The Latina Advantage: Gender, Race, and Political Success 9780292745650

Citation preview

The Latina Advantage

Bejarano-final.indb i

5/10/13 11:54 AM

THIS PAGE INTENTIONALLY LEFT BLANK

The Latina Advantage Gender, Race, and Political Success

Christina E. Beja r a no

University of Texas Press

Bejarano-final.indb iii

Austin

5/10/13 11:54 AM

Copyright © 2013 by the University of Texas Press All rights reserved Printed in the United States of America First edition, 2013 Requests for permission to reproduce material from this work should be sent to: Permissions University of Texas Press P.O. Box 7819 Austin, TX 78713-7819 http://utpress.utexas.edu/index.php/rp-form ∞ The paper used in this book meets the minimum requirements of ANSI/NISO ○ Z39.48-1992 (R1997) (Permanence of Paper).

Library of Congress Cataloging-in-Publication Data

Bejarano, Christina E., 1980– The Latina advantage : gender, race, and political success / Christina E. Bejarano. p. cm. Includes bibliographical references and index. ISBN 978-0-292-74564-3 (cloth : alk. paper) 1. Women legislators—United States. 2. Latin Americans—United States. I. Title. HQ1236.5.U6B46 2013 305.868′073—dc23 2012042768 doi:10.7560/745643

Bejarano-final.indb iv

5/10/13 11:54 AM

I dedicate this work to my family, for their continued support and love: my parents, Benita and Ramon Bejarano, my brother, Chris, and sister-in-law, Jessica Bejarano. Without their constant encouragement, I would not have been able to finish this book or achieve my goals. I am grateful for the strong and wise Latinas who have blessed my life and inspired my work: Benita Bejarano, Jennie Sagaribay, Ramona Hernandez, Sonya Arenivar, and Crystal Alba.

Do not go where the path may lead; go instead where there is no path and leave a trail. R a lph Wa l do Em erson

Bejarano-final.indb v

5/10/13 11:54 AM

THIS PAGE INTENTIONALLY LEFT BLANK

Contents

Acknowledgments

ix

Introduction. Challenges to the Double Disadvantage Theory

1

PART I. Attitudinal Advantages for Latinas

1. Gender and Racial Attitudes in Politics

15

2. Positive Interaction of Gender and Race/Ethnicity

32

PART II. Political and Electoral Advantages for Latinas

3. Predicting Latina Political Office-holding 4. Diverse State Legislators in Texas

53

73

5. Diverse State Legislators in California

101

Conclusion. Explanations for Latina Political Success Appendix

137

Bibliography Index

Bejarano-final.indb vii

131

165

179

5/10/13 11:54 AM

THIS PAGE INTENTIONALLY LEFT BLANK

Acknowledgments

I would like to thank Gary Segura for continuing to serve as an invaluable advisor to me since graduate school. I am thankful for Valerie Martinez-Ebers, who has always encouraged me to pursue my goals as a scholar. I express thanks to the University of Iowa faculty who served on my dissertation committee: Douglas Dion, David Redlawsk, Vicki Hesli, Maria Vélez, and Aimee Carrillo Rowe. I am grateful to Rene Rocha, Holley Hansen, and Stephen Nemeth for their help and support during my dissertation-writing time at the University of Iowa. I gratefully acknowledge the fi nancial support I received from the University of Iowa and the University of Kansas. I thank Trisha Blunt for her research assistance with one of the chapters. I am appreciative of my fellow University of Kansas faculty members, and I am especially grateful for the support and advice of Alesha Doan, Hannah Britton, Don Haider-Markel, Brent Steele, and Jessica Vasquez, who were helpful early in the book process. I thank Kira Sanbonmatsu for providing valuable feedback early in the book proposal stage. I also thank the anonymous reviewers and my editor, Theresa May, for their constructive help with this project.

Bejarano-final.indb ix

5/10/13 11:54 AM

THIS PAGE INTENTIONALLY LEFT BLANK

The Latina Advantage

Bejarano-final.indb xi

5/10/13 11:54 AM

THIS PAGE INTENTIONALLY LEFT BLANK

INTRODUC TION

Challenges to the Double Disadvantage Theory

The political landscape of the 2008 presidential election, with Barack Obama and Hillary Rodham Clinton as major party candidates for president, highlighted a number of important issues concerning the impact of race and gender. Questions arose regarding the predicted level of voter support for women and racial/ethnic-minority political candidates. Would a black or a female presidential candidate be electable in the United States? How does identity impact the level of support for diverse political candidates? This book intends to make a contribution to the political debate on these topics. As the electoral environment becomes more diverse, researchers are challenged to answer questions involving the influence of both gender and racial/ethnic identity. Is the public swayed by a politician’s race/ethnicity or gender as they decide whom to support? Further, how does the interaction of race/ethnicity and gender influence people’s decisions as they assess diverse politicians? How do demographic traits contribute to the continued underrepresentation of both minorities and women? Minority women, as both women and members of a minority racial/ ethnic group, may be treated according to both gender stereotypes and racial stereotypes and expectations. This combination of gender effects and racial effects can pose additional challenges for minority women in their attempt to reach electoral parity. Earlier work on minority women used a “double jeopardy” (Beale 1979) or multiple jeopardy (King 1988) approach to argue that various disadvantages can accumulate and produce distinctive forms of oppression. In line with this, previous work (Carroll and Strimling 1983; Epstein 1973; Githens and Prestage 1977) offered a “double disadvantage” hypothesis, which states that minority

