A Study in Social Identity as a Function of Cross-Category Membership in Certain Ethnic Groups in Bangladesh

Abstract

The present study is an investigation relating to social identity, cross­category membership and intergroup relations. It is an attempt to interrelate the social identity with several variables such as group comparisons, gender identification, age variation and residential differentiation. These A-'our" independent variables were tested with the help of the measures of adjective check list resulting in differential amount of social identity. Group composition represented Santals and Hindus. Gender represented boys and girls. Age represented 13 years age group and 16 years age group. Residence represented urban and rural. The objective of the study was to investigate social identity of Santals as racial ethnic group and Bengali Hindus as religious ethnic group in the cross-category membership or inter-group relations as related to gender, .age and residence. Three hypotheses formulated in the present study were: (I) Bengali Hindus as religious ethnic group would evaluate self and own-group more positively and out-group more negatively as compared to Santals as racial ethnic groups. (2) Boys would evaluate self and own-group more positively and out-group more negatively as compared to girls. (3) Respondents with 16 years age would evaluate self and own-group significantly more positively and the reciprocal out-group more negatively as compared to respondents with 13 years age. The sample of the present study was composed of 240 respondents. 'They were equally divided into Santals as racial ethnic group and Bengali Hindus as religious ethnic group. Each ethnic group was equally divided into boys and girls. They were again equally divided into 13 years age group and 16 years age group. Finally, each age group was equally subdivided into urban and rural according to their residential background. The measure of Adjective Check List was used for data collection. The data were analyzed using the statistical technique of Analysis of Variance (ANOVA). The results were computed in three parts. In the first part, ANOVA was computed separately for Santals. This was done in three sections. In the first section, ANOVA was computed with the scores of self-evaluation representing two level of gender (Boy/Girl), two level of age (13 years/16 years) and two level of residence (Urban/Rural). In the second section, ANOVA was computed with the scores of own group evaluation representing two level of gender (Boy/Girls), two level of age ( 13 years/16 years) and two level of residence (Urban/Rural). In the third section, ANOVA was computed on the scores of out-group evaluation representing two level of gender (Boy/Girl), two level of age (13 years/16 years) and two level of residence (Urban/Rural). Similar procedure was followed in the second part for the respondents of Bengali Hindus. However, the third part of the results contained ANOVA involving group composition (Santal/Bengali Hindu), gender (Boys/Girls) and age (13 years/16 years) regardless of residential background separately on self-evaluation, own-group evaluation and out-group evaluation. The results of the study provided empirical supports to all the hypotheses. It was found that Bengali Hindus as religious ethnic group evaluated self and own-group significantly more positively and out-group more negatively as compared to Santals as racial ethnic group. These findings emerged from the theory of social identity in relation to cross- category membership. Furthermore, the results showed that irrespective of age and residential background. The boys evaluated self and own-group significantly more positively and out-group more negatively as compared to girls. However, this finding was not consistent throughout the results. Hence, it may be said that the second hypothesis was partially supported by the investigation of the study. The third hypothesis that respondents with 16 years age would evaluate self and own-group more positively and reciprocal out-group more negatively has been partially supported by the findings. It was found that Santal and Hindu respondents of 16 years age evaluated out-group significantly more negatively than the respondents of 13 years age. But in case of self-evaluation, the Santal and Hindu respondents of 13 and 16 years age failed to obtain significant difference. Lastly, the findings of the present study provided strong support to the fourth hypothesis. It was found that respondents with urban residential background have evaluated self and own-group significantly more positively and out-group significantly more negatively as compared to the respondents with rural residential background. In conclusion, it might be said that social identity may provide different sources of meaning due to cross-category membership. It is likely that these differences in identities may overlap. Sometimes, they may reinforce each other. Again they may constrain the self and own-group from negative evaluation of the out-group. These conditions have appeared in the present study and they provide important cues for future investigation in the area of social identity of inter-group relation.