the superordinate group (European). Most previous studies focusing on the majority group have provided only correlative evidence of the effect of RIP because successfully manipulating RIP independent from existing group membership is difficult (Weber, Mummendey, & Waldzus, 2002). Therefore, evidence confirming the causal effect of RIP on outgroup attitude remains lacking.

Studies focusing on the effect of RIP on minority groups’ outgroup attitudes are scarce. Although some researchers have found that a lower RIP was associated with Scottish participants’ higher levels of proindependence from Britain when the issue of whether Scotland’s interests were properly represented within Britain was made salient (Sindic & Reicher, 2008), the relationship between RIP and general outgroup attitudes remains unclear because independence is not the only indicator of the relationship between the majority and minority groups. We presumed that the low RIP or outgroup projection of the minority group may result in negative outgroup attitudes. According to Sindic and Reicher (2008), outgroup projection may be explained by an instrumental approach to achieving desired goals. Huynh, Devos, and Altman (2015) found that Asian Americans and Latinos proclaimed that their groups deserved the rights and benefits afforded to all Americans when they showed low RIP. People may stress outgroup superiority to highlight the inequality of the status quo (van Knippenberg & van Oers, 1984). Thus, outgroup projection may become the means by which the minority group can proclaim the following: “this group reflects our rivals; we will always be badly served within it” (Sindic & Reicher, 2008, p. 1426). In particular, when the superordinate group mostly reflects the characteristics of the majority group, minority groups may claim that their ingroup interests, such as achieving a distinctive ingroup identity, are not properly represented within the superordinate group. Thus, the minority group may experience a distinctiveness threat (Reicher, 2004; Wohl et al., 2011) and support independence to preserve the group’s distinctive “way of life,” even at the expense of wealth (Hornsey & Hogg, 1999; Leonardelli, Pickett, & Brewer, 2010; Sindic & Reicher, 2009).

By contrast, a minority group’s high RIP may lead to positive outgroup attitudes. The high RIP may imply that the minority group could have an equal opportunity to project ingroup characteristics to the superordinate group as the majority group, which could enrich and complicate the representation of the superordinate category. Because relative prototypicality is the equivalent of positive distinctiveness (Turner, 1987), the high relative prototypicality of minority groups could make them achieve positive distinctiveness with regard to their ingroup identity. Minority-group members’ (e.g., Asian Americans) greater RIP has been linked to support for ethnic minority language policy or acculturation ideologies (e.g., opposition to assimilation, support for multiculturalism) (Huynh et al., 2015). High RIP could also help minority groups to be accepted and valued by the superordinate group. Because minority groups often seek to be respected by majority groups (Bergsieker, Shelton, & Richeson, 2010; van Oudenhoven, Prins, & Buunk, 1998), it seems that minority groups might consider the perspective of the majority group regarding the minority groups’ prototypicality into account (i.e., metaperceptions) more than the majority groups do. The minority groups may hope to have equal prototypicality with majority groups for representation of the superordinate group (i.e., exhibit high RIP). Therefore, high RIP for the minority group may be a condition that allows for some hope of being recognized, accepted, and valued within the superordinate group. Ufkes et al. (2012) found that minority-group superordinate identification was related to positive outgroup attitudes only when the minority group perceived high ingroup prototypicality. Therefore, the minority group’s high RIP and prototypicality may be associated with positive outgroup attitudes.

The Present Research

The aim of the present research was to explore whether RIP and its effects on outgroup attitudes differ for the majority group and an ethnic minority group. We expected that minority group members would exhibit lower RIP than majority-group members and show outgroup projection, whereas majority-group members would exhibit higher RIP than minority-group members and show ingroup projection. We also anticipated that the effect of RIP on outgroup attitudes might be moderated by group size: a high RIP among the majority group could result in negative outgroup attitudes, whereas a high RIP among the minority group could lead to positive outgroup attitudes. The minority group’s prototypicality may be related to the positive outgroup attitudes of both the majority and minority groups.

We chose Tibetans as the minority group and Hans as the majority group within the Chinese superordinate group because, according to China’s national 2010 census (National Bureau of Statistics of the People’s Republic of China, 2011), the Han group is much larger than the Tibetan group in terms of numbers (91.51% of the Chinese population is Han; 0.46% of the Chinese population is Tibetan). Therefore, in the present study, the distinction between the majority group and the minority group refers only to group size.

We tested our hypotheses in two studies. In Study 1, we used scales to measure the RIP of Hans and Tibetans and to determine the relationship between RIP and outgroup attitudes. In Study 2, we manipulated the RIP by changing the degree of representation of Han and Tibetan cultural prototypicality in Chinese culture by creating two fake “culture comments” on online news boards and explored the moderator effect of group size on the relationship between RIP and outgroup attitudes.

Study 1

Participants and Procedure

The participants were 164 current students (99 women, 65 men, M_age = 19.65, SD_age = 1.40) at two universities in Beijing, China. Eighty-eight participants self-identified as Han, and 76 self-identified as Tibetan. The Han participants were recruited from various campus locations of one university, and the Tibetan participants were recruited via Tibetan clubs at another university. All of the participants completed a questionnaire containing measures of group prototypicality, identification with the ingroup and the superordinate group, outgroup attitudes, and socially desirable responding. After they completed the questionnaire, participants were given gifts in thanks and debriefed.

We conducted a quasi-experiment design (two group sizes: Han, Tibetan) with ratings of prototypicality and outgroup attitudes as dependent variables. Socially desirable responding was a control variable because the relationship between the Han and Tibetan groups is very sensitive.

Measures

Group prototypicality and RIP

We measured participants’ perceived prototypicality for two subgroups (Han, Tibetan) within the Chinese. To this end, a pictorial measure, which was developed by Schubert and Otten (2002) and adapted as a measure of prototypicality by Waldzus and Mummendey (2004), was used to assess the degree to which the ingroup and outgroup were seen as close to (i.e., representation) the superordinate group. This measure consists of two sets of pictures, each with seven pairs of circles varying in their degree of overlap and numbered from 1 (the lowest perceived prototypicality) to 7 (the highest perceived prototypicality). The smaller circle represents the ingroup and the larger circle represents the superordinate group. The same set of circles was presented for the outgroup and the superordinate group. The participants were asked to indicate what pair of circles best represented the typicality of the ingroup (ingroup prototypicality) and the outgroup (outgroup prototypicality) with respect to the Chinese. In line with Kessler et al.’s (2010) study, we computed the RIP as a quotient by dividing ingroup prototypicality by outgroup prototypicality. If the RIP is significantly higher than 1, then the subgroup exhibits ingroup projection, whereas if the RIP is significantly lower than 1, then the subgroup presents outgroup projection.