Bejarano-final.indb 1

5/10/13 11:54 AM

2

The Latina Advantage

women will suffer politically from being women and from being a minority (Darcy, Hadley, and Kirksey 1997). This early research relied on evidence that “black women are less well represented than women generally or blacks generally,” which “appears to confirm the double disadvantage hypothesis: black women suffer politically from being women and from being black” (Darcy et al. 1997:448). There is more recent evidence that demonstrates that black women have made greater gains than white women within their respective electoral environments (Darcy and Hadley 1988; Moncrief, Thompson, and Schuhmann 1991). Research conducted in the late 1990s and early 2000s on Latina elected officials (Montoya, Hardy-Fanta, and Garcia 2000; Takash 1993; Hardy-Fanta 1993), suggested that Latinas are elected at higher rates than their male counterparts (García Bedolla, Tate, and Wong 2005:167). This evidence sparked debate over whether there are possible “overrepresentations” of minority female legislators in their respective minority delegations in the U.S. Congress, as compared to white females in their delegations. This debate includes the question of whether minority women legislators are uniquely positioned to benefit in the legislative process from the intersectionality of their ethnicity and gender, which Fraga et al. (2005) defi ne as their “strategic intersectionality.” I offer a unique contribution to the literature, by proposing a complimentary theory that explains how minority women political candidates, more specifically Latinas, can gain potential electoral advantages from the intersection of their gender and race/ethnicity. I argue and present evidence that counters the theory of double jeopardy or additive electoral disadvantages for minority women. Minority females encounter a positive interaction of their gender and race/ethnicity that results in fewer electoral disadvantages. As a result, they perform better electorally than minority men among some key voters. This interactive dynamic can help explain the dramatic electoral success of minority females in electoral politics in the last ten years.

Political Phenomenon for Latinas Evidence of Latinas’ potential diminished political disadvantages can be seen by examining the current representation of racial/ethnic-minority women in elective office compared to their male counterparts. As of 2005, women of color at the state and national levels made up a larger

Bejarano-final.indb 2

5/10/13 11:54 AM

Challenges to the Double Disadvantage Theory 3

proportion of their minority delegation, compared to their respective minority male counterparts, than white women compared with their white male counterparts (García Bedolla et al. 2005:166). Racial/ethnicminority women currently account for a greater share of minority political representatives than white women do of white elected officials, in both the U.S. Congress and state legislatures. In the 108th Congress (2003–2005), Latinas made up 29 percent of the Latino membership, black women made up 33 percent of the black membership, and Asian American females made up 29 percent of the Asian American membership (1989–1991) (García Bedolla et al. 2005:166). By contrast, white women were only 57 out of 463 total white representatives, or 12 percent of the white membership. In the 109th Congress (2005–2007), women made up 15.1 percent of all members of Congress, or 14 percent of senators and 15.4 percent of representatives (CAWP 2006b). In the 109th Congress eighty-one women were serving, of whom sixty-one were white women (which is 13.2 percent of the total white delegation) and twenty were women of color. This includes twelve African American women, or 28 percent, of the black delegation, seven Latinas, or 27 percent, of the Latino delegation, and one Asian Pacific Islander woman. In the 112th Congress (2011–2012), whose composition is shown in Figure I.1, women held ninety total seats in Congress, with seventythree in the House and seventeen in the Senate. White females were 73.3 percent of the female delegation, or sixty-six seats. In comparison, there were thirteen black females, four Asian females, and seven Latina females in Congress. Latinas first attained national electoral office in Congress in 1989, and have gained increased representation in state legislatures since the 1990s. The first Latina elected to Congress, in 1989, was Ileana RosLehtinen, a Cuban American from Florida (García Bedolla et al. 2005). Over the course of the 1990s, Latina representation in Congress increased 500 percent (from one to six) and their representation in state offices increased 280 percent (from sixteen to sixty-one) (Fraga and Navarro 2004:4). Latina office-holding grew more modestly at the county, municipal, and school-board levels, but at each level of government “Latina increases far outpaced increases in Latina/o representation overall” (2004:4). As of 2011, out of the twenty-eight Latino members of the U.S. Congress only seven were Latinas (from California, Florida, New York, and Washington); five of these were Democrats, and two were Republicans.

Bejarano-final.indb 3

5/10/13 11:54 AM

4

The Latina Advantage

112th U.S. Congress (2011–2012) 90 80 70 60 50 40 30 20 10 0 All Members

White

Latino

% Women

African American

Asian/Pacific Islander

% Men

Figure I.1. Gender in each racial/ethnic congressional delegation.

However, Latinas made up a higher percentage of state legislators nationwide. In 2011, 1,740 women state legislators served nationwide, of whom 347, or 20 percent, were women of color, and 1,393, or 80 percent, were white; the women of color included 241 African American women, 35 Asian American/Pacific Islander women, 63 Latinas, and 8 Native American women (CAWP 2001b). Racial/ethnic-minority women constitute 4.5 percent of the total 7,382 state legislators nationwide. In 2011, 63 out of 245 Latino state legislators were Latinas (CAWP 2011b).1 Twenty Latinas were in the state senates and forty-three in the lower house of the state legislatures; of these the majority, fifty-seven, 1. These women served Arizona, California, Colorado, Connecticut, Florida, Illinois, Indiana, Massachusetts, Maryland, Minnesota, North Carolina, New Hampshire, New Jersey, New Mexico, Nevada, New York, Rhode Island, Texas, Utah, Washington, and Wisconsin.

