Candidate image : factoring in gender

Includes bibliographical references (pages 41-45 ).This research seeks to reveal ways in which political candidate image is influenced by gender; the gender both of candidates and voters. It also explores the joint role of party affiliation and gender in determining candidate image. Improved understanding of the function of gender in determining candidate image may be useful in closing the male-dominated gender gap that characterizes American political arenas currently. Yet, few studies have directly examined the relationship between candidate image and gender in the context of electoral politics. And most studies that have assessed these relationships have been limited by the use of actual candidates rather than supposed candidates. It is likely the results of such research were influenced, to some degree, by fixed notions about the actual candidates. In this study, candidate image was measured along six well-established dimensions using 29 seven-point semantic differential items. The political affiliation and gender of subjects were collected also. Five-hundred-forty actual voters leaving randomly-selected polling sites in the City of San Diego during the 2000 presidential election participated in the study. Voters were asked to examine one of two political mailers; reflecting either a male or female candidate. Factor analysis was conducted on the semantic differential items followed by reliability tests on each set of variables grouped together by factor analysis. Cronbach's alpha reliability coefficient was used to eliminate items that negatively influenced the reliability of the factor scales. The resulting factors were interpreted based on underlying construct differentiating the grouped factors. Factors were averaged and a single composition score assigned for each subject in each set of factors. Once completed, analysis of variance was used to test the hypothesis according to a 95 percent decision rule. The results disconfirmed all three hypotheses: (1) Voters were expected to rate the image of the female candidate lower than their male counterpart; (2) Female voters were expected to rate the female candidate higher than male voters; and (3) Self-identified Republican voters were expected to rate female candidates lower than Democrat voters. According to the findings, gender and political affiliation did not have significant influence on voter perceptions. These results supported an existing rival theory; women are discouraged and/or restricted from becoming political candidate and therefore, rarely make it on the ballot