CHAPTER 2
The Sources of the Gender Gap in Political Participation
For too long, the history of women has been a history of silence.
—HILLARY CLINTON1
ARE WOMEN THE “SILENT SEX”? At first glance, the answer would seem to be no. It is difficult to call to mind a scene in which men outtalk women in everyday life. Women are often thought to be more sociable than men, not less. But when it comes to politics and public affairs, there is reason to take a closer look. Women are much less likely than men to take action to directly influence others. There is reason to wonder, then, whether women are also more passive than men when it comes to public discussion.
The notion that in advanced countries women tend to be less active than men in deciding matters of common concern may strike some as outdated. The voter registration rolls in America reveal clear gender parity. And women are no strangers to volunteerism in public life. Women are not reluctant to take up the cause of civic duty. But the equality we see in these modes of participation disappears when we turn to the forms of participation most concerned with the exercise of power and authority. Thus we have a puzzle on our hands: why do women participate less than men in some ways but not in others, and do so despite massive advances in their standing in society?
Our answer in this chapter is that women have the ability to participate equally, but do not use it. By ability, we mean concrete resources such as time, money, access to social networks, a store of relevant knowledge, and appropriate experience and skills (Verba, Schlozman, and Brady 1995). But ability is not enough. What active political participants need that women tend to lack is the motivation and the opportunity to engage in actions that society associates with authority. Women tend to be more passive than men in stating controversial opinions and directing others to implement those views. That is, they are less motivated to carry influence in the allocation of values and resources, because that activity is deemed masculine territory.
The motivation to be political is socially acquired, and the process of acquisition runs through social identity. “The gender difference in the taste for politics,” Nancy Burns and her colleagues write, may be “rooted in the fact that men have a basis for identifying with the vast majority of the key players in politics” (Burns, Schlozman, and Verba 2001, 341). Because it sends important messages about women’s place in the political world, “the gender composition of the political environment … has consequences for women’s engagement with politics” (356). The result is that women tend to view themselves as outsiders in the locker room of politics, and they do not engage as fully—as authoritatively—as they might across the complete set of participatory activities available to them. In this sense, they are not full citizens. Much the same can be said about women’s participation in public meetings; the political environment may send the message that women are expected to be only partial participants.
Thus women may enter public discussions already inclined against making the full contribution of which they are capable. While women are now better educated and more civically experienced than men, they remain the less authoritative, and therefore relatively “silent,” gender in public affairs. Women have plenty to say in formal meetings, but they do not say it as often as men do, because they have learned to avoid being the leaders—in other words, they have been motivated to be the followers.
THE GAP EXISTS, BUT IT’S NOT BASED IN ABILITY
Are women in fact less likely to participate in authoritative political acts? That is, are women less likely to attempt to voice their opinions to others, to direct others in their implementation, and in various ways, to attempt to influence others to act on them? For our initial answer, we turn to the most comprehensive treatment of gender inequality in political participation to date: Burns, Schlozman, and Verba’s Private Roots of Public Action (PROPA).2 When they combine together a variety of types of participation, such as voting, contributing money to a political campaign, or working on a community problem, they find a moderate gender gap.3 This gap appears to be caused by an accumulation of small differences in the resources that shape “pathways” to participation. For example, a white collar occupation gives skills that allow a person to write a letter to their representative. Earning more income enables political contributions. Education gives people skills, knowledge, and motivations that turn on their radar for politics, enable them to process political information, and drive them to participate. Education gets people to turn on, tune in, and drop in, to modify a phrase from 1960s counterculture.
However, women’s continued lag in participation is puzzling given that women have made enormous strides in employment, earnings, and education. While women are still less likely to be employed full time than men, they now make up about half of the labor force. They still earn less, but now they earn 80% of men’s income when they are employed (US Department of Commerce 2011).4 In fact, according to the US Bureau of Labor Statistics, women now compose just over half of workers in “high-paying management, professional, and related occupations.”5 Consistent with this pattern, young women lag behind young men in political contributions but exceed them in contributing money for charities (Marcelo, Lopez, and Kirby 2007). Women do not have a problem accessing and using money in politics—they have a problem with politics. And the same holds for many of the forms of engagement we wish to see in the citizenry. As we will see, the biggest variable of all—education—now should offset the disadvantages from the small disparities in employment and income, positioning women to equal if not exceed men in political engagement. Given the enormous strides women have made in employment, education, and income, we need to look further for an explanation for why their political participation continues to lag. Only then can we accept the possibility that women may participate less than men in public discussions, and do so despite overall equal resources.6
Adding to the puzzle is that women have completely closed the gender gap in some modes of participation and engagement even as they continue to lag in others. Women have been voting at slightly higher rates than men since the 1980s. This holds not only in the United States but in Western Europe too, and in the 1990s the gender gap in turnout disappeared in a variety of countries worldwide.7 Because women are slightly more numerous than men, women are actually a majority of the electorate—a reversal unthinkable to the suffragists of a century ago (Marcelo, Lopez, and Kirby 2007). Young women outvote young men by more than older women outvote older men, so women’s advantage as voters is trending up (Godsay and Kirby 2010). That is, women today are dutiful when it comes to voting.