Bejarano-final.indb 4

5/10/13 11:54 AM

Challenges to the Double Disadvantage Theory 5

served as Democrats and six as Republicans. The states with the most Latinas in the state legislature were New Mexico (eleven) and Texas (eight). In Texas’s 2011 House, Latinas made up 20.7 percent of the Latino delegation. In California’s 2011 state assembly, Latinas made up 13.3 percent of the Latino delegation (with three Latinas). Overall, Latinas have increased their political presence across a wide variety of state legislatures and “at a rate that outpaces overall gender representation” (Fraga et al. 2006:129). Although Latinas are increasing their representation in elective office, they remain underrepresented compared to their population numbers. Having more women in elective office offers many benefits to the country: legitimate democratic government, and more diverse points of view and leadership styles (Sue and Wilcox 1998). Women and men also tend to have different life experiences and points of reference, which “can translate into a distinctive way of viewing existing legislative proposals and can lead to different agendas” (1998:2). Hawkesworth contends “that the mere presence of women of color in the U.S. Congress is transforming the institution, as they battle stereotypes of minority women and shape the public policy debate on issues pertaining to women and minority groups” (in García Bedolla et al. 2005:171). The growing number of Latina officeholders will also bring leadership differences to both politics and their communities. The presence of more Latina officeholders will have multiple effects on the political environment, including changes to political institutions and government policy. Fraga et al. studied the patterns of gendered representation among Latino state legislators in four states, and found that Latinas often “place greater emphasis on representing the interests of multiple minority groups, promoting conflict resolution, and building consensus in both the legislature as a whole and within the Latino caucus” (2006:122). Further, Latina legislators are also “more likely than Latino men to introduce and successfully pass legislation that addresses the issue agenda held by both Latina and Latino legislators” (122). Latinas have made significant strides at attaining political office in the last ten years, and yet researchers have not been able to explain this phenomenon. Latinas’ political success has surpassed expectations and explanations. Since there are no clear explanations, we cannot know whether this phenomenon will continue in the future or level off. I believe that these exciting political developments provide the perfect opportunity to study the interesting identity dynamics surrounding minority women political candidates.

Bejarano-final.indb 5

5/10/13 11:54 AM

6

The Latina Advantage

Political Phenomenon and Intersectionality My goal here is to offer an “intersecting identity” theory to explain this Latina electoral phenomenon. “Intersectionality” is used as a tool for studying the interactive effects of race/ethnicity and gender in electoral politics, and it challenges the common assumption of compounding electoral disadvantages for minority female political candidates. This identity theory does not assume that intersectional disadvantages do not exist; rather, it asserts that in this instance the electoral resources that come with the intersecting identities uniquely lessen the effect of each additional disadvantage. Interactive identities provide fewer electoral disadvantages and enable Latinas to more readily attain electoral support. In other words, Latinas, as minority women, can capitalize on the crossover appeal gained from “their multiple community identifications” (Smooth 2006:411), which provides them with more potential voters. Several competing hypotheses are set out to explain the positive interactive effects of race/ethnicity and gender for Latinas in political office. Latina political candidates are positively affected by two factors: advantages in terms of attitudinal support and advantages in terms of increased candidate-quality. The attitudinal support results from the softening of perceived racial threat, which provides Latinas with fewer disadvantages in relation to the disadvantages that still affect their male counterparts in respect to election to political office. This challenges scholars to rethink the common assumptions that have been made about female political candidates and demonstrates that gender stereotypes or expectations do not apply equally to all females regardless of their race and ethnicity. This study of Latinas, therefore, offers new alternative possibilities to research on women and electoral politics. The study also offers a more comprehensive examination of the current political status of racial/ethnic minorities, by examining the political characteristics and political outcomes of both minority women and men. Latinas can encounter political advantages from their intersecting identities and successfully attain legislative office.

Overview and Argument The existing literature on women in politics does not provide convincing explanations for minority women’s success in political office. The

