THE ROLE OF GENDER IN DESCRIPTIVE REPRESENTATION

This article broadens consideration of the gender gap from voting diff erences to the larger question of affective preferences for descriptiv e representation (Pitkin 1967). The results, based on a 1993 survey of 416 individuals, suggest that women are far more likely than men to b e ''gender conscious'' in their evaluation of a candidate or a preferr ed representative. Differences among the 224 women in the sample can b e traced to at least four sources. Group interests and feminist attitu des are positive sources of women's preferences for descriptive repres entation. Conversely, conservative political views deter some women fr om supporting women in politics. The results also provide partial supp ort for Carroll's (1987) psychological and economic autonomy thesis. F inally, the results suggest that in part the ''gender gap'' may be a g enerational gap most prevalent among ''baby boomers.''