Jump to ContentJump to Main Navigation
Masculinity, Femininity, and American Political Behavior$

Monika L McDermott

Print publication date: 2016

Print ISBN-13: 9780190462802

Published to Oxford Scholarship Online: August 2016

DOI: 10.1093/acprof:oso/9780190462802.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy).date: 17 October 2018

Gendered Personalities and Political Behavior

Gendered Personalities and Political Behavior

(p.1) Chapter 1 Gendered Personalities and Political Behavior
Masculinity, Femininity, and American Political Behavior

Monika L. McDermott

Oxford University Press

Abstract and Keywords

The book’s introduction provides the theoretical basis for the examination of gendered personalities—masculine and feminine traits—and political attitudes and behavior. It takes the reader through the psychological and political science literature related to this topic, providing a foundation for the book’s overall theory that gendered personalities affect individuals’ politics in multiple ways. Masculine personalities are aggressive, individualistic, and dominant, while feminine personalities are warm, compassionate, and gentle. These personalities translate directly into US politics. Among other things, our political parties are gendered in both their issues and images—Democrats are the feminine party and Republicans are the masculine one. This book’s intuitive argument is that individuals’ masculine and feminine personality traits should affect their political attitudes and party preferences similarly to the gendered nature of our political world—for example, masculinity should contribute to a preference for the Republican Party while femininity should increase Democratic affiliation.

Keywords:   gender, political behavior, personality, masculinity, femininity, gendered personalities, political attitudes

Man up, Harry Reid.

—Sharron Angle (quoted in Hennessy 2010)

Politics is a substantially gendered institution in America. Masculinity and femininity affect and define multiple areas of our politics, including, but certainly not limited to, electoral contests, governing, policies and issues, and the images and positions of the two major political parties. Our culture delineates masculinity and femininity by associating the former with toughness and competitiveness, and the latter with softness and compassion, in both social and political worlds.

The quote in the opening epigraph and others like it demonstrate just how politically prevalent these concepts are. During a public debate between candidates for the United States Senate seat in Nevada in 2008, Sharron Angle told Senate Majority Leader Harry Reid to “man up,” as quoted above. Angle was criticizing Reid’s refusal to debate the solvency of Social Security. The quip was meant to call Reid to task for his unwillingness to discuss this very tough issue (commonly referred to as the “third rail” in politics), which demonstrated, in Angle’s opinion, a lack of masculinity. Angle’s gendered attack got her press attention, but it was far from original in its theme. Questioning the masculinity of political candidates—“wimp baiting,” as one journalist put it (Feinsilber 1989)—is nothing new in American electoral politics. We generally expect our political candidates to display resolution and strength, making the accusation of weakness a traditional favorite and one that can really stick. Perhaps no one knows (p.2) this better than former president George H. W. Bush. In 1987 Newsweek ran a cover story about the then vice president and presidential candidate with the screaming headline: “George Bush: Fighting the ‘Wimp Factor.’” It was not the first time Bush had heard the slur; in fact, the Newsweek story was accompanied by a survey that found a slim majority of Americans believed that Bush’s image as a “wimp” was a problem for his campaign. Despite the wimp factor, Bush won that election, only to lose his bid for reelection in 1992 while again suffering the indignity of the label. Not only did Bush’s opponent, Bill Clinton, publicly accuse him of a “wimp-out” on the issue of crime (The Associated Press 1992), but during the campaign it was also embarrassingly revealed that former president Ronald Reagan had hesitated to pick Bush as his running mate back in 1980 because he thought Bush was a “wimp” (Schwartz and Edsall 1992). While Bush’s presidential advisers fought hard to shake what was essentially a “feminine” image with campaign stunts, such as putting him in boxing gloves for a pre–presidential debate photo opportunity, they largely failed (Ducat 2005). His campaign understood the importance of gendered images for candidates.

Beyond electoral politics, masculinity and femininity also help define the nature of governing. Not only are candidates expected to display strong characteristics, but elected officials need to act on that strength once in office. Politics is a rough world in which conflict reigns. As in the electoral arena, those not up to the task of making difficult decisions in the world of governing can be accused of being soft and lacking in masculinity. Arnold Schwarzenegger, the former Republican governor of California, famously called Democrats in the California state legislature “girlie-men” (over one-quarter of the legislators were actually women) for not passing his belt-tightening budget (Nicholas 2004, B1). The line—which originated from a Saturday Night Live television comedy sketch mocking Schwarzenegger—was such a hit that the California College Republicans sold T-shirts printed with “Don’t be a girlie man: vote Republican.” Another tough-talker, although one of significantly smaller physical stature than Schwarzenegger, 1992 presidential candidate Ross Perot was famous for calling Washington politicians “sissies” for not making the hard choices he thought necessary to improve the country.

Gendered ideals also delineate policy issues in our political world. Political theorist and commentator Irving Kristol assigned gender to policy when he pointed out, “The American welfare state has had a feminine coloration from the very beginning” (Kristol 1996, A16). Kristol was neither the first nor the last to see this connection. In American politics, the goal of helping those in need is seen as a feminine one because it involves (p.3) caretaking and compassion, traits we associate with femininity. For this same reason, issues like health care and education are also seen as feminine. However, masculinity also defines issues in American politics. For example, the arena of national defense is an unmistakably masculine one, centering on keeping our image and stance in the world strong. Even those viewed as foreign policy “doves” see the need to act tough on military issues in America because a “masculine-gendered discourse is the only permissible way of speaking about national security if one is to be taken seriously by the strategic community” (Tickner 2001, 53). This type of masculinity in foreign policy was embodied by President George W. Bush when he landed in a fighter jet on the deck of an aircraft carrier to announce the end of major combat operations in Iraq. Some media and political commentators swooned over the masculinity of the move, comparing Bush to Tom Cruise playing a macho fighter pilot in the movie Top Gun and also to Ronald Reagan, perhaps the ultimate man’s man and foreign policy tough-guy. The aircraft carrier move even inspired an action figure, the “George W. Bush Elite Force Aviator,” wearing a replica of the flight suit Bush wore that day and posed with a helmet tucked under his right arm. The White House was well aware of the domestic benefits of a masculine image in foreign policy and was working to achieve it.

Finally, although certainly not exhaustively, masculinity, femininity, and the policies associated with them play a role in differentiating our two major parties. Americans “associate stereotypically masculine and feminine traits with the Republicans and Democrats, respectively” (Winter 2010, 603). Each party owns gendered issue areas. Republicans own the issues that are viewed as manly and aggressive—such as defense and foreign affairs—while Democrats typically own issues that are feminine and caring, like health care, welfare, and poverty. Liberal commentator Mark Shields put it this way when writing about the 1988 presidential election:

Voters currently trust the Democrats to nourish and nurture the body politic and Republicans to deal with the difficult and hostile forces in the outside world. To oversimplify, the compassionate Democrats are the Feminine Party and the hard-headed Republicans are the Masculine Party. (Shields 1988)

As these various scenarios demonstrate, politicians, pundits, and political scientists alike see the workings of gendered dimensions across a wide range of American politics.

Despite this oft-noted link between masculinity, femininity, and politics, scant research considers whether or not gendered concepts affect an individual’s political behavior. Yet findings from both political science and (p.4) psychology suggest an intuitive, unexplored way in which the gendered dimensions of masculinity and femininity in our society might be expected to affect individuals politically: that is, citizens’ gendered personalities.1 We know from decades of research into personality and politics that an individual’s personality traits affect his or her political predispositions. Personality characteristics stem partly from genetics and partly from socialization, making them early and influential forces in an individual’s life. Extensive psychological research also shows that masculinity and femininity are distinct personality trait dimensions that influence individuals’ social attitudes and behavior. Terman and Miles argued in a formative study that masculinity and femininity “are so deep seated and pervasive as to lend distinctive character to the entire personality” (1936, 1). As such elemental personality forces, gendered personalities are a potentially vital factor in understanding political behavior.