Because these ratio measures resulted in skewed distributions (for the Han sample, skewness = 3.39, SE = 0.26; kurtosis = 12.10, SE = 0.51; for the Tibetan sample, skewness = −0.65, SE = 0.28; kurtosis = −0.33, SE = 0.55), we normalized the RIP scores using the proportion estimation formula (Qiu, 2013), that is, Rankit’s formula¹ (for the Han sample, skewness = 0.14, SE = 0.26; kurtosis = −0.34, SE = 0.51; for the Tibetan sample, skewness = −0.16, SE = 0.28; kurtosis = −0.48, SE = 0.55). Normalized RIPs were entered into the analyses in what follows.

Outgroup attitudes

Attitudes toward the outgroup were measured using an 11-item scale (Waldzus et al., 2003). The items formed four subconcepts, namely, sympathy (e.g., “I like the [outgroup] mentality”), willingness to engage in intergroup contact (e.g., “I think it is important to be in contact with [the outgroup]”), favorable behavioral intentions toward the outgroup (e.g., “If my financial situation permitted, I would donate money to the Society for [the ingroup]/[the outgroup] friendship”), and evaluation of the outgroup’s difference (e.g., “I can easily accept those features that distinguish most [outgroup] from [ingroup]”). Principal component analyses revealed a clear one-factor structure (eigenvalue: 5.05), which explained 45.92% of the variance, and the loadings of all 11 items were greater than 0.47. Thus, it seemed appropriate to use all items as indicators of unitary attitudes toward the outgroup by averaging the responses across items. The reliability of this scale was good (Cronbach’s α = .88). Higher values indicate more positive outgroup attitudes.

Identification with the ingroup and superordinate group

Given that dual identification is often considered a necessary condition for ingroup projection (Wenzel, Mummendey, Weber, & Waldzus, 2003), identification with the ingroup (Cronbach’s α = .63) and with the superordinate group (Cronbach’s α = .85) was assessed using two scales that contained the same five items (Bianchi, Mummendey, Steffens, & Yzerbyt, 2010) but differed in terms of the target group (e.g., “Identify with the [ingroup]/[superordinate group]”).

Socially desirable responding (SDR)

We assessed the socially desirable responding via 16 items (socially desirable responding regarding attitudes and opinions; Schuessler, Hittle, & Cardascia, 1978), such as, “I find that I can help others in many ways,” Cronbach’s α = .78.

All of the variables (outgroup attitudes, identification with the ingroup and superordinate group, SDR) were measured on a 7-point scale (1 = do not agree at all to 7 = agree completely).

Results

Different identification with the ingroup and superordinate group

For the Han and Tibetan participants, both the ratings of ingroup and superordinate group identification were significantly above the scale midpoint of 4, ts(87) > 20.80, ps < .001; ts(75) > 7.10, ps < .001. Thus, both the Hans and Tibetans were dual identifiers. The Han participants reported greater identification with the superordinate group (M = 6.36, SD = 0.75) than the ingroup (M = 6.03, SD = 0.91), t(87) = 4.12, p < .001. By contrast, the Tibetan participants reported greater identification with the ingroup (M = 6.40, SD = 0.81) than the superordinate group (M = 5.03, SD = 1.26), t(75) = 9.19, p < .001.

Different group prototypicality of the Han and Tibetan groups

As predicted, both groups agreed that Hans were more prototypical of the Chinese than Tibetans were (Figure 1). In a 2 (group size: Han, Tibetan) × 2 (target group: Han, Tibetan) repeated-measures generalized linear model (GLM) for scores of close similarity to the Chinese in general, superordinate group identification was included as a covariate, F(1, 161) = 7.64, p = .006, η_p² = .05. The means and standard deviations of all variables are shown in Table 1. We found a significant main effect of target group, F(1, 161) = 59.98, p < .001, η_p² = .27. Hans were regarded as closer to, and thus more prototypical of, the Chinese (M = 6.10, SD = 1.13) than Tibetans were (M = 4.46, SD = 1.76). However, this effect was moderated by a significant interaction effect, F(1, 161) = 4.54, p = .035, η_p² = .03, indicating that the two subgroups differed in their views of how relatively prototypical they were of the superordinate group. More precisely, Tibetan participants perceived Hans to be closer to (i.e., more prototypical of) the Chinese (M = 6.42, SD = 0.82) than the Han participants did (M = 5.82, SD = 1.27), F(1, 161) = 5.90, p = .016, η_p² = .04. Furthermore, Tibetan participants also perceived Tibetans to be more prototypical of the Chinese (M = 4.64, SD = 1.80) than the Han participants did (M = 4.30, SD = 1.71), F(1, 161) = 12.99, p < .001, η_p² = .08. We calculated the difference between Han and Tibetan prototypicalities in general. Although this analysis is redundant to the ANOVA, it is more illuminating. The results indicated that the target group differences were slightly larger for the Tibetan group than for the Han group, M = 1.78, SD = 1.20; M = 1.52, SD = 1.38; F(1, 161) = 4.54, p = .035, η_p²= .03.

Table 1.

Intercorrelations between RIP, Han and Tibetan prototypicality, outgroup attitudes, ingroup and superordinate group identification, and SDR as a function of group size in Study 1.