But duty may be a key word here, since it is a chief reason for the decision to turn out to vote (Achen and Blais 2010). In fact, when asked the reason for their voting turnout, 90% of women mention civic duty. Policy, social, or material motivations come in an eye-squinting distant second, third, and fourth place, respectively (Burns, Schlozman, and Verba 2001, 116).8 Consistent with the duty explanation for women’s political activity is the fact that among the youngest cohort of citizens, women are somewhat more likely than men to participate in charitable activities, but men are more likely than women to participate in electoral activities other than voting (Marcelo, Lopez, and Kirby 2007). That is, women no longer underparticipate men at the ballot box or in other duty-driven acts, but the gender parity does not extend to other types of participation.
Further underscoring the puzzle is the fact that women also have as much as or more than men by way of early experience that trains people to get moving in politics. Women are as likely as men to report that they were “very active” in student government or in clubs during college (Burns, Schlozman, and Verba 2001, 149). Women are more likely than men to recall having been “very active” in high school clubs. They are just as likely as men to recall having been “very active” in high school student government and to take civics classes. In real time, high school girls actually report more of these feeder activities, including taking part in student government, and being active in academic clubs and student publications (Burns, Schlozman, and Verba 2001, 148). That girls are advantaged here, by rates of 50% or even 100%, is surprising. These are powerful factors in fostering an interest in politics in adult life, even decades later (Shani 2010).
And yet women do not seem to be leveraging these formative experiences to the same extent as men, at least not consistently across the forms of political engagement. Burns, Schlozman, and Verba found that the same young women whose participation rates in student clubs and government equal men’s are far less likely than men to indicate that they are interested in “what is going on in government” (a gender gap of 15 points—among the biggest gender gaps one sees [Burns, Schlozman, and Verba 2001, 150]). Niemi and Junn’s large national study found, similarly, that school-age boys are more likely than girls to enjoy civics classes, and boys score higher on measures of political knowledge, despite appropriate statistical controls (1993). While young women are civically active, they are not politically interested. Put differently, women have the resources, the experiences, and the sense of civic duty, and these factors cause them to participate actively in activities deemed dutiful, but women still lack the motivation to engage in activities that are not construed as dutiful.
When we look beyond actions that women may view as their civic duty or which fit closely with women’s existing role as support staff to men, we see continuing trouble spots in women’s steady march forward. These trouble spots tend to occur in the more authoritative political acts. Women are less likely than men to discuss politics, to try to persuade someone to vote for their preferred candidate, or to offer opinions about politics. This pattern conforms to the notion that the difficulty extends to psychological engagement—that is, to motivation. Women are less likely than men to indicate an interest in politics, to learn about politics, to enjoy political discussion, and to feel a sense of political efficacy (the sense that one can do something about politics).9 The magnitude of the gender gap is modest—about 10 percentage points. But it is persistent and difficult to explain (Burns, Schlozman, and Verba 2001, 102–3).10 And in some settings, the gender gap is massive. For example, women submit just 10% of op eds to prestigious news outlets in the United States.11 And even when individual differences between men and women are modest, they cumulate significantly. Women’s relative lack of voice can add up: public officials receive approximately two million more letters or calls from men than from women (Campbell and Wolbrecht 2006, 234). In sum, women lag in areas that constitute the core of political influence—perceiving themselves as influential (as efficacious), expressing opinions about public affairs, and attempting to direct those affairs.
Neither is this gender gap going away any time soon. In a striking illustration of how the more things change the more they stay the same, Atkeson and Rapoport found in their review of a half century of American public opinion surveys that women are less likely than men to offer an opinion when asked by a survey interviewer what they like or dislike about the parties or candidates (2003). They don’t even muster the same level of opinionation as men when asked about candidates for president, many of whom were incumbent presidents. And if any political figure is going to elicit an opinion, it is likely the president. Women are less likely to offer any opinion at all, and offer fewer opinions when they do offer something. As the authors put it, “it is interesting that this difference has shown little change over the past 50 years.” Interesting indeed. Even more telling for our purpose, Atkeson and Rapoport could not make the gender gap go away even after controlling on every political and psychological resource under the sun (Atkeson and Rapoport 2003). Still more interesting, the less respondents talked during the interview about their likes and dislikes about electoral politics, the less they reported trying to persuade others to vote for their preferred candidate—and this effect obtains especially among women. Atkeson and Rapoport conclude that the willingness to express an opinion has important consequences for involvement in discussion and for agenda setting. As PROPA summarizes in its two-decade review of the trajectory of the gender gap in participation, “the lines for women and men move in tandem and almost never converge or cross” (Burns, Schlozman, and Verba 2001, 69). While voting is an exception, the most leader-like activities—offering opinions, attempting to persuade, feeling that it is possible to effect change or make a difference—show little progress despite the dramatic rise in women’s position in society.