One likely reason for a dearth of inquiry into the influence of individuals’ gendered personalities on their political attitudes and behavior stems from the conflation of gender and biological sex in most political behavior research. Before proceeding I must make clear the distinction between the terms “sex” and “gender” and how I use them in this work. I define each term in a standard gender-research sense: “sex” is the biological difference between men and women; “gender” is the construct society has built over time to reflect behaviors and beliefs thought to be typical of, though by no means unique to, the sexes and their roles in society (e.g., Deaux 1985).

Gendered personalities specifically are the sets of personality traits that originated with societal sex roles. Femininity and masculinity comprise two separate personality dimensions, originally based on society’s expected social role for each sex and the personality profile that facilitated filling that role. The masculine dimension encompasses traits that were once associated with the male role of family provider. Masculine individuals are those who are independent, aggressive, competitive, and willing to take risks, among other traits. Femininity, in contrast, is made up of the personality traits expected of the traditional role of mother and caretaker. Individuals with feminine personalities are tender, affectionate, and sympathetic. Each of these personality dimensions is a cohesive compilation of traits reflective of traditional role expectations, but at the same time they are not mutually exclusive within any single individual.

While political science research has long equated sex and gendered personalities—categorizing only men as masculine and only women as feminine—an extensive psychological literature, as well as the original research presented in this book, abounds with studies finding that biological sex and gendered personalities should be considered separate forces (p.5) in shaping individuals’ attitudes. Changes in societal roles now allow, and perhaps even require, men and women to possess traits once thought only appropriate to the opposite sex. I discuss these concepts and their unique features in more detail later in this chapter, and I demonstrate empirically the extent to which gendered personalities do not conform precisely to biological sex in a traditional feminine-woman and masculine-man way.

In this book I address the gap in research regarding masculinity, femininity, and American political behavior. I do so from the point of view that gendered personalities operate similarly to more general personality traits (namely the “Big Five”) in determining political attitudes and choices. Given the relevance of masculine and feminine dimensions to virtually all aspects of our political world, many potential effects are likely. I specifically hypothesize that individuals’ masculine and feminine personality traits substantially influence, respectively, their partisan affiliations, their vote choices, their ideology, their levels of engagement in political life (including political knowledge and interest), and their attitudes toward the roles of the sexes in social and political life. For example, the delineation of our two parties based on masculine and feminine issues should drive more masculine personalities to the Republican Party (GOP) and more feminine personalities toward the Democratic Party. If these effects do exist—and the evidence in this book will demonstrate that they do—such an examination is crucial to our understanding of personalities and politics, gendered personalities generally, gendered elements in politics, and American political behavior overall.

To reiterate, the book’s overarching argument is that masculinity and femininity, as important personality dimensions acquired very early in life, play a strong role in determining a range of individual citizens’ political attitudes and choices. To test this general theory, each of the book’s data chapters (chapters 3 through 6) analyzes a different aspect of political behavior, encompassing all of those mentioned above, by presenting a specific hypothesis (or hypotheses) and rigorously testing it through empirical analysis. The primary data source for these analyses is an original survey, the Gendered Personalities and Politics Survey (GPPS). This 2011 survey provides the first-ever nationwide measure of gendered personalities, and represents the only known measure of gendered personalities and political attitudes in a single survey in the past three decades in America.2

On topic after topic, the results of these analyses demonstrate the significant role that gendered personalities play in shaping political attitudes and behaviors. I demonstrate that masculinity (as the tough personality dimension) increases Republican Party identification and voting for Republican candidates, while femininity (the compassionate profile) (p.6) does the same for Democratic affiliation and voting. In addition, masculine traits boost political engagement. Individuals who are higher in masculine traits, including competitiveness, dominance, and standing up for one’s beliefs, are more likely to show an interest in and knowledge of the political world. Furthermore, an analysis of individuals’ judgments of the appropriate social and political roles for men and women shows that those whose gendered personalities conform to traditional profiles also believe in a traditional division of labor between the sexes—that men belong more in politics while women belong more in the home.

In the process of analyzing each of these important areas of political opinion and behavior, I also examine the role that gendered personalities play relative to that of the more common biological sex distinction. Depending on the political topic under examination, the relative influences of gendered personalities as opposed to sex vary. Gendered personalities can explain some things that sex itself does not explain well (in political preferences); gendered personalities and sex can both contribute, independently, to politics (in political engagement); and sex and gendered personalities can work in combination with each other in the form of gender conformity (in attitudes toward appropriate sex roles in politics and society). In this context, the results not only highlight the importance of gendered influences, they also help to extend understanding of the effects, and limitations, of both sex and gendered personalities in politics. The remainder of this chapter builds a step-by-step foundation for the rest of this book, reviewing the relevant literature on personality and politics and on gendered personalities in psychological research, and explaining the basis for my general hypothesis.

Personalities and Political Behavior

The only form of study which a political thinker of one or two hundred years ago would now note as missing is any attempt to deal with politics in its relation to the nature of man.

Wallas 1921, 35

Since the early 1900s when Graham Wallas lamented the singular focus on rational and intellectual decision-making in the study of politics, research into natural and impulsive influences on behavior has taken root. Because human beings are not necessarily born as the blank slates they were once believed to be, innate predispositions are receiving much more attention. While the debate between rationality and impulse continues in political science, the study of individuals’ personality traits and their relationship (p.7) to political attitudes and behavior has brought a natural, and empirically demonstrated, relationship to the forefront.

The basic theory regarding personality and politics holds that individual personality helps to produce one’s attitudes and subsequent behaviors because personality comes first, temporally, in human development. Personality is “causally prior” to attitude formation (Gerber et al. 2010, 115). In fact, many scholars now argue that at least a portion of personality is heritable, rather than predominantly environmental as once thought (see, e.g., Bouchard et al. 1990). This view places personality among the core elements of an individual’s psyche. Not only is personality there from the beginning, it is also still there at the end. Personality and the composite of traits that form it are consistent throughout the course of an individual’s lifetime, especially from adulthood forward (McCrae and Costa 2003; Roberts and DelVecchio 2000).

Because personality is such a basic and constant force in a person’s life, it has the ability to strongly influence attitudes and decision making that come later. Personality provides individuals with general predispositions that influence their preferences in terms of both opinions and choices and actions. While no single personality trait or dimension of traits can be said to directly cause a specific behavior, personality provides an underlying tendency for different behaviors in different situations (Ajzen 1987).

Relatively early work in personality and politics examines various individual aspects of personality. This research shows that personality traits such as pessimism and alienation (McClosky 1958) and dogmatism (Alker and Poppen 1973) can lead to a more conservative disposition. At the same time, self-confidence (McClosky 1958), a moral outlook (Alker and Poppen 1973), and impulsivity promote a more politically liberal penchant. And of course, work on authoritarian personality has shown a relatively consistent link between that personality type and ideological conservatism (Adorno, Levinson, and Sanford 1950; and more recently, Hetherington and Weiler 2009). This early research is characterized by a focus on ideology as the general political predisposition of interest. The research is also noteworthy, however, for its rather diverse—some might say unfocused—array of potential personality factors. The lack of a defined personality framework at that time prevented any truly cohesive or thorough examination of personality effects on politics.

The recognition in the 1960s and 1970s of broad factor structures in personalities (see specifically Eysenck 1964), made up of multiple individual and related traits, changed the nature of personality research. The development of the Big Five factor structure in particular broke new, fruitful ground. Researchers during this time period discovered consistent clusters (p.8) of personality traits that formed five broad dimensions of personality: openness, conscientiousness, extraversion, agreeableness, and neuroticism (e.g., McCrae and Costa 1987). The openness factor captures personality traits that contribute to an individual being open to new experiences and having wide-ranging interests. Conscientious individuals are those who follow a regimented existence. The extraverted are, as they sound, outgoing, and they frequently enjoy being the center of attention. Individuals high on the agreeableness personality factor are largely sympathetic to other people and their feelings. Finally, neuroticism measures the extent to which an individual is emotionally unstable, experiencing things like mood swings or excessive worry. The neuroticism factor is also frequently measured in the opposite direction and labeled “emotional stability.”

