Testing the multicomponent model of intergroup attitudes and accompanied open-ended measures

According to the multicomponent model of intergroup attitudes our attitudes towards groups are based on emotions toward group members, beliefs about specific characteristics of members of the social group and beliefs about which values are promoted or blocked by the group members. The first aim of this study was to check the validity and reliability of the open-ended measures for assessing components of attitudes towards groups, and the second aim was to test the multicomponent model of intergroup attitudes. The study examined attitudes toward two groups as attitude objects. General attitude was assessed by semantic differential scales. On the bases of data analysis with regard to the content and favorability of subjects' (N = 119) free-responses it was concluded that open-ended measures are valid indicators of attitudinal components. The split-half reliability of open ended measures are satisfactory although lower than mean reliability indices found in literature, A multiple regression analysis revealed that stereotypes provided a significant contribution to the prediction of attitude towards both groups. Affective responses were a significant unique predictor for attitude towards only one group whereas symbolic beliefs in both cases were not predictive of the general attitude. The results do not support the importance of symbolic beliefs as a part of the cognitive basis of attitudes, which is the main new part of the multicomponent model of intergroup attitudes with regard to the general multicomponent model of attitudes.