When it comes to the focus of our study—public meetings—we find strong confirmation of these trends. In their 1990 Citizen Participation Survey, for example, Verba, Schlozman, and Brady asked respondents to imagine that they were attending “a community meeting and people were making comments and statements.” They then asked each respondent to tell them whether he or she could “speak well enough to make an effective statement at such a meeting.” Using these data, Karpowitz (2006) finds that men are significantly more likely to believe that they could effectively speak up at meetings than women, even after controlling for political knowledge, education, vocabulary, and past experience, such as making speeches at work, church, or in other organizations. In other words, the gap in willingness to participate in public meetings has nothing to do with ability or intelligence. And it persists despite the fact that the gender gap in meeting attendance weakens or evaporates in the presence of controls for civic skills and social connectedness.12 Thus even when we control for the factors that help to explain many other types of participation, women are less likely than men to believe themselves capable of speaking up and effectively contributing to the discussion. In this sense too, we see that women participate less than men when it comes to the more authoritative acts, those that carry the most influence.
But what about education? No resource matters more to political participation than education. Surely women close the engagement gap when they close the education gap? Well, actually, not so. Even when women have high levels of education they remain far behind men when it comes to feeling entitled or qualified to talk about and to take meaningful action in politics. In fact, when Hansen (1976) studied the effect of education on women in the 1972 election, she found that education made women less likely to participate relative to their male counterparts (Hansen 1997, 74). This is a head scratcher in political science if ever there was one. If political scientists awoke one morning to the news that social science dictators had taken over the academy and they would be allowed to study just one cause of citizens’ behavior, most would choose education. It is the single most important predictor of political action (Wolfinger and Rosenstone 1980). And yet we see that in some cases, it utterly fails to move women to act. Young women closed and then reversed the college degree gap in the late 1990s.13 Overall, today’s women have the same rate of college graduation as men—30% (2010 US Census).14 But the things that education is supposed to give them—knowledge, interest, and efficacy—did not show a commensurate elevation. Young women are significantly less likely than their male cohort to try to persuade others in an election and to follow news about politics—even though they tend to have more education.15
Put differently, women lag behind men primarily because the resources that forcefully propel men to take action merely give women a gentle nudge. When a man completes college, he becomes much more informed about politics; when a woman completes college, she becomes only a little more informed. Economists talk about this effect as “differences in the returns” on education, since they think of education as an investment that yields high profit to some and low profit to others. For example, Dow found that the primary reason for the 10-percentage-point gap between men’s and women’s political knowledge from the mid-1990s to the mid-2000s is that women enjoy lower “returns” from education (Dow 2009).16 The point is that despite having closed—and even reversed—the gap in higher education, women have yet to benefit fully from this achievement when it comes to politics. College educated women remain behind college educated men in their level of engagement.17 Lower education is not the reason that women are less engaged than men. Women are now better educated, but even at the same level of education, women are less politically engaged than men.18
Much the same applies in the other primary route to political participation—civic organizations. Churches are powerful mobilizers in American politics, and they provide skills to ordinary people who, because of class or race, would otherwise tend to be even more quiescent and politically passive (Verba, Schlozman, and Brady 1995). However, a recent study of gender in churches finds that “the social conditions that at first glance offer the chance for more recruitment of women turn out to have the opposite effect” (Djupe, Sokhey, and Gilbert 2007, 917). Women are not given the same chance as men to participate in politics via churches, even though women are more numerous in the membership, and more active within churches (Djupe, Sokhey, and Gilbert 2007). Access to social networks that recruit people to participate in politics is one of the important resources that form the ability to participate. By providing that access, churches act as an important institution for overcoming the participatory deficits of other social cleavages, such as race and class (Verba, Schlozman, and Brady 1995). But women are not gaining the benefits from their ready access to that resource.
To further test the notion that the gender gap in engagement persists despite equal resources, let us look at the youngest cohort of Americans, ages fifteen to twenty-five. These women are more likely to be educated than the men of the cohort. Moreover, they have not yet experienced the suppressing effects of lower workforce participation and less income characteristic of the cohorts aged thirty+.19 These are women who match, or out-participate, men in student associations and government, who are voting more than men, and who are more active in civic associations—all of which should be directly increasing their involvement in other venues of political participation. We should see these women out-participating men in every form of political engagement.
