In-Group Favoritism Among Native and Immigrant ...

1 downloads 0 Views 652KB Size Report
Apr 2, 2010 - in-group positivity, however, the data about out-group derogation are not as ... with the derogation hypothesis, the native participants judged the ...
The Journal of Social Psychology

ISSN: 0022-4545 (Print) 1940-1183 (Online) Journal homepage: http://www.tandfonline.com/loi/vsoc20

In-Group Favoritism Among Native and Immigrant Turkish Cypriots: Trait Evaluations of In-Group and Out-Group Targets Ahmet Rustemli , Biran Mertan & Orhan Ciftci To cite this article: Ahmet Rustemli , Biran Mertan & Orhan Ciftci (2000) In-Group Favoritism Among Native and Immigrant Turkish Cypriots: Trait Evaluations of In-Group and Out-Group Targets, The Journal of Social Psychology, 140:1, 26-34, DOI: 10.1080/00224540009600443 To link to this article: http://dx.doi.org/10.1080/00224540009600443

Published online: 02 Apr 2010.

Submit your article to this journal

Article views: 68

View related articles

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=vsoc20 Download by: [Dogu Akdeniz University]

Date: 27 December 2016, At: 23:53

The Journal of Social Psvcholoav, 2000. 14N1).26-34

In-Group Favoritism Among Native and Immigrant Turkish Cypriots: Trait Evaluations of In-Group and Out-Group Targets AHMET RUSTEMLI BIRAN MERTAN ORHAN C I F K I Department of Psychology Near East University Lejkosa, Northern Cyprus

ABSTRACT. The authors investigated whether in-group favoritism manifests itself as praise for the in-group or as denigration of the out-group. A total of 450 Turkish Cypriots (248 native, 202 immigrant) judged the applicability of positive and negative trait words to in-group and out-group targets. Both the native and the immigrant groups judged the positive traits as more applicable to their respective in-groups than to the out-group. The native group evaluated the negative social traits as more applicable to the immigrant group. The immigrant group also judged the negative social traits as more applicable to themselves. The two groups did not differ in their judgments for more personal negative traits.

SOCIAL IDENTITY THEORY (SIT; Tajfel & Turner, 1979) has received considerable attention in the study of intergroup discrimination. Social identity theorists assume that the social environment is organized into meaningful categories that ultimately result in intergroup discrimination. According to SIT, the classification of the social world as “us” (in-group) or “them” (out-group) results i n discriminatory behavior and negative attitudes toward out-group members. This discrimination is triggered by self-serving motives: To maintain and enhance a positive self-regard, individuals dispose themselves positively toward members of their own group and discriminate against members of other groups. Researchers have indicated that people have more positive views toward Address correspondence to Biran Mertan, 42 Sehit Ali Cakir Sokak, Lejkosa, Mersin 10, Turkey. 26