Measure	1	2	3	4	5	6	7	M	SD
1. RIP (normalized score)	–	−.12	−.83^***	−.31^**	.20⁺	−.03	−.12	0.01	0.95
2. Han prototypicality	−.40^***	–	.61^***	−.08	.07	−.18⁺	−.11	5.82	1.27
3. Tibetan prototypicality	.90^***	−.02	–	.23^*	−.11	−.10	.06	4.30	1.71
4. Outgroup attitudes	.53^***	−.03	.59^***	–	.02	.17	.22^*	5.08	0.93
5. Ingroup identification	.06	.16	.14	.25^*	–	.58^***	.30^**	6.03	0.91
6. Chinese identification	.59^***	.02	.63^***	.71^***	.28^*	–	.46^***	6.36	0.75
7. SDR	.29^*	.03	.30^**	.21⁺	.25^*	.37^**	–	5.15	0.80
M	−.02	6.42	4.64	4.61	6.40	5.03	5.00
SD	0.89	0.82	1.80	1.04	0.81	1.26	0.63

Note. Intercorrelations for the Han participants (n = 88) are presented above the diagonal, and intercorrelations for the Tibetan participants (n = 76) are presented below the diagonal. The means and standard deviations for the Han participants are presented in columns, and the means and standard deviations for the Tibetan participants are presented in rows. RIP = relative ingroup prototypicality, normalized RIP score; SDR = socially desirable responding.

.05 < ⁺p < .10. ^*p < .05. ^**p < .01. ^***p < .001.

Furthermore, the normalized RIP of the Han participants (RIP = 0.01, SD = 0.95) was significantly higher than the normalized score of 1 (−1.18), t(87) = 11.76, p < .001, whereas the RIP of the Tibetan participants (RIP = −0.02, SD = 0.89) was significantly lower than the normalized score of 1 (0.85), t(75) = −8.51, p < .001.² These results suggest that the Hans demonstrated ingroup projection, whereas the Tibetans demonstrated outgroup projection.

The moderator effect of group size on the relationship between RIP and outgroup attitudes

We next investigated the correlations among RIP, outgroup attitudes, Han and Tibetan prototypicality, ingroup and superordinate group identification, and SDR (Table 1). Consistent with our predictions, the RIP of the Hans was negatively correlated with outgroup attitudes (r = −.31, p < .001). However, the RIP of the Tibetans was positively correlated with outgroup attitudes (r = .53, p < .001). Moreover, we tested the moderator effect of group size using a regression model with the standardized scores of RIP, group size (Han coded as 1, Tibetan as 0), and the hypothesized interaction term as predictors. The dependent variable was outgroup attitudes, and SDR was controlled, β = .24, t(162) = 3.04, p = .003. There was not a significant main effect of RIP, β = .09, t(160) = 1.09, p = .279, but there was a significant main effect of group size, β = .42, t(160) = 2.80, p = .006, demonstrating that the Han participants (M = 5.08, SD = 0.93) showed more positive outgroup attitudes than the Tibetan participants (M = 4.61, SD = 1.04). As previously noted, the RIP × Group Size interaction was significant, β = −.86, t(159) = −5.59, p < .001, ΔR² = .15. Furthermore, we tested the simple effect of RIP for the Han and Tibetan groups separately using two regression models. In line with our hypothesis, after controlling for the effects of SDR, for the Hans, β = .19, t(86) = 2.13, p = .036, ΔR² = .05; for the Tibetans, β = .26, t(74) = 1.87, p = .066, ΔR² = .05, there was a significant negative relationship between RIP and outgroup attitudes among the Hans, β = −.28, t(85) = −2.77, p = .007, ΔR² =.08. For the Tibetans, however, the relationship between RIP and outgroup attitudes was significantly positive, β = .59, t(73) = 4.97, p < .001, ΔR² = .24 (Figure 2).

The different effects of group prototypicalities on outgroup attitudes

Subsequently, we explored the relationship between Han and Tibetan prototypicality and outgroup attitudes. As shown by the correlations presented in Table 1, for both the Han and Tibetan participants, only Tibetan prototypicality was positively related to outgroup attitudes, for the Hans, r = .23, p = .029; for the Tibetans, r = .59, p < .001.

Moreover, considering the interaction effect of superordinate group identification and ingroup prototypicality on outgroup attitudes (Ufkes et al., 2012), we also tested the interaction effect of superordinate group identification and the Han and Tibetan prototypicality on outgroup attitudes. We ran a regression model with group size (Han coded as 1 vs. Tibetan coded as 0), standardized scores of superordinate group identification, Han and Tibetan prototypicality, and the two hypothesized interaction terms as predictors. We used outgroup attitudes as the dependent variable and SDR as the covariate. The results showed that there was a significant three-way interaction of Identification x Ingroup Prototypicality x Group Size, β = −.26, t(154) = −2.55, p = .012. There was also a significant three-way interaction of Identification x Outgroup Prototypicality x Group Size, β = .53, t(154) = 2.87, p = .005. We then performed separate regressions for the Han and Tibetan participants. However, the results showed that for the Han participants, neither the interaction effect of Han prototypicality and superordinate group identification nor the interaction effect of Tibetan prototypicality and superordinate group was significant, β = −.09, t(81) = −0.31, p = .759; β = .37, t(81) = 1.31, p = .194. The same results were found for the Tibetan participants, β = −.11, t(69) = −1.44, p = .154; β = .09, t(69) = 0.99, p = .325.

The results of Study 1 suggested that the findings from the regression focused on the Hans are consistent with the ingroup projection model. However, the results also supported the notion that the Tibetan group showed outgroup projection and low RIP, which were associated with negative outgroup attitudes. Moreover, only Tibetan prototypicality was associated with positive outgroup attitudes for both groups. It seemed that RIP had opposite correlations with the outgroup attitudes of the Hans and Tibetans because perceived Tibetan prototypicality reduced the Hans’ RIP and increased the Tibetans’ RIP. Furthermore, for the Tibetan group, the effect of Tibetan prototypicality on outgroup attitudes was not moderated by identification with the superordinate group, which is in contrast to the findings of Ufkes et al.’s (2012) study.

Consistent with the absence of moderation is that the relation between ingroup prototypicality and outgroup attitudes is much stronger (r = .59, p < .001, for the Tibetans) than in the research by Ufkes et al. (2012; r = .09, p > .100, Study 2). Additionally, consistent with previous studies on minority groups (Elkins & Sides, 2007; Staerklé, Sidanius, Green, & Molina, 2010), the current study found that the Tibetans’ superordinate identification was lower than their ingroup identification. Lower superordinate group identification may imply lower ingroup prototypicality. Thus, Tibetans may question whether they are accepted and valued by the superordinate group. Furthermore, the effects of RIP were correlational due to the survey method used and thus do not allow for inferences of causality. Therefore, we manipulated RIP and explored the causal effects of RIP in Study 2.