And yet while these young women do outvote men and participate in charitable activities more than men, they are less likely than men to try to persuade someone during an election and to follow politics in the news.20 Higher rates of young men than young women report that they “follow what is going on in government and public affairs ‘most of the time,’” while higher rates of women than men report “rarely or never.” That more of these women than men report “rarely or never” following public affairs in the aftermath of 9/11 is a particularly urgent warning bell that we are missing an important factor shaping women’s engagement with their society (Marcelo, Lopez, and Kirby 2007, 16). On the important question of whether the “government typically responds to the genuine need of the public,” women were 11 points more likely to decline to give their opinion, choosing instead to indicate that they “haven’t thought much about it” (Marcelo, Lopez, and Kirby 2007, 15).21
Most relevant to our argument, young women’s motivation gap is evident in self-perceptions of political competence. Rapoport (1981) studied teenagers’ expressions of opinion on issues, some of which were political and others not. He found no gender differences in the proclivity to express opinions on nonpolitical issues. However, boys were more likely than girls to express their views about politics—even when controlling on their actual level of information. Rapoport’s analysis further found that these patterns persist into adult life.22
The heart of the problem is not really women’s inequality of concrete resources, sense of civic duty, faith in government, or skills, whether obtained from education, civic participation, or elsewhere; rather, the problem is women’s lower motivation to be politically authoritative. In predicting overall participation with all the variables that may matter, Burns, Schlozman, and Verba found that concrete resources explain only one-third of the gender gap. Another one-third is unexplained, and the remaining one-third is explained by psychological engagement—interest, knowledge, and efficacy.23 The gender gap is not rooted primarily in the standard pathways to participation. Even when equal to or exceeding men in all these pathways, women are less motivated to engage with politics. Women are notably less likely to think about politics, to seek information about politics, to have opinions about politics, to enjoy politics, to try to influence others about politics, and crucially, to consider themselves able to speak effectively about political issues in public meetings. And that engagement is itself largely unexplained by concrete resources and structural pathways. After conducting the most comprehensive study of political knowledge to date, Delli Carpini and Keeter found that resources and similar structural factors explain only half of women’s shortage of political knowledge (Delli Carpini and Keeter 2005, 29, 205–6). As Atkeson concludes her own study, “Political resources, though important in understanding political engagement, are not enough” (Atkeson 2003, 1052–53). What can explain the gender gap in political engagement is motivation. Because politics involves the exercise of authority, and authority is gendered, women are less likely than men to be motivated to be political.
Women out-participate men in voting and in civic acts of various kinds—even in financial contributions to charities—but not in the activities central to the exercise of political authority. It is not the nature of the act that explains the current gender gap; we see that women perform the same actions when the target is not political. Women are less likely to have opinions and information about politics and to talk about politics and public affairs. As Sapiro argues, this renders women more subjects than citizens (Sapiro 1983). Subjects are dutiful; citizens are willful.
This conclusion raises further, and more difficult, questions. Why are women more prone than men to act as subjects? Why are women the “quiet sex” in public affairs?
INADEQUATE OPPORTUNITY DEPRESSES THE MOTIVATION TO EXERCISE POWER
To understand the gendered nature of the motivation to carry political power, we can begin by asking why women are less motivated from their access to the pathways of participation. A clue to what may be preventing women from benefiting from the feeder activities is what goes on in these settings. As Hansen writes, education may not empower women as expected because “the content of that education and the social setting in which it occurs continue to devalue women’s experiences” (Hansen 1997, 78). We go further than Hansen; these experiences are not only devaluing women’s experiences, they also construct women as less authoritative than men, and that has pernicious civic and political consequences.
In that sense, higher education and civic associations are not reaching their potential as opportunities to inculcate in women the motivation to participate in meaningful political activities. Instead of assisting women in establishing authority in their community, they may be doing the reverse. Here we draw on Delli Carpini and Keeter’s helpful framework for understanding the sources of political knowledge (1996). They argue that political knowledge is based on three factors: (1) the ability to understand politics, (2) the motivation to gain knowledge and to reason about it, and (3) the opportunity to do so. In this section, we argue that as women gain ability—by obtaining higher education and participating in civic organizations—they lose motivation. That is because the gendered experiences in these settings reinforce women’s gendered lack of authority.
A number of reports by academic organizations have suggested that the experience of many girls and women in educational settings is less empowering than men’s. The American Association of University Women, for example, reported that women are less likely to speak in class, and that teachers are more likely to call on boys than girls (AAUW 1992). The general conclusion of the study is that standard practices in educational institutions are better at inviting or recognizing male than female speech (Hansen 1997). These settings represent an opportunity to buttress women’s sense of competence, confidence, and thus, authority. As one female student put it, “Every class discussion where my ideas are respectfully listened to and validated, I grow in confidence and in my grasp of the subject.”24
Several reports on elite universities second this conclusion. We delve in depth into these reports for several reasons. First, they provide apt illustrations of the general phenomenon of the gender gap in educational settings. Second, they suggest what it is about the experience of higher education in particular that may be keeping women back. Third, they offer a revealing glimpse into the group-level dynamics in speaking behavior that will become the heart of our analysis in later chapters.