Rustemli, Mertan, & Ciftci

27

members of their own groups: Even in groups constructed randomly or on the bases of trivial criteria, participants made performance evaluations and reward allocations favoring members of the in-group (Brewer, 1979; Crocker & Luhtanen, 1990; Sidanus, Pratto, & Mitchell, 1994; Vanbeselaere, 1987). This in-group bius is evident also among ethnic, religious, and social groups. For example, Lee ( 1993) found that African American and Chinese American students, two minority groups in the United States, perceived in-group members more favorably than they perceived out-group members. In a study with Catholic and Protestant adolescents in Northern Ireland, the participants attributed violent acts more to external reasons when the offender was an in-group member and more to internal reasons when the offender was an out-group member (Hunter, Stringer, & Coleman, 1993). Although some findings deviate from the predictions of SIT-for example, low-status and minority groups evaluated out-group members more positively than they did in-group members (Hewstone & Ward, 1985; Hinkle & Brown, 1990)-most research essentially supports the predictions of SIT. In the present research, we addressed the issue of whether in-group favoritism manifests itself as in-group positivity, as out-group negativity, or as both. Because t h e term juvoritism is used in conjunction with a comparison of two or more groups, there is a need for clarity about the meaning of the term. Being disposed positively toward one's own group may be qualitatively different from being disposed negatively toward other groups; such a difference may have important implications. Some types of intergroup discrimination cannot be explained easily by unidirectional favoritism. Sidanus et al. (1994) noted widespread instances of out-group hostility, humiliation, and oppression throughout the world. Humiliation and oppression, for example, generally imply power relations between groups. Dominant groups are more likely than dominated groups to use such overt forms of subordination. Favoring in-group members, on the other hand, does not necessarily mean humiliating or feeling hostile toward outgroups; a person may orient himself or herself positively toward members of his or her own group and still stay neutral toward members of the out-group. This assertion, of course, qualifies the usual interpretation-in-group favoritism-of more favorable ratings of the in-group and suggests an alternative interpretation. Rosenbaum's (1986) argument with respect to the relationship between similarity and liking seems also to be valid for in-group favoritism. He argued that ratings that indicate a greater liking for a person similar to the rater do not necessarily mean that similarity leads to liking; the ratings could also mean that dissimilarity leads to repulsion. Similarly, the frequently reported differences in evaluations of in-group and out-group persons in favor of the in-group might well have resulted from feelings of repulsion toward the out-group as well as from feelings of attraction to the in-group. Evaluating groups with a list of positive and negative trait words may be a meaningful way to test the nature of intergroup bias. Because group membership is a constituent of the social self, SIT (Tajfel & Turner, 1986) predicts that the

28

The Journal of Social Psychology

cognitive representation of the social group to which one belongs will be more positive than the cognitive representation of social groups to which one does not belong. The question in the present research was how does being more positive manifest itself-as the attribution of more socially positive traits to the in-group, as the attribution of more socially negative traits to the out-group, or as both? Do people enhance and maintain their self-esteem by praising themselves or by denigrating others? Within this framework, we predicted that participants would assign different values to trait words when describing in-group and out-group targets. We preferred that the direction of this difference should remain as an empirical question. In the present study, there were two groups of participants: native Turkish Cypriots and Turkish immigrants living in Northern Cyprus. Although Cypriots of Greek and Turkish origins have cohabited the island for centuries, events in the second half of the 20th century led to separation of the two communities into two distinct regions in 1974. Greek Cypriots live in the south, and Turkish Cypriots live in the north. Since this division, political and economic circumstances in Northern Cyprus have led to high rates of emigration and immigration, mainly to and from Turkey. At present, the population in Northern Cyprus is composed mainly of native Turkish Cypriots and Turkish immigrants. Although the two groups share the same religion, ethnic origins, and language, their experiences have been quite different. Turkish Cypriots have been exposed not only to Greek Orthodox culture but also to British influences during the colonial period. The immigrants, who come mostly from rural Turkey, constitute a traditional group with limited formal schooling. Because of their ascertainable resemblances and differences, native Turkish Cypriots and Turkish immigrants constitute two distinct groups. Northern Cyprus, with its small and responsive population, was an excellent field for conducting in-depth research among the two groups.

Method Participants

The participants were 450 Turkish Cypriots of both sexes. Of the total, 121 women and 127 men were natives of Cyprus, and 99 women and 103 men were immigrants from Turkey. The mean length of residence in Cyprus for the Iatter group was 17.26 years (SD= 6.16). The ages of the participants ranged from 19 to 70 years (M = 35.74, SD = 11.48). The sample was a nonprobabilistic one taken from convenient rural and urban areas. Instrument

The instrument was a two-part questionnaire. The first part contained questions focusing on such background variables as the participant’s age, gender,

Rustemli, Mertan, & Ciftci

29

education, income, place and year of birth, and length of residence in Cyprus. In the second part, we identified the target group and provided 26 adjectives (13 positive, 13 negative) used in descriptions of people. We had determined the list in a pilot study in which 104 Cypriots (61 native and 43 immigrant) of different ages, gender, and occupational backgrounds judged the favorability of 80 adjectives frequently used in describing persons. We chose for the present study the adjectives that were rated similarly by natives and immigrants and by men and women. The target groups given to each participant were either “native Cypriots” or “immigrant Cypriots” (both of Turkish ethnicity) and either “men” or “women.” The participants indicated to what extent each trait word applied to the target groups by using a Likert-type scale (4 = most applicable, 1 = not applicable at all). Procedure

We collected the data in April-May 1996. We went to different residential and business areas in towns and to different villages on different days. We contacted the participants in their homes, work places, and, in a few instances, at public places such as coffee houses. We did not follow any systematic procedure in selecting the participants.