Study 2

Study 2 aimed to test the contrary causal effects of RIP on the outgroup attitudes of the Han and Tibetan groups. We manipulated RIP and expected that a high RIP among the Hans would result in negative outgroup attitudes, whereas a high RIP among the Tibetans would lead to positive outgroup attitudes.

Participants

The participants were 145 current students (76 women, 69 men, M_age = 19.31, SD_age = 1.63) at two universities. Seventy-five participants who self-identified as Han were recruited from a university in Beijing, China, and 70 participants who self-identified as Tibetan were recruited from a university for nationality in Wuhan, China. Although the participants were from different cities, they were all temporary residents of these cities. Therefore, similar to the participants of Study 1, the participants of Study 2 were from the entire country, including Tibetan areas.

Design and Procedure

The participants were randomly assigned to a prototypicality condition in a 2 (conditions: Hans’ greater prototypicality than or equal prototypicality as Tibetans for the Chinese culture) × 2 (group size: Han or Tibetan) factorial design. The participants were instructed to identify with the ingroup and superordinate group and to respond to the SDR items. Then, they read one of two prototypicality materials, responded to the main ideas of the materials, and completed manipulation check items and a measurement of outgroup attitude. After they completed the experiment, the participants were asked to guess the purpose of the experiment; none of the participants guessed the real purpose. Following these steps, the participants were given ¥15.00 (approximately a good lunch expense at a Chinese university) in thanks and debriefed.

RIP Manipulation

To test the causal impact of RIP on outgroup attitudes, we manipulated RIP based on the argument that subgroup culture prototypicality for the superordinate group culture could affect the subgroup’s RIP beyond group size (Ng Tseung-Wong & Verkuyten, 2010). Moreover, it may be possible for the minority group to claim to be equally prototypical of the superordinate group as the majority group (Ufkes et al., 2012; Weber et al., 2002). In Study 1, we found that 21.6% of Han and 36.8% of Tibetan participants reported equal prototypicality between Hans and Tibetans. These results provided bases for our manipulation of RIP.

Therefore, in the present experiment, for both the Han and Tibetan groups, we manipulated RIP for two conditions by changing the degree to which Han and Tibetan cultures were prototypical of the Chinese culture. In Condition 1, Hans are more prototypical of the Chinese culture than Tibetans (i.e., prototypicality of Han > Tibetan); in Condition 2, Hans are equally prototypical of the Chinese culture as Tibetans (i.e., prototypicality of Han = Tibetan). Overall, for the Han group, Condition 1 was the high RIP condition, and Condition 2 was the low RIP condition. By contrast, for the Tibetan group, the high RIP condition was Condition 2, and the low RIP condition was Condition 1.

Stimulus materials

We created two fake “culture comments” on online news boards on SINA, which is the world’s largest Chinese portal website. The news was depicted as from the “world culture network of England” (the fake website). Both of the culture comments focused on three facets of Chinese culture: language (language characteristics and contributions to Chinese), core culture (cultural traditions and personality characteristics), and customs (festivals and eating habits). We compared Han and Tibetan cultures according to the above mentioned three facets. For the “prototypicality of Han > Tibetan” condition, the news title was “Han culture is the mainstream of Chinese culture,” declaring the dominant influence of Han on Chinese culture. By contrast, for the “prototypicality of Han = Tibetan” condition, the news title was “Han and Tibetan cultures are cultural treasures of Chinese culture,” declaring the equal influence of Hans and Tibetans on Chinese culture. To strengthen the effect of the manipulation, we instructed the participants to cite examples of the cited three facets in the news to illustrate the main ideas.

Measures

Manipulation check items

Given that the indirect and direct measures of RIP were highly correlated (Waldzus et al., 2003), two items directly measured group prototypicality: “To what degree do the Hans [Tibetans] represent the Chinese?” The items were rated on a 7-point scale (1 = cannot represent to 7 = represent completely). The calculation of RIP was the same as that in Study 1. Considering the skewed distribution of RIP, we normalized RIP using the same method employed in Study 1, and normalized RIPs were entered into the following analyses.

One item measured the trustworthiness of the materials on a 7-point scale (1 = do not trust at all to 7 = trust completely). The ratings were significantly above the scale midpoint of 4, indicating that the materials were trusted, M = 4.77, SD = 1.02, t(144) = 9.03, p < .001.

Identification with ingroup and superordinate group

Identification with the ingroup (Cronbach’s α = .77) and with the superordinate group (Cronbach’s α = .81) was assessed using two scales measuring organizational identification. These scales contained the same six items (Mael & Ashforth, 1992) but differed in terms of the target group (e.g., “When someone praises [ingroup]/[superordinate group], it feels like a personal compliment”).

Socially desirable responding (SDR)

We assessed socially desirable responding via 13 items (Marlowe–Crowne Social Desirability Scale; Reynolds, 1982), for example, “There have been times when I was quite jealous of the good fortune of others,” Cronbach’s α = .57). Because SDR was not significantly correlated with outgroup attitudes or group prototypicalities (rs < .18, ps > .113), SDR was not entered into the following analyses.

The items used to measure outgroup attitudes (Cronbach’s α = .87) were the same as those used in Study 1. All of the variables were measured on a 7-point scale (1 = do not agree at all to 7 = agree completely).

Results

Different identification with the ingroup and superordinate groups

As in Study 1, the Han participants’ superordinate identification (M = 4.88, SD = 1.19) was significantly higher than their ingroup identification (M = 4.07, SD = 1.30), t(74) = 7.31, p < .001. By contrast, the Tibetan participants’ ingroup identification (M = 5.06, SD = 0.95) was significantly higher than their superordinate identification (M = 4.74, SD = 1.19), t(69) = 2.49, p = .015. Moreover, the Tibetan participants were dual identifiers because both the ratings of ingroup and superordinate group identification were significantly above the scale midpoint of 4, ts(69) > 5.18, ps < .001. However, the Han participants were more likely to be superordinate identifiers because only the rating of superordinate group identification was significantly above the scale midpoint: superordinate identification, t(74) > 6.39, p < .001; ingroup identification, t(74) = 0.46, p = .648.