When researchers discovered and developed these factors, it was believed that the five alone accounted for the vast majority of an individual’s overall personality. As discussed in the gendered personalities section below, however, there are influential factors that lie outside the Big Five, many of which remain unexamined. Nevertheless, following the discovery and validation of the Big Five factors, personality research turned to predominantly studying them. This provided a much more unified framework for the study of personality effects than work that had come before, but also in some ways it was a more limited one.

Recent research into personality and politics analyzes the Big Five framework almost exclusively. Two of the more recent and prominent examples of this research are by Jeffery Mondak (2010) and Alan Gerber and his colleagues (2010, 2011a, among other pieces). Their respective studies demonstrate how the Big Five factors can influence not only underlying predispositions but also specific political attitudes, partisan affiliation, and political behavior. This research finds that conscientiousness and emotional stability are positively related to conservatism, while openness contributes to liberalism (Gerber et al. 2010 and 2011a; Mondak 2010). In sum (and perhaps not surprisingly), individuals who are more regimented in their lives and less emotionally labile have a higher probability of being ideologically conservative; while those who are less so, on each measure, reflecting a more easygoing lifestyle, have higher probabilities of holding a liberal orientation. The extent to which an individual’s interests range across a variety of areas—the openness factor—has the opposite effect promoting liberal rather than conservative orientations. Personality can also influence attitudes in different policy domains: the general ideological effects are reflected in both social and economic policy preferences. In addition, while not influential on overall ideology, extraversion contributes to economic conservatism. Agreeableness promotes (p.9) economic liberalism but also social conservatism (Gerber et al. 2010, 2011a).

In addition to underlying ideological orientations and issue domains, personality also influences our most important electoral variable: party identification. Not surprisingly, the effects are very similar to effects on ideological leanings. People with conscientious natures and those with emotionally stable personalities are significantly more likely to affiliate strongly with Republicans (Gerber et al. 2011a; Mondak 2010). At the same time, those high in openness are significantly less likely to be Republican (and are therefore more likely to be Democratic).

Further research goes beyond attitudes to examine the link between personality and political behavior—namely voter turnout and choice. While the influence of personality does not guarantee specific voting activity or decisions, by influencing predispositions it certainly makes some outcomes more likely than others. For example, personality affects individuals’ levels of political interest, knowledge, and attention, and in turn also the probability they will turn out to vote. Consistent findings across studies are that individuals who are more open to new experiences are also more likely to be politically engaged (interested in and knowledgeable about politics) (Gerber et al. 2011b; Mondak 2010). More agreeable individuals, on the other hand, are less likely to be politically interested and informed. When it comes to participating in elections, researchers find that the conscientious are less likely to partake in politics not only in terms of voting (Gerber et al. 2011b) but also when it comes to other civic actions, such as attending public meetings or participating in political campaigns (Gerber et al. 2011b; Mondak 2010). Individuals with high levels of openness are more likely to participate in a variety of ways (Gerber et al. 2011b; Mondak 2010). In sum, individuals with broader interests are more likely to vote, while those who follow a more rigid way of life are less likely to do so.3

An additional behavioral aspect—for whom individuals vote—has received less attention in personality and politics research, especially in the United States. There are currently two notable exceptions to this scarcity, both based on Big Five personality studies conducted with very large convenience samples of individuals visiting personality testing websites. In an individual-level study of personality effects in the 2004 presidential election, researchers find significant vote effects for all five personality factors (although the sample size of 5,623 may be responsible for some of the statistical confidence) (Barbaranelli et al. 2007). The results demonstrate that agreeableness and openness both had positive effects on voting for John Kerry, while conscientiousness, energy (the authors’ form of extraversion), and emotional stability all had negative effects. Similarly, in a state-level (p.10) analysis using aggregated personality data, Rentfrow and his colleagues (2009) find that openness contributed to Democratic voting in the 1996, 2000, and 2004 presidential elections, while conscientiousness boosted votes for Republican candidates in these three contests. In the aggregate study, however, extraversion was a consistent boost for Democrats, contrary to the individual-level analysis.4

Research into the Big Five personality factors in politics, despite its relatively recent development, has taught political scientists a good deal about basic human nature and political attitudes and behavior. We now know that ideology, attitudes on economic and social factors, political interest, political activity, and vote choice are all influenced by an individual’s broad personality trait dimensions. What we have yet to fully explore, however, is whether there are personality dimensions that lie outside of the Big Five that also influence politics, despite a call for such research by Big Five authors themselves (Gerber et al. 2011a; Mondak 2010). Of specific interest to this book, of course, psychology researchers have found that the personality dimensions of masculinity and femininity are largely independent of the Big Five factors, explaining aspects of personality that the broad structure does not capture. Additionally, these factors have significant and substantial effects on social attitudes and behavior. Both of these elements make masculinity and femininity promising avenues of political behavior research. In fact, as I demonstrate in this book’s conclusion—chapter 7—gendered personality factors have similarly sized effects on political attitudes and behavior as do the Big Five. In support of this examination, the existing literature demonstrates that masculinity and femininity are well defined and consistent personality dimensions, and as such they impact individual attitudes and behavior in a seemingly endless number of ways.

Gendered Personalities

In my research I treat sex and gender as distinct elements, as does much of the psychological gender and sex research over the past three decades (beginning largely with Unger 1979). As mentioned in the introduction to this chapter, I use the term “sex” to refer to biological differences between men and women. In contrast, “gender” describes the social categories—masculine and feminine—that our culture has constructed to contain the norms once associated with each biological sex. In this framework, an individual’s gender is not restricted to his or her sex at birth (as demonstrated empirically below), leaving gender as a separate influence in our lives.

(p.11) Defining Masculinity and Femininity

One central aspect of gender is personality, specifically masculine and feminine personality traits. Researchers have been studying the idea of gendered personalities for a full century. The modern conception of these personalities is of two distinct dimensions: one reflecting the relative presence or absence of masculine traits, and the other reflecting the relative presence or absence of feminine traits. This particular formulation of gendered personalities has been around, and demonstrated empirically, for decades, much as have the Big Five personality dimensions. As a result, measures developed in the 1970s to capture individuals’ levels of masculinity and femininity—the Bem Sex Role Inventory (BSRI) (Bem 1974) and the Personal Attributes Questionnaire (PAQ) (Spence and Helmreich 1978)—still dominate the field of research today. (The method chapter of this book, chapter 2, includes a brief history on the development of the concepts of masculinity and femininity in psychological research, their stability over time, and how I measure them in this book.) Each of these measures is based on empirically defined ideal characteristics for masculinity and femininity—traits that members of our society traditionally have wanted men and women, respectively, to hold—to which individuals can then be compared to quantify their levels of each broader factor.

While some people validly question the persistence of traditional ideal types in our ever-changing world, existing research demonstrates surprising stability for these ideal types, with some studies even finding intensification of traditional ideals in our society. Demonstrating ideal type stability, Holt and Ellis (1998) retested Sandra Bem’s original measurement (1974) of the ideal traits for men and women more than 20 years after her seminal study (also see Auster and Ohm 2000). Using the same format as Bem, Holt and Ellis found that all but two of Bem’s original forty BSRI traits still demonstrate significant differences between the traits desired for men and those desired for women. Holt and Ellis did find that some of the differences in the traits preferred for one sex as opposed to the other were smaller than Bem’s original measurement, but all differences remained statistically significant, indicating little change overall. Evidence for the intensification of sex-based trait ideals—masculinity and femininity—comes from, among other studies, a review of thirty analyses of the ideal personality traits of males and females (Lueptow, Garovich-Szabo, and Lueptow 2001). The authors of this study find that, despite social changes in men’s and women’s roles, there have been no corresponding changes in sex ideals.5 As they conclude: “the various aspects of gender differentiation are not disappearing, if anything there is an increase in sex-typing” (16). Similarly, in (p.12) an individual-level study, Prentice and Carranza (2002) find evidence for a strengthening of the desirability of traditional personality expectations—ideal types—among both men and women. These studies all support the conclusion that the personality profiles of the ideal man and the ideal woman remain largely unchanged over time, despite changes in men’s and women’s actual roles. While the bulk of this research was conducted around the turn of the century, and not more recently as would be preferable, the more than twenty years of stability in these ideal types, specifically spanning the period of the women’s movement, provides strong evidence for their current relevance.