For example, studies of actual participation in Yale Law School class discussions, conducted in 2002 and again ten years later, find that men are as likely to outtalk women in 2012 as they did in 2002.25 Interviews with female law students suggest that the class dynamics cause some women to lower their self-confidence. As one student said, law school holds up as the ideal lawyer an “image of the dominant male lawyer.” Furthermore, as women grow more silent and men more talkative over the course of their law school experience, this may create a sense among women that the successful and influential members of their milieu are male and that women are outside on the margins. As one female student said, “There’s very much this male in-group here … I feel it’s very fratty and very insular, even more so than the Law School itself.” In other words, women’s experience becomes one of looking on as men talk with other men. This in itself could reinforce women’s feelings of low confidence and failure. Again, we see that despite gains in women’s status in society generally and education in particular, women’s unequal participation in public discussion remains. The gender gap does not disappear with education or among the privileged and elite segments of the population. It is reinforced by the interaction between men and women in settings where talk carries influence, and unequal talk becomes an indicator of unequal status.
An official report by Princeton University also sheds light on current gender gaps and possible reasons why women do not translate experiences in education settings into equal political engagement. Princeton, like the other Ivies, was all male for most of its history. While the university has closed the gap in undergraduate female enrollment, women’s status within the institution continues to lag behind men’s (Princeton University 2011, 21).26 While women do a “large proportion of the important work” in student groups, and while some run for elected office in those groups, women are less likely to run for the most visible posts of these organizations, such as president, and far less likely to occupy the most prestigious campus-wide positions (Princeton University 2011, 5, 67).27 In the 2000s, even as women’s proportion of students was close to 50%, women held only 14% of the most important undergraduate elected posts on campus; this figure represents a 50% decline over the previous decade and a significant decline from the decade before that.28 That is, women’s leadership status declined as women’s presence in the student body rose. As noted in the report, Princeton is far from alone; Harvard’s student government recently passed a resolution to address the large gender gap in its ranks (Princeton University 2011, 71).
All the more striking is the fact that when the entire Princeton freshman class of ’14 was surveyed before arriving on campus (in a survey that enjoyed unusually high response rates), there was no gender difference in intent to seek leadership positions in campus organizations, and women were more likely than men to indicate that they planned to seek leadership roles in campus activities after arriving on campus (Princeton University 2011, 39).29 However, a mere eight weeks later, in their reinterviews, women were about 10 percentage points more likely than men to have changed their minds about seeking leadership posts, and this effect held across income groups (Princeton University 2011, 41). The gender gap had reared its head.
The possible reasons are several. Women who express an interest in prominent leadership positions are “sometimes actively discouraged by other students” (Princeton University 2011, 5). Some women explicitly state that they don’t run because they don’t think they’ll win (69), much as women in politics are more reluctant than comparably qualified men to run for office (Fox and Lawless 2011).30 Women may also be more likely to make self-effacing statements, while men may be more likely to make self-promoting ones, which may explain why people may take women less seriously. For example, men in classes may tend to “speak up more quickly” and express thoughts before they are fully formed, while women may be more likely to speak after they have figured out what to say and how to say it best (Princeton University 2011, 60).
Echoing the findings of the Yale study, students report that in some classes, participation is highly unequal by gender and that those tend to be more conflictual; as one student put it, “there are two kinds of precepts [discussion sections]. Some where everyone participates, and some that are just two people screaming at each other. The two people screaming scenario is always two men” (Princeton University 2011, 60). As in the Yale study, women may be particularly alienated when men dominate discussion because men are taking the floor and because women tend to dislike conflict (more on this in subsequent chapters). Finally, the gap does not come from lower capacity. At Princeton, female GPA slightly exceeds male GPA, and the worst performers are predominantly male (Princeton University 2011, 60). That holds as a general finding at universities: women on average exceed men in academic achievement, suggesting that women are not falling behind in leadership because they are less academically capable (Heckman and LaFontaine 2007).31 A final possible reason for the gender gap in leadership in higher educational institutions is that women may need more mentoring and peer support before they are willing to seek leadership, to enter competitions for fellowships or awards, and to think of themselves as meritorious. Women may especially need support and “affiliation with other women.”
As in studies of national samples, so too in this Ivy League school, we see that women are at least as active as men in some parts of the life of the community but are far less likely than men to lead that community. Women may be less likely than men to self-promote, and more likely to self-efface. Especially relevant to our study, women may often find themselves as observers in classroom discussions, while men carry on the academic debate. Women may also need more encouragement and messages of affiliation than men do. For these reasons, they are not leveraging school experiences with civic participation into adult participation and interest in politics. In the settings where men are gaining authority and preparing to take leadership roles, women are not developing authority or building influence to the same extent.
