It’s more than just ‘how women think’: explaining the nature and causes of gender gaps in political preferences

Women hold political beliefs that differ systematically from those of men, a phenomenon scholars call “gender gaps.” The collective opinions of women tend to favor social welfare policies and spending on these policies, while women oppose the use of force at home and abroad. Far from being inconsequential, empirical research indicates that these gender gaps in political preferences create gender gaps in vote choice and party identification (Kaufmann and Petrocik, 1999; Norrander, 1999; Kanthak and Norrander, 2004). The partisan gender gap has done a great deal to shape modern American politics. A majority of the last five presidents won office with thin margins of victory. More women identifying with the Republican Party or its candidates would have changed electoral returns and, in some cases, the winning candidate. The implications of gender gaps also reach far beyond political science. For instance, some scholars argue that fostering human freedom and individual autonomy is the best way to alleviate poverty, ensure social justice, and cultivate human development (Nussbaum 1995; 1999; 2003; Sen 1990a; 1990b; 1993; 1997; 1999a; 1999b). This human capabilities approach advances human freedom, understood in terms of opportunities available to individuals rather than traditional economic indicators, as a way to evaluate a state's developmental status (Sen, 1990a; 1990b; 1993; 1997; 1999a; 1999b). Other scholars argue that governments must go beyond merely creating the conditions sufficient for an individual to choose to exercise certain capabilities (Phillips, 2001). Distributional differences in economic, social, and political resources limit the ability of some individuals to fully develop these capabilities. Scholars like Anne Phillips argue that this fundamental problem becomes obscured by an emphasis on choice: those concerned with advancing human development must consider not only capabilities, but also the structural inequalities that place some at a political disadvantage in developing or exercising them. This critique focuses scholarly attention on what preconditions must be considered essential for individuals to make a “meaningful” political choice. The practical and normative importance of how choices, or political preferences, emerge suggests the centrality of gender gaps for those studying politics and those trying to advance human development. Despite this importance, much remains unknown about how these gender gaps have changed over time and what factors influence their emergence. This dissertation begins to address these questions. I begin by studying how gender gaps have changed over the last sixty years and by examining how these gaps vary across different demographic groups who possess different levels of politically relevant resources. In Chapter Two, I use American National Election Studies (NES) data from 1948 to 2008 to determine how gender gaps in policy preferences have changed as women have made significant economic, political, and social gains. My results indicate that gender gaps have been decreasing in the areas of civil rights and government function, while holding steady or increasing slightly on preferences related to morality, social policy, and foreign policy. The next section of Chapter Two uses data from the 2000 and 2004 National Annenberg Election Surveys (NAES) to determine the size of gender gaps across demographic groups. Empirical political science research suggests that low levels of political knowledge may impede the functioning of one capability, an individual’s ability to exercise control over his or her environment (Nussbaum, 2000). I examine whether four demographic characteristics correlated with levels of political knowledge alter the size of gender gaps among citizens: gender, race, income, and education. I find that gender gaps in these years vary based on characteristics that may place some at a political advantage relative to others. The largest gender gaps tend to emerge among citizens of higher socioeconomic status, a surprising finding since research suggests that these individuals are best able to form political preferences that reflect their underlying political predispositions. These results provide empirical evidence of how gender gaps vary with different distributions of politically relevant resources, a central concern for critics of the human capabilities approach. After describing gender gaps in multiple issue areas over a sixty year period and gender gaps within demographic subgroups, Chapter Three begins to explore the various biological, sociological, and political frameworks scholars use to explain gender gaps. Chapter Three also draws testable hypotheses from each framework. By drawing together these disparate theories and areas of research, this chapter facilitates more complex theoretical tests of women’s preference formation than political scientists have previously articulated. I conduct such tests in Chapters Four and Five. Chapters Four and Five explore the nature and causes of gender gaps in two issues areas where gender gaps appear most prevalent and stable: foreign policy and social policy. Using data from the 2000, 2004, and 2008 presidential elections, I demonstrate that the opinions expressed by women during the three presidential election years follow the expectations that emerge from the literature on gender gaps. Women surveyed across these presidential elections express less support for military interventions than men. Women also express more support for social welfare programs than men. One potential cause of these gaps may be different attitude structures used by men and women to form political preferences. I analyze attitude structures, providing evidence men and women use similar considerations when forming policy preferences. These chapters then go on to use data simulations to estimate how much gender differences in the factors identified in Chapter Three affect the size of gender gaps (Althaus, 1998; 2003; Bartels, 1996; Delli Carpini and Keeter, 1996). This analysis suggests that the uneven distribution of these factors between men and women contributes to the development of gender gaps. Yet different factors emerge as more important contributors to the development of gender gaps in different issue areas. Biological considerations and political knowledge appear to contribute to gender gaps in foreign policy, while feminist consciousness and gender role socialization appear to contribute more to gender gaps in social policy. The concluding chapter of this dissertation outlines the findings of the three empirical chapters. It also returns to a discussion of the practical and normative significance of these findings. The empirical analyses in this dissertation trace gender gaps across time and issue domains, identify the factors that may underlie them, and estimate how these factors alter political preferences and, consequently, the size of gender gaps. This approach provides political scientists with a better understanding of the nature and origins of gender gaps. Studying these different factors together begins to move political scientists away from a fragmented study of women’s behavior in separate topical domains toward broader theoretical tests of gender gaps in preferences, tests that recognize the complex interplay between gender and political behavior. This approach also ties questions about gender gaps to those in fields such as evolutionary biology and economics, which ask important questions about how material and social differences condition women’s political behavior. A richer theoretical and empirical study of gender gaps can improve our understanding of women’s political behavior, and remains a fascinating area of political science that warrants further study