A chief reason for the stability in the conception of gendered personality traits over time is that they originally developed in relation to the traditional roles played by men and women in our society (Eagly 1987). Each sex’s historic—and to some extent biological—role dictated a set of traits believed necessary for an individual to play that role effectively. As Spence and Helmreich (1978) explain:

[M]‌en and women are typically assumed to possess different temperamental characteristics and abilities—distinctive sets of attributes whose … inculcation is believed to be necessary if members of each sex are to fulfill their assigned functions. (4)

Because these gendered trait dimensions stem from societal sex-roles, each is defined by traditional social expectations of that role and the ideal set of traits that should accompany it.

Research into conceptions of model masculinity repeatedly demonstrates that two of the most important elements in this dimension are orientations toward achievement and dominance (Cicone and Ruble 1978). The male sex role includes the expectation that men will be the family breadwinners and the societal leaders, and those filling such roles need to have psychological traits that facilitate the role’s accompanying tasks. Masculinity is thereby defined and measured by an individual’s levels of assertiveness, leadership, and resoluteness, among other traits. This personality dimension may also be described as “instrumental.”6

Society’s beliefs about the ideal female sex role also have common underlying elements throughout psychological research over time. Because the key components of the traditional female role are motherhood and nurturing, the common elements in female sex role expectations are warmth and expressiveness (Broverman et al. 1972). Women are expected to be tender and loving, as well as sympathetic and soothing. These traits are seen as requisite to fulfilling the traditional female sex role and they thereby define (p.13) and measure femininity as a personality dimension. Feminine traits are also occasionally described as “expressive” traits.

While concepts (as ideals) of masculinity and femininity have been largely stable over the past century, society’s conceptions of the appropriate sex roles for men and women have changed a great deal. It is increasingly acceptable for women to have careers, and even be the family breadwinners, and for men to stay home with the children (Bolzendahl and Myers 2004). Views about the roles that are socially suitable for men and women have not only gotten more progressive, they have become more uniformly so across the population, as measured through shrinking standard deviations on sex role measures by Bolzendahl and Meyers (2004).

The American National Election Studies (ANES) provide similar direct evidence of these changes in the conceptions of sex roles. Figure 1.1 contains the average scores of ANES respondents on a measure of the appropriate role for women over time. The question asks whether women should have an equal role, or whether their place is in the home, on a 1- to 7-point scale. Higher values indicate a more traditional role attitude. As the data show, traditional role attitudes—at least toward women—have dropped substantially (and significantly in an independent sample means test), moving from a populace almost evenly divided in opinions in 1972 (3.5 average) to one clearly in favor of an equal role for women (1.8 average) (p.14) in 2008. In further support of the Bolzendahl and Myers findings, the standard deviation of traditional sex role opinions in the ANES has decreased over time, from 2.3 in 1972 to 1.4 in 2008. The sum of these findings demonstrate that as a society we are moving toward broad agreement that it is no longer appropriate to require or even expect conformity to traditional sex roles (although chapter 5 demonstrates how and why some people still support conformity).

Gendered Personalities and Political Behavior

Figure 1.1: Attitudes toward a Traditional Sex Role for Women, ANES Cumulative File.

Note: Dashed lines indicate ± one standard deviation.

Changes in the social acceptability of cross-sex roles for men and women have been accompanied by demonstrable changes in the actual distribution of gendered personality traits within the American public (Twenge 1997a). Given that some substantial portion of personality is heritable (as much as 48% for gendered personalities, according to Mitchell, Baker, and Jacklin 1989), and therefore should not change easily or quickly in society, such potential changes beg the question of how and why this is so. There are two possible mechanisms through which this change could be happening. First is the change in the social environment. Changes in attitudes toward the acceptability of nontraditional gendered personality profiles should naturally result in changes in the environmental factors that influence personality development—namely socialization. For example, empirical evidence shows that families with more sex-egalitarian views are significantly more likely to raise children who are less oriented toward sex-typing (Weisner and Wilson-Mitchell 1990). Because society as a whole has become less restrictive over time regarding sex roles, strictly sex-typed personalities should be less common in general. Indeed, research on twin pairs indicates that broad social forces may be responsible for a substantial part of the nonhereditary aspects of gendered personalities (Mitchell, Baker, and Jacklin 1989; and, more recently, Knafo, Iervolino, and Plomin 2005).

The second potential reason for changes in the distribution of gendered personality traits over time is the weakening of social desirability attachments to femininity for women and masculinity for men. This increase in acceptance of nontraditional profiles may make individuals with nontraditional personality profiles more honest in their self-assessments, resulting in higher observed levels of masculinity for women and femininity for men. After all, society’s ideas about what is appropriate and individuals’ resulting desires for conformity can affect answers to self-assessment questions of all kinds (Paulhus 1991). But as these pressures ease, some individuals may now be more comfortable identifying themselves as a blend of the two personality dimensions (Auster and Ohm 2000). In other words, it is increasingly acceptable to individuals to identify themselves, to themselves or to others, as diverging from the typically sex-defined personality (p.15) boundaries. If the result is more honest self-assessment, then we would expect to measure less conformity over time.7

Regardless of the mechanism at work, there is little doubt that gendered personalities are no longer as sex-conforming as they once were. It has become increasingly likely for individuals to take on (or admit to) nontraditional gendered personality traits (Twenge 1997a). To be clear, masculinity and femininity as personality dimensions still capture meaningful aspects of personality and are important predictors of attitudes and behavior, as discussed later in this chapter. Possession of each set of traits, however, is no longer as highly correlated with one’s biological sex as it may have once been.

The result of these societal developments is change in aggregate levels of masculinity and femininity within the population. For example, conformity of men to a masculine personality profile and of women to a feminine one has dramatically decreased over the past few decades. In a meta-analysis of twenty years of BSRI studies, Twenge (1997a) shows that levels of masculine personality traits have risen significantly among women (also see Spence and Buckner 2000). At the same time, men’s possession of both masculine and feminine personality traits has increased. Both sexes are now, on average, more likely than in the past to possess the personality traits of the sex role opposite the one once dictated to them by tradition.

The observation that gendered reality has been moving away from sex conformity since the 1970s is not restricted to academic studies; it has pervaded concurrent American culture, both political and popular, as well. Since at least the early 1970s, cultural conservatives have loudly bemoaned the gendered changes—a loss of sex conformity—in American society, while social progressives have repeatedly promoted them (e.g., Schlafly 1972). These changes were also debated, and noted, in popular culture. One of the more obvious examples of this blurring of gender lines and the debate surrounding it is the ever-popular 1970s television sitcom All in the Family. In the introductory song to the series, family patriarch Archie Bunker pines for the days when “girls were girls and men were men,” and struggles in each episode to accept the feminine, caring side of his hippie son-in-law. On the surface Archie advocated for the good old days of gender conformity, often to great comedic effect. But at the same time, Bunker’s blatant bigotry made his overt message unappealing, revealing the gendered tensions of the time. On a multimedia front, in 1972 television star Marlo Thomas released a Grammy-nominated album, as well as an accompanying children’s book and later a television special, called Free to Be … You and Me. These materials explicitly promoted the crossing of gendered personality lines with songs like “William’s Doll.” Thomas reports the project was inspired by a shopping (p.16) trip she took to buy a book for her niece, when she discovered “that all the children’s books I found reinforced old gender stereotypes of what girls and boys were supposed to be or ought to be. None of them talked about all the possibilities of what girls and boys could be.”8 It is important to note that during what was the era of feminism, the focus was not merely on females becoming more masculine; these examples also demonstrate an increasing acceptance of males becoming more feminine. The reality of an American society in which gendered personalities were no longer restricted by sex, whether culturally celebrated or condemned, was evident.