The Media and Elections as a Missed, or Realized, Opportunity
Women’s motivation to exercise power rests not only on their interactions within civic institutions but also on the subtle signals they receive from the information environment. The lag in women’s motivation is rooted in part in the gendered portrayals of who exercises power. Despite progress over time, the media, and election campaigns, still frequently links authority with men. On the flip side of the coin, the opportunity to gain motivation can come in the face-to-face settings that women frequent and in the cues that trickle down to them as they consume the news or entertainment media.
Women continue to be portrayed in the news and entertainment media in heavily gendered ways. Although the overt sexism of the past has weakened, media coverage of candidates tends to be quite gendered. Coverage of gubernatorial, senatorial, and presidential candidates dwells more on women’s than on comparable men’s politically irrelevant personality traits, family, and appearance.32 Striking examples include commentary in the Washington Post and New York Times on presidential candidate and senator Hillary Clinton’s low neckline and “over-reliance” on suits. Similarly, media commentator Bill Maher commented on Clinton’s display of emotion during the 2008 primaries, that “the first thing a woman does, of course, is cry” (Lawless 2009, 71). This emphasis can be so skewed that coverage of the candidate’s appearance or personal behavior can dominate the coverage of her issue positions (Aday and Devitt 2001). The amount of coverage can fall short of the candidate’s standing in the polls, as in the case of Elizabeth Dole, who in the 2000 primaries was the first prominent female candidate to seek the Republican nomination for president (Heldman, Carroll, and Olson 2006). Female candidates’ viability appears more fragile than comparable men’s.33
These media representations matter. Kim Kahn’s experimental evidence shows that this “distorted” media coverage shapes perceptions of women’s authority (Kahn 1992; 1994b; 1996). The typical media stories of male candidates produce higher viability ratings and leadership ability ratings than the typical media stories of female candidates, which emphasize lack of viability and give relatively little attention to the candidate’s issue positions (Kahn 1992). This evidence is consistent with the notion that voters tend to view men as more quintessentially leader-like in part because they view coverage that emphasizes this trait. Not surprisingly, while female candidates are perceived by voters as better able to handle care issues (health, education, social welfare), they are viewed as less capable on issues deemed men’s domain (crime, military, foreign affairs, financial matters); and most important for our point, female candidates are viewed as having a more compassionate personality than men, while men are viewed as more assertive and confident, “tougher and more decisive”—that is, men are deemed better at the core traits of authority.34 The masculine traits are perceived as better indicators of competence on a wider range of issues, that is, they are closer to the traits of a desirable leader (Huddy and Terkildsen 1993a; 1993b). Masculine traits are also deemed a more important qualification than feminine traits for the most powerful positions—national and executive offices (Huddy and Terkildsen 1993a, 518). Finally, even the seemingly positive coverage of female heads of state as the “first woman” in a high-level office could backfire by presenting women as anomalies in the masculine domain of power (Norris 1997a; 1997b). While the personnel who operate the news media often attempt to adopt a gender-neutral approach, the media continues to send often-subtle signals that women and political power fit together uneasily (Braden 1996).
At least as much as news coverage of leaders, everyday popular culture affects women’s motivations to exercise power. As Murphy put it, “Perhaps nowhere is the view of minorities and women more partial and inadequate than in the mass media. … Moreover, the women who do appear are typically portrayed as passive, overemotional, dependent on men, and inordinately concerned with ‘getting rings out of collars and commodes.’”35 As one scholar of gender and media summarizes, “the media reinforces the notion that for women the secret to true power lies not in personal achievement or economic independence but in the guise of beauty” (Roessner 2012, 330).36 Moreover, Murphy’s experiments demonstrate that stereotypic media portrayals of women prompt people to make more gender-stereotypic judgments of specific individual women in prominent cases of males accused of wrongdoing. And the clincher is that even when told that the female characters they viewed are fictitious, the effect persisted (1991).
Further evidence that media representations undermine women’s authority comes from a clever social psychology experiment—one with special relevance to our interest in discussion groups—in the tradition known as “stereotype threat.” Paul Davies, Steven Spencer, and Claude Steele (2005) showed college students either a set of television product ads without people, or the same set of ads plus ads displaying women in traditional gender roles, such as women who aspired to be a homecoming queen. Then the students were asked to work in small groups to solve a common problem. They were asked whether they were interested in functioning as either a leader or a follower in the group. The women exposed to the gender-stereotyped ads indicated less interest in serving as the group leader and more interest in being a follower, while the ads had no effect on men. Merely reminding women in a subtle, indirect way that women fill feminine roles may signal to women that society values them for their femininity and that they are expected to engage in feminine behaviors. This may depress their motivation to take leadership roles. Society may be continually sending such signals to women and thereby creating the sense that leadership is not feminine and thus not valued for women. The opportunity available to women to view themselves, and to be viewed, as properly exercising authority is often limited.