Manipulation check

A manipulation check was performed using a 2 (conditions: Hans’ greater prototypicality than or equal prototypicality as Tibetans for the Chinese culture) × 2 (group size: Han or Tibetan) between-subjects univariate analysis of variance. In terms of normalized RIP, the results showed no significant main effect of condition, F(1, 141) = 2.38, p = .126, or of group size, F(1, 141) = 0.08, p = .779. However, the results revealed the expected effect for the interaction of these factors, F(1, 141) = 23.68, p < .001, η_p² = .14. The Han participants reported lower RIP in the prototypicality of Han = Tibetan condition (M = −0.43, SD = 0.62) than in the prototypicality of Han > Tibetan condition (M = 0.50, SD = 0.96), F(1, 73) = 25.14, p < .001, η_p² = .26, even though they showed ingroup projection in both conditions: both RIPs were significantly higher than −1.24 (normalized score of 1), ts > 8.00, ps < .001. In contrast to the Han group, the Tibetan participants reported higher RIP in the prototypicality of Han = Tibetan condition (M = 0.24, SD = 0.76) than in the prototypicality of Han > Tibetan condition (M = −0.24, SD = 1.09), F(1, 68) = 4.59, p = .036, η_p² = .06. In the prototypicality of Han > Tibetan condition, the Tibetan participants exhibited outgroup projection: RIP was significantly lower than 0.23 (normalized score of 1); t(34) = −2.59, p = .014. However, in the prototypicality of Han = Tibetan condition, the Tibetan participants did not display outgroup projection or ingroup projection: RIP was not significantly different from 0.23; t(34) = 0.04, p = .968.

Moreover, in terms of Han and Tibetan prototypicality, the results revealed a significant main effect of condition on Tibetan prototypicality, F(1, 141) = 29.69, p < .001, η_p² = .17. Both the Hans and the Tibetans showed higher Tibetan prototypicality in the prototypicality of Han = Tibetan condition (M = 4.31, SD = 1.34) than in the prototypicality of Han > Tibetan condition (M = 3.20, SD = 1.56). However, for Han prototypicality, there was no significant main effect of condition, F(1, 141) = 0.08, p = .789 (see Table 2). Therefore, the Han and Tibetan participants were significantly influenced by our manipulation only for Tibetan prototypicality.

Table 2.

Means and standard deviations of Han prototypicality and Tibetan prototypicality for both Han and Tibetan groups under two conditions.

Condition	Han group (n = 36)		Tibetan group (n = 35)
Condition	Han prototypicality	Tibetan prototypicality	Han prototypicality	Tibetan prototypicality
Han > Tibetan	5.78 (0.93)	2.28 (1.34)	4.74 (1.25)	4.14 (1.17)
Han = Tibetan	5.51 (0.91)	3.79 (1.36)	4.91 (1.10)	4.89 (1.08)

Note. “Han > Tibetan” is prototypicality of Han > Tibetan condition, “Han = Tibetan” is prototypicality of Han = Tibetan.

The effect of RIP and group size on outgroup attitudes

The dependent measures were analyzed using a 2 (conditions: Hans’ greater prototypicality than or equal prototypicality as Tibetans for the Chinese culture) × 2 (group size: Han or Tibetan) between-subjects univariate analysis of variance. In terms of outgroup attitudes, condition had the expected main effect, F(1, 141) = 10.48, p = .002, η_p² = .07. The participants reported more positive outgroup attitudes in the prototypicality of Han = Tibetan condition (M = 5.34, SD = 1.00) than in the prototypicality of Han > Tibetan condition (M = 4.81, SD = 0.97). The main effect of group size, F(1, 141) = 1.08, p =.302, and interaction effect, F(1, 141) = 0.14, p = .711, were not significant. To test the effect of RIP and group size on outgroup attitudes, we used a 2 (RIP: high or low) × 2 (group size: Han or Tibetan) between-subjects analysis. We coded high RIP as 1 and low RIP as 0. The results showed that the main effect of RIP, F(1, 141) = 0.14, p = .711, and the main effect of group size, F(1, 141) = 1.08, p = .302, were not significant. However, as expected, the RIP × Group Size interaction was significant, F(1, 141) = 10.48, p < .001, η_p² = .07 (see Figure 3). The Han participants reported less positive outgroup attitudes in the high RIP condition (M = 4.93, SD = 0.88) than in the low RIP condition (M = 5.39, SD = 0.97), F(1, 73) = 4.82, p = .031, η_p² = .06. In contrast to the Han participants, the Tibetan participants reported more positive outgroup attitudes in the high RIP condition (M = 5.29, SD = 1.05) than in the low RIP condition (M = 4.69, SD = 1.05), F(1, 68) = 5.59, p = .021, η_p² = .08.

The results of Study 2 indicated that both the majority and minority groups expressed more positive outgroup attitudes when the degree to which they were prototypical of the superordinate group was similar. The results implied that in this condition, the majority group could share their prototypicality and respect the distinctive contribution of the minority group’s prototypicality, whereas the minority group could be respected or gain more status. The results also supported the causal effect of RIP on outgroup attitudes and the moderation of group size on this effect, which is consistent with the results of Study 1.

General Discussion

In two studies, we demonstrated that the majority group showed high RIP and ingroup projection, whereas the minority group demonstrated low RIP and outgroup projection. We also found that the effect of RIP on outgroup attitudes was moderated by group size: a high RIP among the majority group resulted in negative outgroup attitudes, whereas a high RIP among the minority group led to positive outgroup attitudes.