Bejarano-final.indb 6

5/10/13 11:54 AM

Challenges to the Double Disadvantage Theory 7

work only examines women candidates in general and provides competing ideas about women’s underrepresentation in political office. The unsettled debate in the literature concerns the question of women’s electoral parity—whether women still face obstacles to representation compared to men. One side of the debate includes researchers who claim that women no longer suffer the same electoral disadvantages that have hurt them in the past and are therefore on equal standing with men. The researchers do not find evidence of gender-based obstacles relating to the campaign, nor can gender alone explain differentials in winning percentages, vote totals, and fundraising receipts (Burrell 1998; Seltzer, Newman, and Leighton 1997; Chaney and Sinclair 1994). The other side in the debate includes researchers who argue that women’s drastic underrepresentation in electoral office continues as a result of gender stereotypes and gender biases in the political and electoral system (Fox 1997; Kahn 1996; Huddy and Terkildsen 1993a and 1993b). The stereotypes and biases disadvantage women and create multiple obstacles to their electoral success. This debate remains unsettled mainly because of the scarcity of research that focuses on the complex circumstances that women face in the electoral environment. Taking a position within this debate, I hypothesize that women’s underrepresentation in electoral office as compared to men is related to disadvantages in terms of voter biases and campaign obstacles. In evaluating this hypothesis, I follow Duverger’s valuable framework (1955) of the three main obstacles that women face in the electoral environment: voter hostility, male conspiracy (lack of party leaders’ support), and electoral arrangements (Duverger 1955; also Thomas and Wilcox 1998 and 2005). The addition of race/ethnicity to the equation can change the outcome of the standard electoral obstacles encountered by women. An interesting new trend in this gender research has introduced evidence that demonstrates that female political incumbents are increasingly considered to possess higher candidate-quality than male incumbents (Milyo and Schosberg 2000). Unfortunately, the sobering explanation for this finding is that female political candidates often underestimate their qualifications and therefore only highly qualified females are elected to political office (Anzia and Berry 2011; Fox and Lawless 2004; Fulton 2011). I extend this new trend in gender research by examining the electoral experiences of racial/ethnic-minority women. Minorities are still underrepresented in political office in relation to the majority population (Hero and Tolbert 1995; Rule 1994; Pachon and

Bejarano-final.indb 7

5/10/13 11:54 AM

8

The Latina Advantage

DeSipio 1992). Grofman identifies principal barriers to greater representation for minorities as “electoral arrangements (including at-large elections and gerrymandered districts) and the continued reluctance of white/Anglo voters to vote for minority candidates” (Grofman, Handley, and Niemi 1992:1). The existing literature on minority women in electoral politics also does not provide any generalizable answers; rather, extant work predominantly examines case studies of minority women or only examines one particular racial/ethnic group at a time. For instance, the double disadvantage hypothesis has mainly been examined for black women and has not yet been tested among other groups of minority women. As mentioned earlier, it has been argued that the main reason for Latina electoral success comes from the softening of the racial threat factor and their increased candidate-quality characteristics. This book introduces a new twist to the theory of racial threat as it applies to white voters’ decreasing support of minority candidates, by examining how minority women political candidates encounter changing campaign dynamics through the intersectionality of their gender and race/ethnicity. On the one hand, the negative effects of stereotype and bias on women and racial/ethnic minorities are undeniable, and they uniquely disadvantage minority women. For example, minority women can be disproportionately affected by the experience of racism and sexism due to labor market segregation and relative economic disadvantage (Reynoso 2004). On the other hand, the possibility of softening perceptions of threat and cross-group gender solidarity may actually reduce the net effect of race and gender. As a result, minority women may not experience the same level of hostility or negative racial stereotypes concerning their race/ethnicity. This analysis also engages in the contemporary debate over female candidates and aspects of their candidate-quality (Lawless and Fox 2005). It offers a more elaborate examination of candidate-quality characteristics that highlight the contributions that minority women bring to politics. The characteristics include the candidate’s previous membership or leadership roles in community organizations, previous political experience in lower-level political offices, and other preparatory background characteristics. Latinas, with the help of these two political factors, experience fewer electoral disadvantages through the availability of more voter coalitions. Focusing specifically on Latina electoral behavior is useful because their style of politics is a departure from traditional style. Latinas have a more

Bejarano-final.indb 8

5/10/13 11:54 AM

Challenges to the Double Disadvantage Theory 9

participatory vision of politics than Latino men, which leads Latinas to take a more successful community-oriented approach to political mobilization and participation efforts of Latinos (Hardy-Fanta 1993). Overall, this study of Latinas contributes needed data to the relatively little research done to date that specifically examines the experiences of minority women in politics, across ethnic groups and states.

Overview of Chapters This examination of the intersectionality of race/ethnicity and gender for political candidates crosses multiple disciplines, including political science, race/ethnicity studies, and gender studies. There is growing curiosity about the role of intersectionality of race/ethnicity and gender in U.S. politics, as well as a desire to explain the increasing number of Latino (Casellas 2010) and minority female officeholders (Hardy-Fanta et al. 2006). The following chapters engage in several contemporary political debates, such as the renewed scrutiny of the role of women’s candidate qualifications and office-holding, and the popular and scholarly interest in Latino politics. The 2010 Census data show that the country’s Latino population represents 16.3 percent of the total population, and grew by 43 percent in the last decade. This population growth has only fueled the interest in Latino politics. The book is divided into two parts; the first includes an investigation of the public’s willingness to support women and minorities in politics, by using two national public opinion datasets, and the second attempts to identify the political factors that would contribute to Latina electoral success. Chapter 1 uses the General Social Survey (GSS) cumulative data from 1972–2010 to evaluate the public’s attitude toward the role of women and minorities in politics. The questions assess the respondents’ attitudes toward the role of women, women’s participation in politics and business, and their willingness to vote for a woman for president. The chapter then considers questions that assess the respondents’ view of the role of minorities in politics. The results demonstrate that white women are more willing than similarly situated white males to cross racial barriers and support a minority candidate for political office, in this instance a black candidate for president. White women are more likely than white men to support progressive racial attitudes, which can lead