However, there are cases where women do enjoy the opportunity to raise their sense of authority. That women’s motivation to engage with politics can rise with opportunity is keenly apparent in the effect of visible female candidates. When women run for a high-level office, female voters increase their sense of efficacy and their level of political information, while men do not (Burns, Schlozman, and Verba 2001; Hansen 1997). The higher the number of women who run for important offices, the stronger this effect becomes (Burns, Schlozman, and Verba 2001). The visibility of female candidates influences the behavior we have focused on as an indicator of the exercise of authority—political proselytizing (Atkeson 2003; Hansen 1997). Even more telling for our argument, Campbell and Wolbrecht found that “over time, the more that women politicians are made visible by national news coverage, the more likely adolescent girls are to indicate an intention to be politically active” as adults (2006, 233). The effect works in part by elevating adolescent girls’ political discussion, especially within the family.37 These results add to the evidence we have reviewed so far to form a consistent pattern. Women benefit from opportunities to view their influence as women as legitimate. These opportunities do their work by signaling to women that they do have the necessary authority. As Atkeson writes, “Despite women’s legal ability to participate in the political system for over 80 years, the lack of visible female political players has helped the gender gap in political engagement to persist. The cues provided by minimal female representation in politics and campaigns are that women are not full citizens and that they are not welcome in the political world” (2003, 1053).
CONCLUSION
Women have certainly come a long way in Western society and politics. American women, like their counterparts in other advanced economies, have narrowed or even reversed the gap in important resources that form the prerequisites of political participation. Women have become reliable voters and civic activists. In other words, they are the foot soldiers of democracy. But they are not the generals. Women lag considerably on the more consequential forms of political activity. They have not turned their growing progress along the pathways to political participation into commensurate political influence. Why is that?
In this chapter we argued that the answer does not lie primarily in the standard list of participatory resources. The United States has experienced a shrinking gender gap in workforce participation, occupational status, and income. When it comes to the single most important antecedent of participation, namely, education, American women have even achieved a higher level than men. In addition, they are more involved than men with voluntary associations during childhood and adulthood. So women are actually advantaged over men in these two crucial pathways to participation. All these resources and opening pathways have given women the ability to participate. Education and civic experiences have provided women with cognitive resources and skills that enable them to engage with politics—should they choose to do so. Yet women continue to lag behind men in making that choice. Put differently, there is no longer a gender gap in ability, but there remains a gender gap in motivation and opportunity.
The opportunities to participate in the public sphere have bifurcated effects on women. The institutions that provide women the ability to exercise political influence may interfere with their motivation to do so. By implication, women can close the gender gap with men in participation and influence when the institution provides an opportunity to do so. The key is to structure the institution so that it elevates women’s motivation to participate. The way to do so is by elevating women’s authority in those institutions and to provide women with opportunities to view women as properly exercising power.
Our story in the next chapter takes off from this point of departure. We turn back from an analysis of participation in general to focus on participation in public discussion specifically. We will explore in greater depth how and why women may have a motivation gap with men, and why they may experience an opportunity deficit relative to men. Moreover, we will explore how opportunity and motivation could work in tandem. As the college studies showed, lower motivation to engage, and relative passivity in public discussions, may arise from women’s reactions to the gender dynamics they experience in the discussion itself. We then turn to the question of how the opportunity to contribute to discussion is structured by the group itself, and what types of motivations may contribute to the gender gap.
1 “Women’s Rights Are Human Rights,” speech by Hillary Clinton, United Nations Fourth World Conference on Women, Beijing, China, September 5, 1995. http://www.columbia.edu/cu/augustine/arch/hrclinton.txt.
2 The previous classic work on the subject, Voice and Equality, reported that women participate only slightly less than men overall (Verba, Schlozman, and Brady 1995, 254).
3 Other acts included in this participation measure are working on a campaign, serving as a volunteer on a local government board, contacting public officials, membership or contributions to organizations that take stands in politics (Burns, Schlozman, and Verba 2001, 64). In this chapter, quantitative reports of individuals’ patterns of political activity or psychological engagement are typically based on a person’s self-report of their activity and engagement (e.g., their level of political interest). The exception is the level of political knowledge, which is usually measured by asking factual questions and noting the accuracy of the answer. When self-reported turnout is validated by checking actual turnout records, the same variables that affect one affect the other, albeit somewhat less so (Achen and Blais 2010). We do not know if actual vote matches self-reported vote in the same way for women and men.