Group Projection for the Majority and Minority Groups

In Study 1, we found that the Han group showed ingroup projection and high RIP, which is consistent with the ingroup projection model (e.g., Wenzel et al., 2007). As expected, the Tibetan group showed outgroup projection and low RIP. Moreover, our results indicated that the Tibetan participants perceived the Hans to be more prototypical of the Chinese and perceived more differences in Han prototypicality relative to Tibetan prototypicality than did the Han participants. These findings differ from the results Waldzus et al. (2004) obtained in a study of Germans. These authors found that the minority group (East Germans) regarded the majority group to be less prototypical and perceived less of a difference in majority prototypicality relative to minority prototypicality than did the majority group (West Germans). West Germans and East Germans may differ in terms of political power and population, but their cultural core characteristics (e.g., language and religious faith) do not differ greatly. In our studies, there were major religious, cultural, and lifestyle differences between the Tibetans and Hans in China; for instance, Tibetan Buddhism was the predominant religious faith of the Tibetans but not the Hans. Therefore, cultural differences may strengthen outgroup projection and the representation of higher outgroup prototypicality. When the superordinate group mostly reflects the cultural characteristics of the Han group, greater differences from the Han culture are associated with higher Han prototypicality represented by the Tibetan group.

To the best of our knowledge, our study may be the first successful research to separately manipulate RIP for majority and minority groups. In contrast to most previous studies that directly measured RIP (e.g., Ullrich et al., 2006; Waldzus et al., 2003), we provided the Tibetan and Han participants with an opportunity to have equal status in Chinese culture through the use of fake “culture comments” on online news boards in Study 2. The Han participants showed low RIP, and the Tibetan participants showed high RIP and did not exhibit outgroup projection under this condition. The explanations for this result are that the complex representation of the superordinate group lowered the RIP of the majority group (Waldzus et al., 2005), and the social identity complexity enhanced positive outgroup attitudes for both the minority and majority groups (Brewer, Gonsalkorale, & van Dommelen, 2013). Both social identity complexity and the complex representation of the superordinate group may imply more equal status among the included groups. The Hans may recognize and value the important effect of the Tibetan’s distinctive culture on Chinese culture, and the Tibetans may achieve more respect and consideration. As Wenzel et al. (2007) presumed, the majority will most likely perceive this as a relative status loss, whereas the minority will likely view it as a status gain.

RIP and Outgroup Attitudes

Our two studies supported the negative relationship between high RIP and the outgroup attitudes of the majority group, which is also consistent with the ingroup projection model (Kessler et al., 2010; Waldzus et al., 2005; Waldzus et al., 2003). Moreover, extending the correlation evidence found in previous research (e.g., Waldzus & Mummendey, 2004), we provided evidence of the causal effect of RIP on the majority group’s outgroup attitudes by manipulating RIP in Study 2.

We also found that Tibetan prototypicality was associated with positive outgroup attitudes for both Han and Tibetan participants. These results indicated that the minority group’s prototypicality was correlated with positive outgroup attitudes not only for the majority group, which is consistent with a meta-analysis of 27 studies (Wenzel et al., 2007), but also for the minority group. Furthermore, as expected, we found that RIP increased the positive outgroup attitudes of the minority group, which was not explored well in previous studies (Wenzel et al., 2007). Having equal cultural status in Chinese culture (Study 2) may be the maximum extent of Tibetan prototypicality of the Chinese. In such a condition, Tibetans may feel that their distinctive culture is respected and well represented in Chinese culture. Thus, Tibetans could achieve a positive, distinctive ingroup identity and feel included and valued by China. Hence, a high RIP among Tibetans leads to positive attitudes toward the Hans.

Our results also suggested that for the minority group, a high RIP led to positive outgroup attitudes without significant ingroup projection or greater RIP. The high RIP in the present study indicates that the minority group has an equal opportunity to project ingroup characteristics to the superordinate group as the majority group has. The operational definition of the high RIP condition for the Tibetan participants in Study 2 was prototypicality of Han = Tibetan. The Tibetan participants did not report significant differences between ingroup (M = 4.91, SD = 1.10) and outgroup prototypicality, M = 4.89, SD = 1.08; t(34) = 0.20, p = .845; therefore, they did not show significant ingroup projection under this condition. Ufkes et al. (2012) found that when the minority group perceived high ingroup prototypicality, superordinate identification was related to positive outgroup attitudes. The minority-group participants also reported no significant differences between ingroup (M = 4.16, SD = 1.82) and outgroup prototypicality, M = 4.16, SD = 1.68; t(44) < 1.00, p = 1.00 (Ufkes et al., 2012). The minority group also did not show significant ingroup projection in Ufkes et al.’s study. However, if the minority group shows significant ingroup projection, we cannot determine whether the positive effect of high RIP on outgroup attitudes would still exist because we did not test this situation. Therefore, future research should investigate the effect of the minority group’s significant ingroup projection on outgroup attitudes.

RIP of the Minority Group and Multiculturalism

The endorsement of multiculturalism could increase positive outgroup attitudes because individuals learn to appreciate and value different groups’ positive cultures by developing an understanding of the lives and perspectives of different racial and ethnic groups (Correll, Park, & Smith, 2008; Ryan, Casas, & Thompson, 2010). However, multiculturalism may ignore the differences in RIP between groups; that is, the majority group’s culture may represent more of the superordinate group culture than the minority group’s culture. Emphasizing the distinctive culture of different ethnic groups and failing to connect the minority’s culture with the superordinate group culture (RIP of the minority) might be a form of “isolated” respect for the minority. Indeed, multiculturalism may also result in negative outgroup attitudes (Ryan et al., 2010; Ryan, Hunt, Weible, Peterson, & Casas, 2007). Therefore, if the RIP of the minority group on the superordinate group culture is improved by giving the minority group equal status with the majority group, multiculturalism can play a positive role in outgroup attitudes.

Limitations and Future Directions

Despite its strengths, this study also has limitations. We found a positive effect of RIP on outgroup attitudes for the minority group with a very distinctive culture, but we did not test this effect for the minority group without a distinctive culture. A distinctive culture may moderate the effect of RIP on the outgroup attitudes. When minority groups have a distinctive culture and pride in their ingroup culture, they are eager to project ingroup characteristics to the superordinate group (i.e., high RIP) to achieve a positive distinctive ingroup identity and be valued by the superordinate group. However, if the minority group does not have a very obvious distinctive culture, there may be no relationship between RIP and outgroup attitudes because the minority group may be merged into the mainstream culture shaped by the majority group. Additionally, members of the minority group may even evaluate the outgroup positively when they consider their inferior position legitimate (Brewer, 2007). Therefore, future studies should investigate the moderator effect of a distinctive culture for the minority group.