Bejarano-final.indb 9

5/10/13 11:54 AM

10 The Latina Advantage

to their support of different types of minority candidates for different levels of office. Women in general are also more likely to support progressive gender attitudes, which can lead to more support for female political candidates. This chapter also demonstrates that minorities provide support for the importance of co-ethnic candidates, with minorities showing increased support for a black candidate for president. Chapter 2 uses national public opinion data from the American National Election Studies (ANES) to examine the public’s support of women and minority U.S. incumbent House representatives in the period 1992–2008. The analysis examines the respondents’ support of their congressional incumbent. The argument is that on average minority female incumbent national legislators receive more support than minority male incumbent national legislators. This chapter introduces strong explanations for Latinas’ increased political success by accessing multiple voting coalitions, showing that if the respondents do not share the race/ethnicity of their incumbent, they will overcome their potential racial and gender bias and favor the female incumbent. As a result, minority females receive increased support from their co-ethnic minorities and from white females over minority male incumbents. The intersection of race/ethnicity and gender does not create a double disadvantage for minority females in electoral politics; instead it creates a positive interaction and electoral success. Moving to part II, we find in chapter 3 an examination of the statelevel factors that influence Latina political office-holding across the country. No previous studies look at the political experiences of Latinas across states. Earlier research focuses on white women, black women, or racial/ethnic minority-women in the aggregate. This chapter looks at the experience of Latino male and female state legislative incumbents across thirty-six U.S. states. My analysis explores political factors that can explain the variation in the number of Latina state legislators across the country. Chapters 4 and 5 move to Texas and California, where there is clear evidence regarding the significant political advances made by racial/ethnic minorities since the 1960s and 1970s. Although Texas has the most Latino state legislators and a high number of Latina elected officials, there are no Texan Latinas in the U.S. Congress. In contrast, California has the most Latinas in the U.S. Congress. I offer two unique datasets, encompassing all candidates who ran for the Texas and California lower houses in a single election year (2004), thus facilitating direct comparisons. The datasets provide support to the argument that minority fe-

Bejarano-final.indb 10

5/10/13 11:54 AM

Challenges to the Double Disadvantage Theory 11

male political candidates benefit from their “high-quality” characteristics. Chapter 4 provides evidence of increased electoral support in Texas for the minority candidates running for state legislature, especially for Latinas, while chapter 5 shows that campaign contributions proved to be the most important deciding factor influencing the electoral support of all candidates in the California election. A closer examination of the specific political races in the two states reveals clear differences between the level of candidate-quality characteristics and importance of district makeup for candidates in both Texas and California. Candidate-quality characteristics helped successful Latino candidates overcome potential district obstacles, such as districts that did not include a majority-minority population (wherein the majority of individuals are of racial or ethnic minorities) or a district makeup that did not include their political party in the majority of the population. The case studies also demonstrated that the successful Latinas had more previous political experience, community experience, and higher campaign contributions than their opponents. In the conclusion I affirm that explanations for the Latina political phenomenon come from a variety of sources. Attitudinal factors enable Latinas to receive increased electoral support. The fi ndings introduce a twist to the theory of double disadvantages, wherein there is a positive interaction of race/ethnicity and gender that results in fewer electoral disadvantages. My findings also add a new dynamic to the existing debate on the political status of women, since the Latina political experience differs from that of white women. My work also calls for a reexamination of the concept of racial threat to include the impact of gender. It advises scholars to rethink the common assumptions made about minority political candidates, since minority stereotypes do not apply equally to all minorities regardless of their gender. It would be useful for future studies to incorporate a wider variety of characteristics, in order to identify the qualities that help minority female candidates emerge as more appealing to the voters.

Bejarano-final.indb 11

5/10/13 11:54 AM

THIS PAGE INTENTIONALLY LEFT BLANK

CHAPTER 1

Gender and Racial Attitudes in Politics

Voter bias is an influential dynamic that has the potential to impede the electoral success of both women and minorities in the United States. Previous researchers have focused their analysis of voter bias and political disadvantages on either women or, separately, racial/ethnic minorities. Bias can also have an interactive influence, with gender influencing pro-social (or progressive) racial attitudes and race influencing progressive gender-role attitudes. Racial/ethnic minorities and women are politically influenced by their shared experiences of discrimination in society at large. This chapter includes an examination of racial/ethnicminority female and white female racial and gender attitudes, as a means of assessing whether these groups are more likely than white males to support racial/ethnic minorities and females in political office.

Impact of Gender on Progressive Racial Attitudes Some early studies conducted mainly by sociologists and psychologists examined the impact of gender on social tolerance. Additional studies considered the influence of gender on support for progressive government racial policies (Poole and Ziegler 1985; Steeh and Schuman 1992; Bobo and Kluegel 1993; Tuch and Hughes 1996; Kane and Kyyro 2001). Gender has also been linked to social distance questions (Schuman et al. 1997; Johnson and Marini 1998). A predominant strand in this literature derives from social structural theories (Blumer 1958), which combine race and class to explain the social positions or attitudes of a racial group (Hughes and Tuch 2003). So-

