While previous research indicates that voters hold gender-based stereotypes of women and men candidates for elected office, the degree to which candidate actions contribute to these views is less well known. The research reported here attempts to determine whether candidates appeal to gender-based stereotypes by choosing to campaign on issues that are in line with voter expectations. Specifically, it examines whether women candidates for Congress in 2000 and 2002 presented campaign issues that were different from those presented by their male opponents and whether these issues conform to expected gender stereotypes, and then compares these findings to that of a men-only race comparison group. Content analysis of campaign web sites is employed to examine the campaign images presented by these candidates. Contrary to assumptions, women in 2000 and 2002 did not focus their issue priorities on a set of gender-stereotyped issues, but instead campaigned on a set of topics that were similar to those of their male opponents.
Even though women candidates are becoming more common in American politics, how they are perceived by voters is often an open question. One thing that research does show clearly is that the public often sees women candidates through the lens of gender stereotypes. Women candidates are thought to be more compassionate and honest than men, while men are viewed as stronger leaders and better decisionmakers (Alexander and Andersen 1993; Huddy and Terkildsen 1993; King and Matland 2000; Koch 1999). In terms of issues, the public tends to see women as more liberal than men, more interested in issues that affect women, children, education, and health, and less focused on issues such as business, the economy, and foreign affairs (Brown, Heighberger, and Shocket 1993; Kahn 1992; McDermott 1998). The focus of this research is to examine whether women candidates for Congress contribute to these impressions by campaigning on a set of issues that conform to these gender-based stereotypes. For this purpose, an analysis of the campaign websites of women and men candidates for the U.S. House of Representatives and Senate in 2000 and 2002 will be employed.
THE POTENTIAL IMPACT OF STEREOTYPES
While demonstrating that voters employ these genderbased stereotypes is relatively straightforward, evaluating their impact is less clear. Indeed, use of these stereotypes can cut both ways for women candidates, leaving them to face what Jamieson (1995) refers to as a "double bind." Generally, work on women candidates has shown that voters favor more "male" traits in their elected leaders, which may lead them to choose men candidates over women (Huddy and Terkildsen 1993). But, in 1992, the outsider status of many women candidates actually worked in their favor, as many voters who were fed up with political incumbents chose women, whom they perceived as more honest and concerned about the public good.
Issue stereotypes can work the same way. Analysts of congressional elections point to the elections of 1990 and 1992 as examples of how vote choice can depend on which issues the public values at the time of the election. In 1990, as the U.S. prepared for the Gulf War, foreign affairs, a set of issues on which women candidates are given little credibility, were central to political discourse. In 1992, however, when a record number of women were elected to the House and Senate, domestic issues had surged to the fore. While not the sole explanation for the increase in women elected to Congress in 1992, it is clear that voters who were concerned with these issues were more likely to choose women candidates, perhaps because these voters assumed women would give them more attention (Dolan 1998).
Because of the complexities involved in the public's perception of women candidates, decisions about how to present themselves to voters can be tricky for these women. Running "as a woman" and playing to the stereotypes the public holds can work in their favor when these attributes are valued by the public. …