Moreover, future studies could explore the effect of the minority group’s ingroup projection on outgroup attitudes. Spoor, Jetten, and Hornsey (2014) found that the minority group tended to be less favorable toward overrepresentation of the superordinate group. The minority group’s ingroup projection may challenge the status quo (Wenzel et al., 2007) and, thus, may be related to negative attitudes toward the majority group (Ng Tseung-Wong & Verkuyten, 2010).

Additionally, future studies should examine the mechanism underlying the process by which RIP influences the minority group’s outgroup attitudes. High RIP may decrease the distinctive threat and improve positive outgroup attitudes or may increase the legitimacy of status differences and lead to more positive outgroup attitudes when the superordinate category is positively evaluated.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Beijing Well-Being Foundation(No. 00203442015-01-005), and the Fundamental Research Funds for the Central Universities.

Notes

References

Bergsieker

H. B.

Shelton

J. N.

Richeson

J. A.

(2010). To be liked versus respected: Divergent goals in interracial interactions. Journal of Personality and Social Psychology, 99, 248–264. doi:10.1037/a0018474

Bianchi

Mummendey

Steffens

M. C.

Yzerbyt

V. Y.

(2010). What do you mean by “European”? Evidence of spontaneous ingroup projection. Personality and Social Psychology Bulletin, 36, 960–974. doi:10.1177/0146167210367488

Brewer

M. B.

(2007). The importance of being we: Human nature and intergroup relations. American Psychologist, 62, 726–738. doi:10.1037/0003–066X.62.8.728

Brewer

M. B.

Gonsalkorale

van Dommelen

(2013). Social identity complexity: Comparing majority and minority ethnic group members in a multicultural society. Group Processes & Intergroup Relations, 16, 529–544. doi:10.1177/1368430212468622

Correll

Park

Allegra Smith

(2008). Colorblind and multicultural prejudice reduction strategies in high-conflict situations. Group Processes & Intergroup Relations, 11, 471–491. doi:10.1177/1368430208095401

Devos

Banaji

M. R.

(2005). American = White? Journal of Personality and Social Psychology, 88, 447–466. doi:10.1037/0022–3514.88.3.447

Elkins

Sides

(2007). Can institutions build unity in multiethnic states? American Political Science Review, 101, 693–708. doi:10.1017/S0003055407070505

Gaertner

S. L.

Dovidio

J. F.

(2000). Reducing intergroup bias: The common ingroup identity model. Philadelphia, PA: The Psychology Press.

Gaertner

S. L.

Dovidio

J. F.

(2009). A common ingroup identity: A categorization-based approach for reducing intergroup bias. In Nelson

(Ed.), Handbook of prejudice (pp. 489–506). Philadelphia, PA: Taylor & Francis.

10.

Hornsey

M. J.

Hogg

M. A.

(1999). Subgroup differentiation as a response to an overly inclusive group: A test of optimal distinctiveness theory. European Journal of Social Psychology, 29, 543–550. doi:10.1002/(SICI)1099–0992(199906)29:4<543::AID-EJSP945>3.0.CO;2-A

11.

Huynh

Devos

Altman

H. R.

(2015). Boundaries of American identity: Relations between ethnic group prototypicality and policy attitudes. Political Psychology, 36, 449–468. doi:10.1111/pops.12189

12.

Kessler

Mummendey

(2009). Why do they not perceive us as we are? Ingroup projection as a source of intergroup misunderstanding. In Demoulin

Leyens

J. P.

Dovidio

J. F.

(Eds.), Intergroup misunderstandings: Impact of divergent social realities (pp. 135–148). New York, NY: Psychology Press.

13.

Kessler

Mummendey

Funke

Brown

Binder

Zagefka

. . . Maquil

(2010). We all live in Germany but . . . Ingroup projection, group-based emotions and prejudice against immigrants. European Journal of Social Psychology, 40, 985–997. doi:10.1002/ejsp.673

14.

Leonardelli

G. J.

Pickett

C. L.

Brewer

M. B.

(2010). Optimal distinctiveness theory: A framework for social identity, social cognition, and intergroup relations. Advances in Experimental Social Psychology, 43, 63–113. doi:10.1016/S0065–2601(10)43002–6

15.

Mael

Ashforth

B. E.

(1992). Alumni and their alma mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior, 13, 103–123. doi:10.1002/job.4030130202

16.

Mummendey

Wenzel

(1999). Social discrimination and tolerance in intergroup relations: Reactions to intergroup difference. Personality and Social Psychology Review, 3, 158–174. doi:10.1207/s15327957pspr0302_4

17.

National Bureau of Statistics of the people’s Republic of China. (2011). National census 2010 in China. Retrieved from http://baike.baidu.com/link?url=zARfTHa3bkWWe2OpyPCHyfTU1c9y6cqd1-RofawrtVeXTcGrQhq_6Lk7TK51gFgo6zzjqoya9zwXh0zMyLQ_sK

18.

Ng Tseung-Wong

Verkuyten

(2010). Intergroup evaluations, group indispensability and prototypicality judgments: A study in Mauritius. Group Processes & Intergroup Relations, 13, 621–638. doi:10.1177/1368430210369345

19.

Peker

Crisp

R. J.

Hogg

M. A.

(2010). Predictors of ingroup projection: The roles of superordinate category coherence and complexity. Group Processes & Intergroup Relations, 13, 525–542. doi:10.1177/1368430209360205

20.

Putnam

R. D.

(2007). E pluribus unum: Diversity and community in the twenty-first century. The 2006 Johan Skytte Prize lecture. Scandinavian Political Studies, 30, 137–174. doi:10.1111/j.1467–9477.2007.00176.x

21.

Qiu

Z. H.

(2013). Data processing and data check. In Qiu

Z. H.

(Ed.), Quantitative research and statistical analysis—A case study of SPSS data analysis (pp. 87–88). Chongqing, China: Chongqing University Press.

22.

Reicher

(2004). The context of social identity: Domination, resistance, and change. Political Psychology, 25, 921–945. doi:10.1111/j.1467–9221.2004.00403.x

23.