Bejarano-final.indb 15

5/10/13 11:54 AM

16

Attitudinal Advantages for Latinas

cial structural theories focus on the group members’ perceived relationship to other groups, which influences their racial attitudes. White men and women should have the same racial attitudes since they share the same racial position (2003:385). For example, a white woman’s position in the white racial group should be more important than her experience as a woman in determining her racial attitudes (386). Contrary to this social structural hypothesis, another strand in the literature highlights a significant gender difference in whites’ progressive racial attitudes (Johnson and Marini 1998; Schnittker 2000; Hughes and Tuch 2003): white women expressed more favorable racial attitudes than white males (Johnson and Marini 1998:247). This key difference has been attributed to women’s generally more pro-social attitudes, which include their views on race (1998). Women showed a greater concern for others and focused on relationships, which fostered more favorable racial attitudes (1998:247). Further, this group of researchers (Baxter and Lansing 1983; Shapiro and Mahajan 1986; Steeh and Schuman 1992) found that women are also generally more liberal on “racial policy aimed at achieving equality” (Johnson and Marini 1998:248). Gender has been shown to have a significant impact on progressive racial attitudes relating to politics. However, previous research focused primarily on the impact of gender on whites’ racial attitudes, and should be expanded to include racial/ethnic-minority groups in this examination.

Impact of Race on Progressive Gender Attitudes Next, I turn to another key dynamic: the impact of a person’s race or ethnicity on her or his gender-related attitudes. An analysis of gender attitudes typically includes questions that assess a person’s attitude toward set gender roles in society. Gender-role attitudes range from very strict and traditional ideas about the “proper place” for men and women up to nontraditional, or egalitarian, attitudes. “Traditional sex-role socialization” refers to women being taught to be passive individuals who should focus on family responsibilities and building a strong home life, while men are taught to be assertive, independent, and goal-oriented. Enacting these distinct roles can cause women to have lower levels of political interest and aspiration because they are taught by society to view their roles differently. From this perspective politics can be considered to be a man’s world and inappropriate for women.

Bejarano-final.indb 16

5/10/13 11:54 AM

Gender and Racial Attitudes in Politics 17

Many previous findings on gender-role attitudes among racial/ethnic minorities are often contradictory (Kane 2000). The first group of research includes authors that argue that Latino men (Wilkie 1993) and women (Harris and Firestone 1998), as well as black men (Sigelman and Welch 1984; Wilcox 1992), all demonstrate more traditional genderrole attitudes than other racial groups. Ransford and Miller analyzed the General Social Survey data on gender-role attitudes, as well as race and gender differences in sex-role outlooks (1983). They found no difference between black and white women’s feminist outlooks. However, they did find that black men had more traditional views than white men in terms of the view that “the woman’s place is in the home” and concerns about the “emotional suitability of women in politics” (Ransford and Miller 1983:51). Another group of researchers argues that minority men and women do not demonstrate significantly more traditional gender-role attitudes than whites (Kane 2000). Further conclusions have been that male dominance is not significantly greater among Latinos than whites (Baca Zinn 1980; Gonzalez 1982; Montoya 1996; Vazquez-Nuttall et al. 1987; Ybarra 1982), and that racial/ethnic-group membership can shape gender-related attitudes (Kane 2000:422). Additionally, blacks do not have more traditional gender-role attitudes than other groups (Hatchett and Quick 1983; Hunter and Sellers 1998; Welch and Sigelman 1989), and in fact demonstrate more egalitarian gender-role attitudes than whites (Kane 2000:422). This racial dynamic has also been examined in respect to specific political questions (Ransford and Miller 1983; Sigelman and Welch 1984), including public support of a female candidate for president. Previous research does not find significant racial differences in men’s attitudes to this issue (Kane 2000:425). However, African American men appear generally less likely than African American women to support a woman candidate for president (Sigelman and Welch 1984). Racial/ethnic differences in gender attitudes can be attributed to blacks’ increased experience with inequality and their history of moreegalitarian family practices (Kane 2000:426). This experience can lead blacks (men and women) to “greater criticism of gender inequality in the country” (2000:426). Other minority group members’ additional experiences with racial and ethnic discrimination can also lead to greater awareness of gender inequality (427). This increased understanding can produce more support for social action among minority women and black men (Fulenwider 1980), through such devices as the women’s

Bejarano-final.indb 17

5/10/13 11:54 AM

18

Attitudinal Advantages for Latinas

movement (Kane 2000:428). It is clear that race and ethnicity can have a significant impact on progressive gender attitudes relating to women’s role in politics.

Theory and Argument There are interesting dynamics involved in exploring the influence of a person’s gender or race on their own racial or gender attitudes. Gender can have a softening influence on racial views among whites, which can lead to white females’ increased support of progressive racial attitudes. Race/ethnicity can also have a softening effect on gender attitudes of minorities, which can lead to minorities’ increased support of progressive gender attitudes. These two theories on gender and racial attitudes are combined in this chapter in order to examine the interactive effects of both. First it is necessary to discern the respondents’ general attitude toward minorities in politics and their racial attitudes, and then test for gender differences in relation to racial/ethnic groups and racial differences in relation to gender. The expectation is that minority respondents will be more supportive of progressive racial attitudes than white males. Further, white females are expected to be more sympathetic than white males to minorities’ attempts to enter politics. This leads to hypothesis 1: H1: White women are more likely than white men to hold progressive racial attitudes.