Results Instead of analyzing the responses adjective by adjective, we first subjected the data to factor analysis, to categorize them into meaningful groups. An initial analysis using principal components with varimax rotation resulted in six factors that accounted for 61.0% of the total variance. After repeated analyses with restrictions on the number of factors, we chose as the best summary of the data the four-factor solution that explained 53.1% of the total variance. Factor 1, Social-Negative, portrayed a socially negative personality (variance = 32.7%; a = 3 7 ) . The adjectives (loadings in parentheses) that defined Factor 1 were conservative (.82), modern (-.74), ignorant (.7l), unmannerly (.67), dirty (.66), rude (.60), polite (-.55), and aggressive (.48). The adjectives (loadings in parentheses) that defined Factor 2, Social-Positive (variance = 8.7%; c1 = .81), were loyal (.74), honest (.70), reliable (.60), selfish (-.53), tolerant (.43), and lovely (.41). Factors 3 and &Personal-Negative and Personal-Positive, respectivelywere composed of more personal positive and negative terms. The terms (loadings in parentheses) that defined Factor 3 (variance = 6.2%;a = .64) were pessimistic (.69), lack of confidence (.60), gossip (.57), bad-tempered (.55), disrespectful (.39), and lazy (.43). The terms (loadings in parentheses) that defined Factor 4 (variance = 5.5%;a = .71) were initiative (.62), brave (.61), successful (.56), proud (.5 l), friendly (.44),and skillful (.42).

30

The Journal of Social Psychology

We conducted a multivariate analysis of covariance on the computed factor scores by using participant group (native-immigrant), participant gender, target group (in-groupout-group), target gender, trait type (social-personal), and trait status (negative-positive) as variables with repeated measures on the last two and with age as a covariate. Because the results of this analysis indicated no meaningful effects for the variables of gender and age, we pooled the data for these variables and reanalyzed them. The main effects for trait type and trait status, the within-subject variables, were significant. Personal traits, F(1,404) = 126.88, p c .001, and positive traits, F( 1,404) = 213.22, p < .001, were judged as more applicable than social and negative traits to the target individuals. The interaction effect (see Table 1, last row) for those variables was also significant, F(1, 406) = 20.92, p < .001. Positive-social and positive-personal traits were judged equally applicable to target persons, whereas negative-social traits were judged as significantly less applicable than negative-personal traits (Ms = 2.1 1 and 2.35, SDs = 0.67 and 0.55, respectively), t(428) = 11.05,p < .001. The critical effect for the present research was the interaction between target group and trait status. This interaction reached significance, F(1,406)= 9.60, p < .002. Positive traits were judged as being more applicable to in-group targets ( M = 2.97, SD = 0.44) than to out-group targets (M= 2.75, SD = 0.46), r(424) = 4.39, p c .001; the reverse was the case for negative traits. The applicability ratings of negative traits for in-group targets ( M = 2.19, SD = 0.50) were significantly lower than they were for out-group targets (M= 2.29, SD = 0.56), t(426) = 2.14, p < .033. However, the significant triple interaction of those two variables with participant group, F(1, 404) = 8 . 9 3 , ~< .003, required qualifications of this TABLE X Mean Applicability Ratings and Standard Deviations of In-Group and Out-Group Targets, by Participant Status, Trait Type, and Trait Status