The current result of these changes is an American society with little sex and gendered personality conformity. To demonstrate the extent to which gendered personalities are currently detached from sex, Table 1.1 reports the recent percentages of respondents falling into each of four different gendered personality categories—based on a standard BSRI sex-type measure9—from two different sources. The first source is the GPPS employed throughout this book (and explained in detail in chapter 2). The second dataset comes from an independent online personality testing website, personality-testing.info. This site conducts online personality tests and collects, compiles, and releases the data for public use.10 These data are from adults who visited the website (a nonrandom, self-selected sample) and took the BSRI between 2011 and 2012. Using one data presentation method common to BSRI analysis, gender conformity (see n. 9), I classify male and female survey respondents into one of four categories, according to their combination of masculine and feminine traits: “traditional” gendered personalities—individuals who are above the median measure for their own sex’s traditional traits while below the median on the opposite (p.17) sex’s; “cross-typed” personalities—those high on the opposite sex’s personality dimension while low on their own; “androgynous” personality profiles—individuals above the median on both masculinity and femininity, regardless of sex; and “undifferentiated”—men and women below the median on both measures (Bem 1977).

Table 1.1. Traditional and Nontraditional Gendered Personalities among Men and Women, from the BSRI Short-Form

GPPS 2011

Personality-testing.info 2011–2012

Gender type



































As is clear from the table, few American adults from either data source conform to a traditional gender/sex personality profile. In the nationally representative GPPS, only one-third each of men and women possess predominantly the traits that society has traditionally associated with their sex. This leaves 67 percent each of men and women that possess non-sex-conforming gendered personality profiles—presenting a far more unsexed gender personality profile in America than a conforming one. In the self-selected personality-testing.info sample, conformity is even lower, with roughly one-quarter of both men and women falling into the traditional gendered personality classification. In line with Twenge’s (1997a) findings that members of each sex increasingly possess the traits traditionally expected from the opposite sex, androgynous profiles are nearly as common among both sexes as traditional, conforming profiles are. More to the point, in both datasets roughly half of each sex possesses high levels of the traits that society once expected exclusively from the opposite sex.

These results depict a society in which masculine and feminine personality traits are not the sole domains of men and women, respectively; indeed they are far from it. These personality dimensions are gendered (following classic ideals of masculine and feminine), but not sexed (restricted by biological sex). As such, they have the potential to influence attitudes and behavior apart from any potential effects of biological sex. All of the findings discussed and demonstrated in this section leave modern gender research with two clearly and consistently defined (and measured) personality dimensions that can play determinative roles in shaping attitudes and behavior. The resulting psychological literature on gendered personalities demonstrates just how important these influences can be in shaping our personal and social lives.

Established Effects of Masculinity and Femininity

As discussed above, from the early days in personality research scholars have argued that personalities provide us with stable attitudinal and behavioral predispositions; and that these predispositions help us to form relatively consistent decisions in diverse contexts (e.g., Allport 1955). Our personality traits, and more importantly the cohesive dimensions they (p.18) form, are the guiding forces in our lives. While situational factors influence decisions as well, the underlying tendencies provided by personality are relatively constant across situations (Mischel 2004) and over one’s lifetime (McCrae and Costa 2003). The gendered personality dimensions of masculinity and femininity exhibit precisely these features, providing individuals with gendered predispositions brought to bear in attitude formation, decision making, and behavior, as evidenced by extensive research.

The masculine and feminine personality dimensions play a unique role in shaping an individual’s character and thereby his or her attitudes and behavior. Researchers find that gendered personality dimensions lie outside of other, supposedly comprehensive, personality dimensions, including the ubiquitous Big Five. Factor analysis demonstrates that gendered personality factors load onto factors that are distinct from those of the Big Five (e.g., Noller, Law, and Comrey 1987; Paunonen 1993). Gendered dimensions of personality are not only separate from such broad measures, but there is also evidence that they are equally as influential in defining an individual’s overall personality (Paunonen and Jackson 2000). In direct support of this, chapter 7 presents an analysis of the effects of gendered personalities on political attitudes relative to those of the Big Five (as found in Mondak 2010 and Gerber et al. 2011a and 2012). The analysis demonstrates that gendered personalities have similar effect sizes to the five broad factors typically studied in political research. As Lee, Ogunfowora, and Ashton (2005) note, in a study that finds that masculinity-femininity (measured on a continuum) does not fit well within the Big Five, nonbroad factor personality measures such as gender can be “useful in predicting important real-life criteria, by assessing variation that is not shared with the broad factors of personality” (1455).

These “real-life criteria” cover a great deal of our lives and span numerous cultures, according to existing research. A wide array of studies demonstrates gendered personality effects in varied attitudinal and behavioral areas, including values formation, self-esteem, relationships, psychological and physical well-being, and many others. Below is a sampling of only the most recent areas in which researchers have found masculine and feminine personality effects—specifically measured by the BSRI—on attitudes, behavior, and even health outcomes.

  • Binge drinking: Masculinity contributes to binge drinking among both male and female college students (Peralta et al. 2010).

  • Mental health: High levels of both masculinity and femininity—androgyny—contribute to lower levels of depression and better coping strategies in stressful situations (Cheng 2005).

  • (p.19) Intelligence ratings: Masculine individuals view themselves as more intelligent and verbally skilled than average, while those high in femininity see themselves as more socially and emotionally skilled (Szymanowicz and Furnham 2013).

  • Sex and romance: Individuals with higher levels of feminine traits prefer romantic situations over sexual ones, both implicitly and explicitly (Thompson and O’Sullivan 2012).

  • Workplace leadership style and success: While masculine traits are typically associated with leadership, feminine traits among managers foster communal methods of leadership and contribute to more satisfied employees (Gartzia and van Engen 2012).

  • Health and wellness: Men with higher levels of feminine personality traits have a lower risk of death from coronary heart disease (Hunt et al. 2007).

Masculinity and femininity not only explain a wide array of social and psychological attitudes and behaviors, they also have relevance to a diverse assortment of cultures. Researchers in many other parts of the world have made use of these concepts, specifically as measured by either the BSRI or the PAQ, to examine a host of topics. A sample of such studies includes: a study in Israel of the effects of gendered traits on athletic involvement (Rubinstein and Lansisky 2013); a Nigerian analysis by Ogunleye (2012) testing the effects of gendered personalities on sexist attitudes; Zimmerman, Sieverding, and Müller’s (2011) comparison of the effects of masculinity and femininity on alcohol use in Spain and Germany; research into the self-esteem of university students in China based on their gendered traits (Huang et al. 2012); and, finally, a personal favorite (arguably of less social import than the others cited), a Brazilian study of the degree to which gendered personalities affect an individual’s appreciation of domestic cats as pets (Ramos et al, 2012).

Naturally, studies that analyze the effects of gendered personalities do so while controlling for potential biological sex effects, but some research goes still further by explicitly comparing the relative effects of sex as opposed to gender. Scholars occasionally find that some individual differences we have historically attributed to biological sex may actually be better explained by masculinity and femininity. For example, studies of educational motivation and achievement gaps between the sexes (McGeown et al. 2012; Pajares and Valiante 2000), religiosity differences between women and men (Francis and Wilcox 1998; Thompson 1991), and the higher ethical standards usually found among women relative to men (McCabe, Ingram, and Dato-on 2006) all find that, both logically and empirically, gendered personalities (p.20) better account for individual differences than does sex. In fact, in each of these studies the authors find that sex effects disappear once they include gendered effects in their models. In some instances, although certainly not all, gendered personalities explain attitudes and behaviors better than biological sex does.