Next, the respondents’ general attitude toward women in politics— their gender attitudes—must be determined to test for gender and racial differences among all racial/ethnic groups. Females are expected to be more sympathetic than males to females’ attempts to enter politics. Minority respondents are also expected to be more supportive than white males of progressive gender attitudes, which is hypothesis 2. H2: Minorities are more likely than white men to hold progressive gender attitudes.

This analysis provides timely insight on the public’s attitude toward underrepresented groups in politics. It offers support to the theory that

Bejarano-final.indb 18

5/10/13 11:54 AM

Gender and Racial Attitudes in Politics 19

minority female candidates, including Latinas, can expect to receive increased electoral support from both minorities and white females.

Data National public opinion data from the General Social Survey (GSS) cumulative dataset—compiled from 1972 until 2010—is used to test the two hypotheses. The GSS is an annual interview survey of U.S. households conducted by the National Opinion Research Center (NORC). This data “tracks the opinions of Americans over the last four decades” (GSS 1998). This analysis uses weights to provide an accurate account of the makeup of the national population. Racial Attitudes My first analysis investigates the influence of gender and race/ethnicity on respondents’ progressive racial attitudes. The GSS question seeking to determine the respondent’s willingness to vote or support blacks in politics is worded as follows: (a) If your party nominated a (black/African American) for president, would you vote for him if he were qualified for the job? This question is coded as a dichotomous (“yes”/“no”) variable, with “one” signifying support of a black candidate for president. This is the type of question predominantly used in all national surveys to discern the level of support for any minority in politics; it is also the only question of its kind currently available in surveys. This question was previously only hypothetical, but became realistic with the 2008 election of President Barack Obama. The survey goes on to present four questions that assess the respondent’s racial attitudes toward various racial/ethnic groups, testing for gender/racial differences in the process. These questions were used by the GSS for their 2002 wave and include (b) Feelings towards Latinos, blacks, Asians, or whites. They are recorded on a scale of “nine”—representing very warm feelings toward the particular racial/ethnic group— to “one”—representing very cool feelings. Gender Attitudes The second section of the survey utilizes several questions that assess the respondents’ view of the role of women in politics, or their progres-

Bejarano-final.indb 19

5/10/13 11:54 AM

20

Attitudinal Advantages for Latinas

sive gender attitudes. The GSS question asks about the respondent’s willingness to support a female candidate for President: (a) If your party nominated a woman for president, would you vote for her if she were qualified for the job? The question is coded as a dichotomous variable, with “one” signifying support for such a candidate. This question is also the one used in all national surveys to examine the support of women candidates. Further, it is now more realistic, after Hillary Clinton’s candidacy for president in 2008, to ask the public if they will support a female candidate for president. Three additional GSS questions focus on gender-role attitudes in respect to women working outside of the home and participating in politics and the community. Support for females in politics is measured by pro-female or progressive gender responses to the following questions: (b) Do you agree or disagree with this statement? Women should take care of running their homes and leave running the country up to men. [progressive gender response is “disagree”] (c) Do you approve or disapprove of a married woman earning money in a business or industry if she has a husband capable of supporting her? [progressive gender response is “approve”] (d) Tell me if you agree or disagree with this statement: Most men are better suited emotionally for politics than are most women. [progressive gender response is “disagree”]

The gender-role questions were previously used in other studies (Barnartt and Harris 1982; Ransford and Miller 1983; Harris and Firestone 1998) to examine gender-role ideology; intercorrelations among the items range from 0.17 to 0.54. The four gender questions have not been combined into an index for this analysis and instead are examined separately to provide a clearer idea of individual progressive gender attitudes. The four questions provide the most appropriate avenue for assessing gender views, since they are the only questions of this kind that ask about support for women in politics. The responses to the four questions will be used to test for the racial and gender differences in progressive gender attitudes among all the respondents. Independent Variables Descriptive identity characteristics of each respondent—gender and race/ethnicity—are used to provide a clearer picture of how identity can

Bejarano-final.indb 20

5/10/13 11:54 AM

Gender and Racial Attitudes in Politics 21

influence gender and racial attitudes. Dichotomous variables are coded for “white female,” “white male,” “black female,” “black male,” “‘other race’ female,” “‘other race’ male,” “Latina,” and “Latino male.” The dataset from the years under study includes 24,260 (44 percent) male respondents and 30,827 (56 percent) female respondents. The racial makeup is 44,873 white respondents (81.5 percent), 7,625 black respondents (13.8 percent), and 2,589 “other race” respondents (4.7 percent). The GSS began including an ethnicity question for “Hispanic (Latino)” in their 2000 wave; the dataset includes 1,715 of the Latino respondents (3.1 percent). There are other demographic variables that can influence a respondent’s support for either progressive racial or progressive gender attitudes. Marital status is expressed as a dichotomous variable for “married”; the dataset includes 29,861 married respondents (54.2 percent). Age is an ordinal variable with three categories of age groups, including “one” for “18–34 years” (32.4 percent), “two” for “35–50 years” (30.4 percent), and “three” for “51+ years” (37.1 percent). The respondent’s highest level of education is coded as an ordinal variable with “zero” representing “some high school education” (22.5 percent), “one” for “high school degree” (51.6 percent), “two” for “some college” (5.3 percent), “three” for “college degree” (13.9 percent), and “four” for “graduate school degree” (6.7 percent). Income is coded as an ordinal variable with “one” representing “low income” (less than $10,000/year) (29.1 percent), “two” for middle income” ($10,000–24,999/year) (35.1 percent), and “three” for “high income” ($25,000/year and higher) (35.8 percent). Religion is another potential influence on gender/racial attitudes and is expressed in this analysis as the two most reported religious identifications, with dichotomous variables for “Protestant” (59.1 percent) and “Catholic” (24.5 percent). Gender and racial attitudes can also be influenced by political party affiliation; the dataset includes political party dichotomous variables for “Democrat” (37.4 percent), “Republican” (25.8 percent), and “Independent” (35.4 percent).