Participant status Native In-group Out-group Immigrant In-group Out-group Total

Social traits Positive Negative M SD M SD

Personal traits Positive Negative M SD M SD

2.99 2.66

0.48 0.64

1.84 2.34

0.53 0.72

2.98 2.81

0.44 0.46

2.33 0.56 2.37 0.58

2.90 2.75 2.83

0.54

0.51 0.56

2.25 1.93 2.11

0.53 0.43 0.62

2.95 2.77 2.86

0.50 0.40 0.46

2.34 0.53 2.37 0.52 2.35 0.55

Nore. Higher values indicate higher applicability

Rustemli, Mertan, & Ciftci

31

two-way interaction (see Figure 1). As indicated in the figure, both the native group and the immigrant group were more positive toward in-group targets than toward out-group targets. With respect to negative traits, however, the two groups differed in their judgments; the native group judged the negative traits as more applicable to out-group targets (M = 2.36, SD = 0.61) than to in-group targets ( M = 2.09, SD = 0.49), t(239) = 5.08, p < .001, whereas the immigrants judged the negative traits as more applicable to in-group targets (M = 2.30, SD = 0.50) than to out-group targets (M= 2.15, SD = 0.43), t( 185) = 2.86, p < .005. The triple interaction of the grouping variables with trait type was also significant, F( 1,404) = 22.19, p < .001. Because that interaction was disqualified by a higher order interaction with trait status, detailed analyses are reported on the latter interaction. Table 1 contains the means for the significant four-way interaction among the grouping and trait variables, F( 1,404) = 77.44, p < .001. Both the native group and the immigrant group judged positive traits (social-positive and personal-positive) as more applicable to in-group targets than to out-group targets. Neither the native nor the immigrant group showed any difference in judging the applicability of personal-negative traits to in- and out-group targets. The mean ratings for in- and out-group targets were almost the same (Table 1, last col-

FIGURE 1. Mean applicability ratings for positive and negative traits, according to participant group and target.

32

The Journal of Social Psychology

umn). The native group and the immigrant group did differ, however, in their judgments of social-negative traits. The native group assigned a higher degree of applicability of social-negative traits to the out-group (M= 2.34, SD = 0.72) than to the in-group (M= 1.84, SD = 0.53). The immigrants also differentiated between in- and out-group targets in respect to social-negative traits (Ms = 2.25 and 1.93, SDs = 0.53 and 0.43, respectively), but this difference was in favor of the outgroup, t(191) = 5 . 1 5 , ~< .001.

Discussion We investigated whether in-group favoritism manifests itself as praising the in-group or as derogating the out-group. The results are clear about the praising hypothesis: Both the native and the immigrant respondents evaluated positive traits as more applicable to their own groups than to the other group. Contrary to in-group positivity, however, the data about out-group derogation are not as clear. First, the participants did make a distinction between personal attributes and more general or social attributes. Traits such as pessimism, lack of confidence, bad temper, laziness and disrespect were not easily generalized to all members of a large social group. This seems to be the case, especially when members of social groups interact with each other in daily routines and have opportunities to know the other group’s members as individuals who deviate from group stereotypes. Both the native and immigrant participants evaluated these traits as equally applicable to members of their own group and to members of the out-group. The two groups, however, differed on the factor composed of the traits “conservative,” “ignorant,” “unmannerly,” “not modern,” and “not polite.” Consistent with the derogation hypothesis, the native participants judged the foregoing traits as more applicable to immigrant targets than to native targets. The immigrant group also differentiated between in-group and out-group targets on those traits, but in the opposite direction; the immigrant participants evaluated these traits as more applicable to their own group than to the other group, showing in-group derogation. The results suggest that in-group favoritism may operate both as praising the in-group and derogating the out-group but under different conditions. The former appears to be the basic and general manifestation of intergroup bias. In the present study, both the native participants and the immigrant participants not only judged the positive traits as more applicable to their own groups but, at the same time, as more applicable than the negative traits to both target groups. Those findings are congruent with the notion that group members are motivated to work together and try to enhance their objective position in the society, and, thus, their social identity (Tajfel & Turner, 1986). Derogation of out-groups, on the other hand, appears to be related to group status. In the present study, the native Cypriots were a political and numerical majority with higher social status than the immigrant group. The native group