While this short review cannot begin to provide a comprehensive accounting of the multitude and variety of gendered personality studies conducted both in the United States and abroad, it should present a sense of both the topical and cultural range of masculine and feminine personality influences. Masculinity and femininity are widely accepted as powerful, independent forces in shaping individual character and action across a host of attitudes and behavior, with one glaring exception: there have been minimal tests of their effects in the realm of politics. The prevalence of gendered distinctions in existing political science research, however, leads to a reasonable expectation that these gendered traits would affect political attitudes and behavior.

Gendered Politics

The extent to which US politics is infused with gendered concepts strongly suggests the merit of pursuing my investigation into an as yet unexplored facet of those politics: individual gendered personalities in the public at large. As stated in the introduction, masculinity and femininity are common forces in our political world. Existing studies analyzing gendered effects in politics, however, have done so primarily from a top-down perspective: analyzing gendered images of candidates and political parties and how those images affect judgments by individual citizens. While these studies do not address individuals’ gendered personalities, they do demonstrate the prevalence of gendered personality concepts in politics. In addition, four noteworthy studies take a look at individuals’ gendered personalities and political preferences, although unfortunately only from a gender-conformity angle. I discuss each of these literatures—candidate gender stereotypes, gendered images of political institutions, and individuals’ gender conformity and politics—in turn.

Gendered Candidates

The most prolific research area in gendered politics is that of gendered images of candidates. The most common approach within this area (p.21) examines candidates in terms of sex-stereotyping and the gendered traits voters assign to male and female candidates. As a result, much of the research deals with gendered ideals (as stereotypes), and focuses on the interaction between candidate gender and candidate sex—whether women candidates are stereotyped as more feminine and less masculine than men candidates because of their sex, and whether men are assumed to be more masculine and less feminine.

Research demonstrates that voters do stereotype women candidates as more feminine than men not only on personality traits but also based on issue competence stemming from traditional sex roles. Sapiro’s (1981–1982) innovative study of the stereotypes voters have of men and women in politics demonstrates that society’s gendered expectations influence experimental subjects’ views of a candidate’s leadership potential based on sex. A hypothetical politician is rated more competent on “feminine” compassion issues like health care and education when presented as a woman than when presented as a man. Research since Sapiro’s has consistently shown these same effects: voters view women candidates and politicians as more feminine on both gendered issues and gendered traits than they view male candidates; at the same time, voters view political males as possessing more masculine traits and issue competencies than females do (Alexander and Andersen 1993; Huddy and Terkildsen 1993a; McDermott 1998; and Sanbonmatsu 2002, to name just a few).

More recent research on candidate stereotypes reinforces the early work that gender stereotypes affect judgments of candidates, although it also displays some important and perhaps recently developed limits of stereotyping and its effects. For example, research finds that women candidates are still stereotyped in terms of gendered traits, but not uniformly (Schneider and Bos 2013). Women politicians remain disadvantaged—rated lower on politically desirable masculine traits—relative to male politicians. At the same time, however, female politicians are not as stereotypically judged on feminine traits—some of which can be undesirable in politics—as women in general are. By virtue of being a politician, a women candidate may partially overcome the feminine stereotype, at least compared to women not in the political arena.

Research also demonstrates some weakness in the effect of gendered stereotypes relative to political party identification. Dolan (2010) finds that potential voters do still stereotype hypothetical male and female candidates as holding masculine and feminine traits and issue competencies, respectively, but that when candidate party identification is included in the mix, stereotype results weaken considerably. When it comes to voting in actual races, she finds that the primary influence of gendered stereotyping (p.22) is through issue competence expectations, with little effect related to trait inferences. Multiple studies have noted this limitation of gendered stereotypes vis-à-vis partisan candidate stereotypes (assumptions about Republican candidates and Democratic candidates) (for recent tests see Dolan 2014 and Hayes 2011), bringing into question the electoral impact of sex stereotypes, if not their existence.

The measured weaknesses in gender-stereotype effects should come as little surprise given the weakening ties between sex and gender, and sex stereotypes that are potentially waning as a result (Diekman and Eagly 2000). But such findings do not necessarily reflect a diminishing role for masculine and feminine influences. There may still be a part for gendered traits to play in candidate stereotyping, it simply may not be a wholly sex-restricted one in which only one sex owns traditional traits. A unique study by Huddy and Terkildsen (1993a) shows that when candidates of either sex are described in gendered terms—compassionate and family-oriented versus tough and ambitious—experimental subjects still assign gendered trait stereotypes, but they assign them in accord with the candidate’s gendered personality profile rather than by the candidate’s sex. As the authors conclude: “The gender-linked adjectives used to describe candidates made a difference politically” (132). Voters make decisions based on gendered trait profiles, even when that gender does not match the traditionally expected biological sex. While this is only one study, the results show that the application of gendered personality concepts in politics is likely to be broader than sex alone.

Gendered Institutions

Further support for the existence of a broader, nonsexed role for gendered ideals in the world of politics stems from research into judgments of political institutions. While institutions are inherently sexless, they may not be genderless. Research shows that Americans’ judgments of our two major political parties are, at least in part, driven by gendered images of them. Voters also translate these gendered images into gendered traits that they assign to candidates based on each candidate’s party identification. And, separate from these party images, Americans also consider different types of political offices to be gendered—with some considered more masculine and others more feminine.

Americans view the Democratic and Republican parties in gendered terms, as summarized by Mark Shields in his quote earlier in the chapter. Research by Winter (2010) establishes empirically what Shields and other (p.23) pundits have discussed for decades: the Democratic Party has a feminine profile while the Republican Party has a masculine one. Winter demonstrates that public images of the parties carry distinct gender-trait dimensions. Voters discuss their likes and dislikes of the parties in gendered terms. For example, voters describe the Democratic Party as compassionate and caring (feminine personality traits) while viewing Republicans as independent and decisive (distinctly masculine personality traits).

This research into gendered parties reinforces earlier work on party issue ownership. Each of the major American parties has a central message on policies and issue stands, and these messages are, whether by design or not, gendered. As a result of the parties’ messages, the public believes each party “owns” certain issues. For example, the Democrats own compassion issues, and the Republicans are the party of military and law-and-order issues (Petrocik 1996). Pioneer of neoconservativism Irving Kristol pointed out as much after the 1996 Democratic presidential convention, arguing that the Democrats had become the feminine party by virtue of their focus on the welfare state (Kristol 1996). Analysis of the Republicans’ 2008 convention shows a similar strongly gendered message for the GOP, but on the masculine side, with an emphasis not only on the rugged individualism and toughness of their candidate John McCain, but also on the party’s policies (Gibson and Heyse 2010). In a twist on the familiar sex and gender script, this 2008 masculine convention message was brought home perhaps most forcefully by the female Republican vice-presidential candidate, Sarah Palin.

Gendered effects may also extend beyond issues and party images through party-based stereotypes of candidates’ traits. Research shows that parties own traits as well as issues, and that when a candidate campaigns on his or her party’s issues, voters then assume the candidate possesses personality traits they associate with the party (Hayes 2005). A Democratic candidate who toes the party line will be viewed as compassionate (feminine), and a Republican who campaigns on GOP-owned issues will be seen as a relatively strong leader (masculine). The feminine image of the Democratic Party evidently transfers to its candidates, as does the masculine image of the GOP. While not explicitly discussing the party images as gendered, this research presents another level in which masculinity and femininity are relevant to politics.