Data Analysis and Results: U.S. Gender and Racial Attitudes The influence of gender and race/ethnicity on both progressive racial and gender attitudes can be traced over time by means of the GSS data from 1972 until 2010. The 1960s and 1970s saw dramatic changes in both racial and gender attitudes in U.S. society. The feminist and civil rights movements challenged traditional gender roles and racially biased

Bejarano-final.indb 21

5/10/13 11:54 AM

22

Attitudinal Advantages for Latinas

attitudes. Therefore, it is expected that respondents will hold more progressive gender and racial attitudes after the 1970s. Support for Black and Female Candidates for President White female respondents show a higher level of support, in Figure 1.1, for a black candidate for president than do white male respondents, and the level of white support also increases steadily from 1972 through 2010. Black and “other race” male respondents started off with less support than white male respondents for a female candidate for president in the 1970s, but come to report higher levels of support after 2000. The respondents’ level of support for both a black and a female candidate for president has increased steadily over time for each racial/ethnic group (see also Table A.1 in the appendix). Next, a multivariate analysis—utilizing data on respondent attitudes from the GSS for 1982–2010—provides a stronger idea of progressive gender and racial attitudes after the 1970s. Logistic regression was used to examine the respondents’ likelihood of supporting a black and a female candidate for president, and the results are reported in Table 1.1, with unstandardized logit regression coefficients, standard errors, and the estimated changes in predicted probabilities. White male respondents, compared to the other racial/ethnic groups and white females, are less likely to support a black candidate for president. White females and “other race” females are about 2 percent more likely than white males to support a black candidate for president, supporting H1. Black females and black males are also 6.4 to 7.1 percent more likely than white males to support a black candidate for president. The relationships hold even when we control for the demographic variables used in the model, including the respondent’s marital status, age, education, and religion. Protestants and older respondents are less likely to support a black candidate for president. Respondents who are married and have a higher education are positively associated with supporting a black candidate. Democrats and Independents are also more likely than Republicans to support a black candidate. The respondents in the later GSS waves are more likely than respondents in earlier waves to support progressive racial attitudes toward a black candidate for president. White females and black females are also more likely than white males to support a female candidate for president, by about 2 percent. “Other race” males are less likely than white males to support the female candidate, by 5.5 percent. Respondents who are older and Prot-

Bejarano-final.indb 22

5/10/13 11:54 AM

Gender and Racial Attitudes in Politics 23

White Support for a Black Candidate for President 100 95 90 85 80 75 70 1972–1978

1980–1989 White Female

1990–1998

2000–2010

White Male

Male Support for a Female Candidate for President 100 95 90 85 80 75 70 65 60 1970s White Male

1980s Black Male

1990s

2000s

Other Race Male

Figure 1.1. Support for a black and female candidate for president, 1972–2010.

Source: GSS, 1972–2010.

estant are negatively associated with support for a female candidate for president. Respondents who have a higher education level and a higher income, as well as being Democratic or Independent, are also positively associated with supporting a female candidate for president. The year of the GSS wave is again a significant factor, with respondents in the most recent GSS waves, closer to the year 2010, reporting more progressive gender attitudes toward a female candidate for president than in earlier waves.

Bejarano-final.indb 23

5/10/13 11:54 AM

Table 1.1. Respondents’ Support of Black and Female Candidates for President, 1982–2010 Support Black Candidate for President Model White female Black female Black male “Other race” female “Other race” male

Min → Max

.323*** (.079) 2.33*** (.286) 3.52*** (.583) .506* (.304) .053 (.231)

.018 .064 .071 .023 .003

Support Female Candidate for President Model .436*** (.085) .401*** (.156) .243 (.166) −.087 (.254) −.736*** (.196)

Min → Max .023 .019 .012 −.005 −.055

Demographics: Married Age Democrat Independent Education Income Protestant Catholic

.183** (.074) −.089* (.047) .187** (.089) .367*** (.088) .552*** (.042) .016 (.053) −.355*** (.114) −.008 (.126)

.011 −.010 .010 .020 .114 .002 −.020 −.001

−.049 (.075) −.269*** (.046) .428*** (.091) .549*** (.091) .313*** (.039) .294*** (.053) −.294** (.115) −.026 (.128)

−.003 −.031 .022 .028 .063 .033 −.016 −.002

Time: GSS Year Constant Pseudo R 2 Chi-square N

.410*** (.057) .327 (.218) 0.0932 581.73*** 10,412

.042

.473*** (.060) .549** (.225) 0.0605 380.77*** 12,294

.047

Significance: *p