Rustemli, Mertan, & Ciftci

33

was better educated, x*(4)= 97.98, p < .001, and had higher monthly incomes, p < .001, than the immigrant group. As social dominance theory ~ ~ (=429.23, ) suggests (Sidanus et al., 1994), the members of the present socially dominant group did not hesitate to denigrate the targets of the lower status immigrant group. In addition to self-serving motives triggered by social categorization, the competition created by the immigrants for rare jobs in a declining economy may have contributed to the out-group denigration. The members of the subordinate group, on the other hand, agreed with this stereotyping and perceived the ingroup as socially more negative than the out-group. This in-group inferiority is at odds with the predictions of SIT (Tajfel & Turner, 1979, 1986) and suggests that intergroup bias is not independent of the relative positions of the groups in society. Low-status groups may feel themselves under threat and experience fear and anxiety. The derogation of the in-group may be a manifestation of such a fear and the desire to avoid conflict with the dominant group that would undermine social harmony (Hewstone &Ward, 1985). We obtained the present results from between-groups data. A great majority of the earlier studies on intergroup bias used within-group comparisons. Although within-group procedures are effective in tapping perceptual differences between in- and out-group targets, there are some drawbacks. Asking participants to make multiple comparisons increases both the complexity of the rating task (Brown & Turner, 1979) and the saliency of social categorization, which, in turn, accentuates perceived differences between groups. Although in daily life, people make many comparisons of the self and others, such comparisons are seldom conscious and consecutive. Rather, the comparisons are mediated most of the time by a cognitive anchoring process that is beyond awareness. Therefore, data from independent groups may be immune to the effect produced by the artificially increased saliency of the comparison process intrinsic to consecutive multiple comparisons; therefore, such data may be a better approximation of the discriminatory process eventuated by social categorization. However, new research is needed to compare data from within-group and between-groups procedures in the study of intergroup bias. Gender had no effect on perceptions of in- and out-group targets. Both the men and the women perceived the given traits as equally applicable to the targets differing in gender and place of birth. One possible explanation for the lack of gender effect on judgments is that that variable did not attain saliency. Native or immigrant status was more important than gender in the changing character of the society. In summary, the present results suggest that in-group favoritism may vary according to group, depending upon how one group relates to another. Members of dominant groups may engage in self-enhancement not only by praising their own groups but also by derogating subordinate groups. Subordinate groups also use self-praise for the maintenance of their social identity but refrain from denigrating members of dominant groups.

34

The Journal of Social Psychology

REFERENCES Brewer, M. B. (1979). In-group bias in the minimal group situation: A cognitivemotivational analysis. Psychological Bulletin, 86,307-324. Brown, R. J., & Turner, J. C. (1979). The criss-cross categorization effect in intergroup discrimination. British Journal of Social and Clinical Psychology, 18, 37 1-383. Crocker, J., & Luhtanen, R. (1990). Collective self-esteem and in-group bias. Journal of Personality and Social Psychology, 58, 60-67. Hewstone, M., &Ward, C. (1985). Ethnocentrism and causal attribution in SoutheastAsia. Journal of Personality and Social Psychology, 48, 614-623. Hinkle, S., & Brown, R. J. (1990). Intergroup comparisons and social identity: Some links and lacunae. In D. Abrams & M. A. Hogg (Eds.), Social identi@ theory: Constructive and critical advances. London: Harvester Wheatsheaf. Hunter, J. A., Stringer, M., & Coleman, J. T. (1993). Social explanations and self-esteem in Northern Ireland. The Journal of Social Psychology, 133, 643-650. Lee, Y. T. (1993). Ingroup preference and homogeneity among African American and Chinese American students. The Journal of Social Psychology, 133. 225-235. Rosenbaum, M. E. (1986). The repulsion hypothesis: On the nondevelopment of relationships. Journal of Personality and Social Psychology, 51, 1156-1 166. Sidanus, J., Pratto, F., & Mitchell, M. (1994). In-group identification, social dominance orientation, and differential intergroup social allocation. The Journal o/ Social Psychology, 134, 151-167. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.). The social psychology of inrergroup relations (pp. 3347). Monterey, CA: Brooks/Cole. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago, IL: Nelson-Hall. Vanbeselaere, N. (1987). The effects of dichotomous and crossed social categorizations upon intergroup discrimination. European Journal of Social Psychology, 17, 143-1 56.

Received September 8, 1997 Accepted January 8, I998