Gendered expectations extend to another sexless institution—political office. Different levels and types of offices, for example, can be classified as either masculine or feminine by both researchers and the public based on the duties of the office (Fox and Oxley 2003; Rosenwasser and Dean 1989). Scholars classify attorneys general and governors as masculine positions (p.24) because they deal with crime and executive leadership, respectively. In contrast, elective offices that deal with education—such as state superintendents of education—are considered feminine. Of course, researchers view the highest office in the land—the presidency—as the epitome of a masculine position because dominance plays such a powerful role in the duties and practice (Duerst-Lahti and Kelly 1995). Importantly, the public agrees with these gendered assessments (Rosenwasser and Dean 1989). The gendered images of these offices are vital to electoral politics not only because of public judgments but also because strategic candidates appear to consider these images when deciding for which offices to run. Male candidates more frequently run for masculine types of office, while women typically choose to vie for feminine offices (Fox and Oxley 2003).

The gendered nature of politics is not limited to images or stereotypes, although these are the dominant focus in the existing political behavior literature. Gendered effects have also been found in a wide variety of political areas, such as political rhetoric and style (Dow and Tonn 1993); candidate campaign strategies (Wadsworth et al. 1987); the formation of attitudes related to politics, like racial diversity and sex equality (Wade and Brittan-Powell 2001); and the actions of our political leaders (Coe et al. 2007). Finally, a very limited literature looks at voters and their gendered personalities, the subject of my investigation.

Gender Conformity and Politics

While research into individuals’ own gendered personalities and their political behavior is largely absent from the academic literature, a few isolated studies do exist and they lend further support to this book’s general argument that individuals’ gendered personalities matter politically. The studies do not provide strong evidence, however, because three of the five date from the 1960s and 1970s, and also because they all focus solely on gender conformity—the extent to which an individual fits the personality profile for their biological sex—rather than analyzing gendered personalities in and of themselves. Nevertheless, these few studies demonstrate that gendered personalities can affect individual attitudes.

The early studies all test the same basic hypothesis, albeit in different ways, that individuals who do not conform to the personality traits associated with their sex are more likely to be ideologically liberal than those who do conform. The earliest of these analyses (Ferdinand 1964) looks at feminine traits among males, and the effects such femininity has on liberalism (technically defined as “humanism”), lack of social conformity (p.25) (“aconventionalism”), and support for government public assistance (“welfarism”). The results demonstrate that men with feminine aspects do indeed have more liberal tendencies on these three factors. A somewhat later study from 1970 finds that effects of nonconformity apply to both sexes and their politics (Cottle, Edwards, and Pleck 1970). Men and women who principally hold the gendered traits of the opposite sex—masculine women and feminine men (with sex role identification defined through drawing tests)—are more likely to agree with politically liberal statements than other individuals. Both studies provide evidence that nonconformity with the gendered personality dictated by one’s sex may lead to more liberal views. Hershey and Sullivan (1977) also analyze gender nonconformity and ideology, although they use the BSRI as their measure of gendered traits, providing a more defined (and currently relevant) view of gendered personality. They test not only the role of gender nonconformity in determining political views, specifically liberal policy positions (use of military troops and governmental guaranteed income), but also views of political figures (specifically Gerald Ford and Bella Abzug) and individuals’ party identification. They find mixed results overall, but they do find evidence, as did the previous two studies, that nonconformity generally relates to ideological liberalism.11

The primary exception to the preceding decades-old studies of gender personality conformity and politics comes from a 2012 article by Hatemi and his colleagues. The authors study how the electoral preferences of Australians (in a study of twins) are affected by individuals’ conformity to the expected personality for their sex. They find that adult gender conformity, measured on separate scales for men and women (unfortunately resulting in noncomparability across the sexes), can influence vote choice. Specifically, less conforming individuals are more likely to support the left-leaning party—results similar to the early studies above. While their research takes a step toward examining gendered personalities, as with the early research they limit the definition of gender by biological sex—women are measured only on femininity and men solely on masculinity—and thereby study only gender/sex personality conformity. Continuing to define gender in relation to sex, however, ignores the empirical reality presented in Table 1.1 that sex does not define gendered trait possession, and that therefore biological sex should not restrict the potential political influence of masculinity and femininity.

The sum total of the research presented in this section demonstrates the inherently gendered nature of politics. Individuals apply gendered stereotypes to candidates, based on both candidate sex and candidate trait descriptions. They also apply gendered issues and traits to sexless institutions in politics, like our two major political parties and different types of (p.26) political offices. Additionally, limited research into voters’ gendered personalities demonstrates that gender conformity or nonconformity has some, albeit limited, effects on individual political preferences. This literature lays important theoretical and empirical groundwork for the extension of gendered personality effects into individual political opinions and behavior—my goal in this book.

The Gendered Personalities and Politics Argument

This book’s central argument is that the two dimensions of gendered personality traits—femininity and masculinity—influence individual political attitudes and behaviors. We know from existing research that, first, personality affects political opinions and actions (largely in terms of the Big Five factors); second, gendered personalities are an independent and strong influence on social actions and beliefs; and third, that politics is an inherently gendered institution. From these three points it seems intuitive that gendered traits should substantively affect individual political preferences and actions. This section presents a step-by-step articulation of this argument, summarizing the evidence amassed to this point through review of the literature. The remainder of the book tests the arguments in important and relevant areas of political attitudes and behavior.

Solid and wide-ranging research supports Terman and Miles’s (1936) early assertion that gendered personalities are central to the character of an individual. Gendered personalities help determine who we are and what we do across a broad range of areas in our daily lives. For example, we know that individuals who are high in feminine traits behave more communally, caring more about, and more for, others. At the same time, strongly masculine personalities believe more in individualism and as a result behave competitively and aggressively. We also know the two gendered personality factors are not mutually exclusive. An individual can have high levels of both masculinity and femininity, and even low levels of both. All combinations are possible (as demonstrated in Table 1.1).

I argue that the gendered personality dimensions of masculinity and femininity affect political attitudes and behavior—much as they do social attitudes and behavior—because they are so important to individual psyches and because evidence shows that our psyches affect our politics. Personality is a strong determinant of one’s views in politics. Additionally, politics is naturally infused with gendered elements, making the connection a natural one, if overlooked. The entirety of the research discussed (p.27) so far makes these points clear, and it is the collective origin of my overall hypothesis.

Various aspects of gendered personalities and of politics dictate the specific hypotheses tested in this book. I have chosen three separate areas of political research on which to test my general argument. The hypotheses and rationale in each area are outlined briefly here, and discussed in detail in the respective chapters for each.

First, gendered personalities should affect political preferences and electoral choice. Given the research discussed in the sections above, there is a clear expectation regarding the relationship between gendered personalities and party identification, ideology, and vote choice. The literature shows that the traits associated with masculinity and femininity map nearly perfectly onto the images of, and issues owned by, the two major parties in America. Because individuals with feminine personality traits are largely compassionate, sensitive, and concerned with others, femininity should lead to Democratic affiliation and voting. At the same time, an individual who possesses masculine traits is aggressive, individualistic, and forceful, predicting a more Republican outlook because the GOP is the party of tough defenses and bootstrap beliefs.

Second, gendered personality traits should also influence individuals’ levels of political engagement. We know that politics is not for the faint of heart—it is by its very nature competitive, aggressive, and tough. It requires individuals who are willing to strongly defend their own beliefs. This is the masculine personality in a nutshell. Given the purely voluntary nature of political action in America, the more masculine an individual is, the more he or she should engage with the political world. (This does not automatically mean, however, that feminine individuals will not engage in politics, since femininity is not the opposite of masculinity.)

Finally, I contend that masculinity and femininity should affect individuals’ opinions on which sex is better suited for politics (i.e., running for office) and which for home life. In this case, the hypothesis for the effects of gendered personalities on sex roles involves individuals’ own sex and gender conformity. Previous research (and possibly common sense) demonstrates that individuals who fit the personality orientation for their sex relatively well also expect others to do the same (e.g., Hoffman and Fidell 1979). Here again this larger social effect should apply to the political realm. Specifically, individuals whose personality profiles conform to that expected of their sex—that is, feminine women and masculine men—should themselves expect sex role conformity in politics and society (men better suited for politics and women for the home). In contrast, (p.28) nonconformists are likely to be more supportive of nontraditional sex roles (men at home and women in politics).

