Voting the Gender Gap

9. The Gender Gap: A Comparison across Racial and Ethnic Groups

9 The Gender Gap A Comparison across Racial and Ethnic Groups m. margaret conway The term “gender gap” has been used to describe differences between men and women in vote choice, voter turnout, other types of political participation, policy preferences, and public opinion differences. Regardless of the topic studied, almost all research focusing on the United States has examined the gender gap either exclusively among whites or just within the white and African American groups. In part this paucity of research can be attributed to the small numbers of minority group members included in most national surveys. Only a few studies have expanded analysis of the political gender gap to include Hispanics, Asian Americans, and Native Americans. The extent to which a political gender gap occurs within different racial and ethnic groups merits further study. This article focuses on the gender gap among several racial and ethnic groups in the United States, comparing patterns found within those groups. It first reviews prior research on the political gender gap; then the gender gap patterns within white, African American, Hispanic, Asian American, and American Indian racial or ethnic groups are examined, using data from several surveys conducted between 1990 and 2004. Prior Research Survey data collected since 1952 in the American National Election Studies (ANES) provide evidence for the existence of the political gender gap among whites and of change over time in its patterns. For example, prior to 1964, women more frequently voted for Republican Party presidential candidates i-xii_1-220_Whit.indd 170 1/10/08 9:18:00 AM m. margaret conway · 171 than did men.1 After 1964, women were more likely than men to vote for the Democratic Party presidential candidates.2 Examining the gender gap in the 1992 presidential election, Franco Mattei suggests the most powerful contributor to the gender gap in voting behavior is beliefs about the proper role of federal government.3 Other research based on survey data collected during the 1984, 1988, and 1992 elections points out that different levels of salience on economic issues contributed to the gender gap in voting behavior.4 Since 1980, a gender gap has existed in voting for members of the House of Representatives, with women being more supportive of Democratic Party candidates.5 Previous research suggests that a gender gap also exists in other forms of political participation. In their study of political activity, Burns, Schlozman, and Verba found whites to be more politically active than African Americans , who were more politically active than Latinos. Within each racial or ethnic group, men were more politically active than women.6 However, the disparities between groups in factors affecting political activity were structured more by race or ethnicity than by gender.7 Differences have also existed in patterns of party identification. Since 1964, a greater proportion of women than men have identified with the Democratic Party, and, since 1988, a greater proportion of men than women have reported a Republican Party identification.8 Other studies have examined political ideology and party support among racial and ethnic groups. A study of Hispanic Americans using data from exit polls conducted in 1980, 1984, and 1988 concluded that Hispanic women were more liberal and more supportive of the Democratic Party than were Hispanic men.9 Another study of the gender gap suggests that the shift in male-female differences in both partisanship and presidential voting may be explained by differences in policy attitudes. An alternative explanation points to the shifts occurring because of different weights that men and women apply to selected attitudes when making decisions about partisan identification and presidential vote choice. Research using survey data from the 1992 and 1996 election studies provides evidence that both explanations contribute to understanding citizens’ choices .10 Other factors besides the gender gap, including educational attainment, level of income, religious affiliation and religiosity, and organizational memberships , appear to influence party identification.11 Differences in men’s and women’s evaluations of presidents also exist. What might explain a gender gap in presidential evaluations? An examination of variations in support for Ronald Reagan, using data from the 1982 NES study, concludes that the most important component of the gender gap i-xii_1-220_Whit.indd 171 1/10/08 9:18:00 AM 172 . the gender gap in presidential approval was generated by differences in opinions on social and military issues.12 These differences in political attitudes may arise from a variety of sources...