Social representations of aggression: Crossing the sex barrier

Controversy over Moscovici's concept of social representations has focused upon the extent to which they can be viewed as enduring cognitive structures characterizing social groups and whether individual members are ‘prisoners’ of their social representations, unable to duplicate the social representations of other social groups. Previous research has established a consistent gender difference in orientation toward aggression with men viewing it as an instrumental act of coercion and women as a temporary loss of self-control. These two social representations, originally recovered from spontaneous conversation, have been measured with a psychometric instrument called Expagg. To examine the mutability of these representations, men and women in the present study were asked to complete the questionnaire either spontaneously or as they believed a member of the opposite sex might respond. Under conditions of same-sex responding the usual significant sex difference appeared. When asked to respond as a member of the opposite sex, men accurately mirrored women's higher expressive total score on the questionnaire but psychometric analysis revealed that there was no similarity in terms of item–total correlations. Women grossly overestimated the degree of men's instrumentality but item–total total correlations revealed a considerable degree of similarity with men's structure. The male representation whether natural or assumed showed higher internal consistency than did the female mode. The results are discussed in terms of differential modes of access to gender-linked representations and the cultural dominance of a masculine and instrumental representation of aggression.     

RESPONSE‐DEPENDENCE THEORY AND EMPIRICAL CLAIMS FOR THE SOCIAL SCIENCES

Abstract: The analysis here is an attempt to show how the current epistemological theory of response‐dependence (R‐D) may be relevant to understanding putative ontological claims of the empirical social sciences. To this end I argue that the constitutive features of human response, central to R‐D theory, can be made explicit for social science. I conclude that for the empirical social sciences the implication of combining R‐D and certain forms of statistical analyses leads to the possibility of an events‐based ontology.