Much of the research discussed to this point in the chapter demonstrates separate roles for sex and gender in determining attitudes and behavior, and even a stronger influence of gender than sex in some instances. As a result, with the exception of the hypothesized combined effect of sex and gender on sex role attitudes, I expect that gendered personalities will play a role distinct from, and even more powerful than, biological sex in influencing political opinions and actions. In this book I systematically test each of these expectations.

Layout of the Book

The remainder of the book is divided into six chapters. Chapter 2 describes the Gendered Personalities and Politics Survey (GPPS), which forms the backbone of the empirical analyses. In addition, I review the methodological issues involved in measuring gendered personality traits. I discuss not only the gendered personality measure I employ here, the BSRI, and its accompanying strengths and weaknesses but also the primary alternative measure, the PAQ. The chapter also includes a discussion of the alternative variable constructions of the BSRI in the existing research; and I explain why I choose the specific formulations I employ in different analyses throughout the book.

Chapter 3 contains an analysis of one of the biggest issues in American politics when it comes to sex and gender: the partisan “gender gap.” I analyze the role of gendered personalities in shaping partisan preferences, in terms of both party identification and vote choice (in 2008 and 2010), as well as political ideology. Analysis of the GPPS shows a substantive and significant effect of gendered personality traits on political and partisan preferences—increased masculinity leads to Republicanism while femininity leads to Democratic leanings. The analysis also looks at the concurrent effects of individuals’ sex on their preferences in order to address the traditional sex gap literature. Significantly, it finds that once gendered personalities are accounted for, sex no longer matters to these preferences (consistent with some of the psychological literature). The result demonstrates a “gendered” gap rather than the traditionally conceived sex gap.

I next investigate individual differences in political interest, activity, and knowledge. Chapter 4 contains an analysis of how masculinity and femininity affect political engagement. Here again, conventional wisdom has defined evident differences between men and women as sex-determined. (p.29) Using the GPPS I show that increasing masculinity leads to increasing interest in and knowledge of politics, demonstrating that politics holds more attraction for those whose personalities are grounded in aggression, independence, and competition. The results of this analysis also show that, in this case, gender and sex operate independently in determining political engagement. Gendered personalities, however, have the far stronger effect.

Chapter 5 takes one of the original Bem classification schemes of masculine, feminine, androgynous, and undifferentiated as categorical distinctions and analyzes the categories’ effects on the same political factors as chapters 3 and 4: partisan preference and political engagement. The benefit of such an analysis is investigating androgyny, a popular concept throughout gendered personality studies, as well as undifferentiation, a category less popular in research than androgyny but nevertheless important. Androgynous individuals are those who possess above the median levels of both masculine and feminine traits; and the undifferentiated are those with lower than the median levels of these trait factors. Scholars have argued, and at times shown, that the two “balanced” gender categories—androgynous and undifferentiated—are unique precisely because of their balanced masculine and feminine trait levels. I analyze this logic empirically and find that the primary difference lies in strength of partisan ties. The androgynous are significantly more partisan than, and the undifferentiated significantly less partisan than, the unbalanced groups of masculine and feminine.

Chapter 6 takes the examination of gendered personalities and politics one step further, analyzing the role these personality factors play in influencing attitudes about sex roles in society and politics. Specifically, I explore the role of gendered personalities in forming individual judgments about women’s and men’s appropriate roles in politics and family life. I demonstrate not only that Americans retain a reluctance to see women in a political role but also that this reluctance is influenced by individual gendered personalities. Specifically, this analysis captures an interactive effect of gendered personalities and sex. The more an individual’s gendered personality conforms to their biological sex (i.e., feminine women and masculine men), the less accepting they are of a nonconventional role for women relative to men. Those who are themselves less conventional in their gendered orientation are more accepting of revised sex roles in politics and the family. Thus existing research on gender conformity receives support in the context of modern American politics when the issue is one of tradition, rather than partisanship or political activity. In this chapter I also analyze androgyny and undifferentiation as categories, compared to the conforming personality category. Perhaps not surprisingly, given existing (p.30) psychological literature on the androgynous, they prove to be the least supportive of traditional sex roles. (I explain fully in chapter 2 the difference in the categorical schemes of gendered personalities, chapter 5’s subject, and gendered conformity, used in Table 1.1 and analyzed in chapter 6.)

Chapter 7 concludes my study by tying together the various findings throughout the book on the role of gendered personalities, including a discussion of the research’s strengths and weaknesses. In this chapter I include an analysis of the empirical effect size of gendered personalities compared to those of the Big Five (as found in Gerber et al. 2011a and 2012 and Mondak 2010). Replicating existing Big Five models with masculinity and femininity in place of the Big Five traits demonstrates that gendered personalities have similar effect sizes to the more standard Big Five. I also reassess the conventional wisdom that defines gender in political science based on biology rather than personality. In light of the evidence presented in the book, the latter may provide more leverage for understanding important political phenomena or, at the very least, add significantly to our understanding of gender writ large. Finally, I discuss possible avenues for future research in which gendered personalities should play a role in explaining and understanding not only individual political behavior and attitudes but also candidates’ gendered images and choices in electoral campaigns and their potential reception.


(1.) To be clear, masculinity and femininity are defined here as personality traits not physical traits.

(2.) Chapter 2 contains full details on the survey and its method.

(3.) Mondak (2010) finds the negative participation effects of conscientiousness counterintuitive. In subsequent analysis of the relationship, he finds that the lowered effects are primarily among those high in conscientiousness but low in external efficacy. In other words, a conscientious person will not be conscientious about electoral participation if they do not view such effort as making a difference.

(4.) This difference may well stem from differences in how the two studies measured their factors. Barbaranelli and colleagues (2007) used a translated version of their Italian five-factor measure, while Rentfrow (2009) used the standard Big Five Inventory (BFI). As a result, the traits presented to respondents were not identical across the two studies.

(5.) While Lueptow, Garovich-Szabo, and Lueptow (2001) did find studies that indicate a diminution in the differences between the ideal types of men and women, these studies were deemed (by the authors’ standards) less reliable than those that found stability or intensification.

(6.) Researchers often generalize the differences between these role attributes according to Bakan’s (1966) classification of personalities as either instrumental or expressive. Instrumentality is action-oriented and independent, not involving the reaction or concerns of other individuals in society. The resulting traits include industrious and forceful. In contrast, expressiveness is primarily concerned with the actions of and interactions with others in the social sphere. As a result, expressiveness includes the traits sympathetic, warm, and understanding (Johnson et al. 1975) Many researchers concur that the fundamental masculine dimension is related to instrumental (also called “agentic”) traits, while the feminine dimension correlates with expressive (or “communal”) traits (Bakan 1966).

(7.) I should note that this idea is merely speculative on my part. There is no evidence that individuals have ever been less than honest in their BSRI self-assessments. This does not mean, however, that one should not allow for the possibility.

(9.) Bem and subsequent researchers use two classification schemes, one that focuses on gender conformity—used here in Table 1.1—and one that focuses on gendered personalities. The analysis in chapter 6 looks at the former, while chapter 5 analyzes the effects of the latter. The difference in meaning and purpose for the two schemes is explained fully in chapter 2.

(10.) The tests all follow the common academic measurements for each personality element (including the Big Five Inventory), and the data have been used in multiple, recent publications (e.g., Mahalakshmi and Sornam 2013; Perwez et al. 2013). The website provides no information on sponsorship.

(11.) The role of nonconformity—in the form of androgyny and undifferentiation—in political views is analyzed in chapter 5.