Reynolds

W. M.

(1982). Development of reliable and valid short forms of the Marlowe–Crowne Social Desirability Scale. Journal of Clinical Psychology, 38, 119–125. doi:10.1002/1097–4679(198201)38:1<119::AID-JCLP2270380118>3.0.CO;2-I

24.

Ryan

C. S.

Casas

J. F.

Thompson

B. K.

(2010). Interethnic ideology, intergroup perceptions, and cultural orientation. Journal of Social Issues, 66, 29–44. doi:10.1111/j.1540–4560.2009.01631.x

25.

Ryan

C. S.

Hunt

J. S.

Weible

J. A.

Peterson

C. R.

Casas

J. F.

(2007). Multicultural and colorblind ideology, stereotypes, and ethnocentrism among Black and White Americans. Group Processes & Intergroup Relations, 10, 617–637. doi:10.1177/1368430207084105

26.

Schubert

T. W.

Otten

(2002). Overlap of self, ingroup, and outgroup: Pictorial measures of self categorization. Self and Identity, 1, 353–376. doi:10.1080/152988602760328012

27.

Schuessler

Hittle

Cardascia

(1978). Measuring responding desirably with attitude–opinion items. Social Psychology, 41, 224–235. doi:10.2307/3033559

28.

Sindic

Reicher

S. D.

(2008). The instrumental use of group prototypicality judgments. Journal of Experimental Social Psychology, 44, 1425–1435. doi:10.1016/j.jesp.2008.05.007

29.

Sindic

Reicher

S. D.

(2009). “Our way of life is worth defending”: Testing a model of attitudes towards superordinate group membership through a study of Scots’ attitudes towards Britain. European Journal of Social Psychology, 39, 114–129. doi:10.1002/ejsp.503

30.

Spoor

J. R.

Jetten

Hornsey

M. J.

(2014). Overplaying the diversity card: When a superordinate group overrepresents the prevalence of a minority group. Group Processes & Intergroup Relations, 17, 161–177. doi:10.1177/1368430213497063

31.

Staerklé

Sidanius

Green

E. G. T.

Molina

L. E.

(2010). Ethnic minority–majority asymmetry in national attitudes around the world: A multilevel analysis. Political Psychology, 31, 491–519. doi:10.1111/j.1467–9221.2010.00766.x

32.

Tajfel

Turner

J. C.

(1979). An integrative theory of intergroup conflict. In Austin

W. G.

Worchel

(Eds.), The social psychology of intergroup relations (pp. 33–48). Monterey, CA: Brooks/Cole.

33.

Turner

J. C.

(1987). A self-categorisation theory. In Turner

J. C.

Hogg

M. A.

Oakes

P. J.

Reicher

S. D.

Wetherell

M. S.

(Eds.), Rediscovering the social group: A self-categorisation theory (pp. 42–67). Oxford, UK: Blackwell.

34.

Ufkes

E. G.

Otten

van Der Zee

K. I.

Giebels

Dovidio

J. F.

(2012). Urban district identity as a common ingroup identity: The different role of ingroup prototypicality for minority and majority groups. European Journal of Social Psychology, 42, 706–716. doi:10.1002/ejsp.1888

35.

Ullrich

(2009). Reconsidering the “relative” in relative ingroup prototypicality. European Journal of Social Psychology, 39, 299–310. doi:10.1002/ejsp.540

36.

Ullrich

Christ

Schlüter

(2006). Merging on Mayday: Subgroup and superordinate identification as joint moderators of threat effects in the context of European Union’s expansion. European Journal of Social Psychology, 36, 857–875. doi:10.1002/ejsp.319

37.

Van Knippenberg

van Oers

(1984). Social identity and equity concerns in intergroup perceptions. British Journal of Social Psychology, 23, 351–361. doi:10.1111/j.2044–8309.1984.tb00651.x

38.

Van Oudenhoven

J. P.

Prins

K. S.

Buunk

B. P.

(1998). Attitudes of minority and majority members towards adaptation of immigrants. European Journal of Social Psychology, 28, 995–1013. doi:10.1002/(SICI)1099–0992(1998110)28:6<995::AID-EJSP908>3.0.CO;2–8

39.

Waldzus

Mummendey

(2002, June). Advanced findings on ingroup projection: Causal effects and side effects. Paper presented at the 4th Jena Workshop on Intergroup Processes, Jena, Germany.

40.

Waldzus

Mummendey

(2004). Inclusion in a superordinate category, in-group prototypicality, and attitudes towards out-groups. Journal of Experimental Social Psychology, 40, 466–477. doi:10.1016/j.jesp.2003.09.003

41.

Waldzus

Mummendey

Wenzel

(2005). When “different” means “worse”: In-group prototypicality in changing intergroup contexts. Journal of Experimental Social Psychology, 41, 76–83. doi:10.1016/j.jesp.2004.05.006

42.

Waldzus

Mummendey

Wenzel

Boettcher

(2004). Of bikers, teachers and Germans: Groups’ diverging views about their prototypicality. British Journal of Social Psychology, 43, 385–400. doi:10.1348/0144666042037944

43.

Waldzus

Mummendey

Wenzel

Weber

(2003). Towards tolerance: Representations of superordinate categories and perceived ingroup prototypicality. Journal of Experimental Social Psychology, 39, 31–47. doi:10.1016/S0022–1031(02)00507–3

44.

Weber

Mummendey

Waldzus

(2002). Perceived legitimacy of intergroup status differences: Its prediction by relative ingroup prototypicality. European Journal of Social Psychology, 32, 449–470. doi:10.1002/ejsp.102

45.

Wenzel

Mummendey

Waldzus

(2007). Superordinate identities and intergroup conflict: The ingroup projection model. European Review of Social Psychology, 18, 331–372. doi:10.1080/10463280701728302

46.

Wenzel

Mummendey

Weber

Waldzus

(2003). The ingroup as pars pro toto: Projection from the ingroup onto the inclusive category as precursor to social discrimination. Personality and Social Psychology Bulletin, 29, 461–473. doi:10.1177/0146167202250913

47.

Wohl

M. J. A.

Giguère

Branscombe

N. R.

Mcvicar

D. N.

(2011). One day we might be no more: Collective angst and protective action from potential distinctiveness loss. European Journal of Social Psychology, 41, 289–300. doi:10.1002/ejsp.773