4 According to a Bureau of Labor Statistics report, in 2010, 59 percent of working-age women in the United States were in the labor force—employed or actively seeking employment (US Department of Labor 2011).
5 According to Pande and Ford 2011, 2, citing US Department of Labor, Bureau of Labor Statistics, Employment and Earnings, 2009 Annual Averages and the monthly Labor Review, November 2009.
6 We can also rule out an additional possibility—that women are too busy with their domestic duties. This plays only a small role in explaining the gender gap in participation (Jennings and Stoker 2000; Burns, Schlozman, and Verba 2001).
7 Desposato and Norrander 2009; Inglehart and Norris 2003; Verba, Schlozman, and Brady 1995, 255.
8 The importance of duty also applies to men. Women are also about as likely as men to work on campaigns and to protest (Burns, Schlozman, and Verba 2001, 65; Verba, Schlozman, and Brady 1995). Women are no more likely to participate in local affairs than men, less likely to discuss local politics than men, and are less efficacious than men about local politics, even though they are as interested in local affairs as men and are more likely than men to know the name of the local head of the school system (Burns, Schlozman, and Verba. 2001, 67, 102–3). Women are equally knowledgeable with men regarding local politics (Delli Carpini and Keeter 2005).
9 These gender gaps in discussion, persuasion, opinionation, and psychological engagement are documented in Atkeson and Rapoport 2003; Burns, Schlozman, and Verba 2001, 102–3; Conover, Searing, and Crewe 2002; Delli Carpini and Keeter 1996; Delli Carpini and Keeter 2005; Hansen 1997; Huckfeldt and Sprague 1995. Not only does gender affect the average level of engagement, it also affects extremes: men are more likely to be news and political junkies (Burns, Schlozman, and Verba 2001, 105). Women are also far less likely to make big campaign donations (Burns, Schlozman, and Verba 2001; and here is one anecdote from 2008: http://www.usatoday.com/news/politics/election2008/2008-07-30-gender_N.htm).
10 Women score higher on measures of knowledge based on visual recognition of political actors than on verbal measures (Prior 2004, 19). On verbal measures, men’s knowledge is frequently at least 1.5 times greater than women’s (Delli Carpini and Keeter 2005). A very important caveat is that when studies contain sufficient numbers to analyze separately African American and Hispanic youths, they find that the gender gap reverses, with girls more politically knowledgeable and efficacious than boys (Gimpel, Lay, and Schuknecht 2003, 87, 92).
11 The 90% figure comes from a study by the Washington Post that tracked five months of submissions to the op ed pages of that publication. This was originally reported in the Post by Deborah Howell in May 2008, according to http://www.theopedproject.org/index.php?option=com_content&view=article&id=70&Itemid=82. The Op Ed Project reports similar percentages at other prestigious news outlets.
12 In their 1990 Community Participation Survey, Verba, Schlozman, and Brady ask about attendance at “any official local government board or council that deals with community problems and issues such as a town council, a school board, a zoning board, a planning board, or the like.” With these data, Karpowitz (2006) finds that men are more likely to report attending such meetings than women, but the gap falls just short of statistical significance in the presence of controls for political experience and social connectedness. These data are self-reports of meetings generally and do not necessarily reflect the gender composition of any single meeting.
13 Burns, Schlozman, and Verba 2001, 144; US Department of Commerce 2011.
14 http://www.census.gov/newsroom/releases/archives/education/cb11-72.html. The percentage is among those twenty-five years or older. Women twenty-five and older were more likely than men twenty-five and older to have completed at least high school, at 87.6% versus 86.6%. Among the youngest cohort (ages twenty-five to twenty-nine), women finish college more than men (36% vs. 28%). Interestingly, a study by the National Bureau of Economic Research finds that women are returning to pre–World War II levels in this regard; women composed about half of college students from 1900 to 1930, mainly in teacher-training colleges. The low point for women in college attendance was shortly after World War II, when the GI Bill subsidized (mostly) men; men exceeded women by 2.3 to 1 (Goldin, Katz, and Kuziemko 2006).
15 Results from a 2006 national survey of young people conducted by CIRCLE at the University of Maryland. The Civic and Political Health of the Nation Survey (2006 CPHN) interviewed 1,674 young people and 547 adults on civic and political activities performed over the past twelve months, measuring nineteen activities. It oversampled nonwhite youth (Marcelo, Lopez, and Kirby 2007).
16 When we calculate the gender gap in political knowledge from table 5.5 of Delli Carpini and Keeter, we see clearly that at every education level, women know less than men, by a magnitude of 7 to 18 percentage points. For example, the chance of a college educated person correctly naming both of their US senators is 18 percentage points higher for men than for women (1996, 198).
17 Delli Carpini and Keeter 1996, table 5.5, 197–98.
18 One possible solution to the puzzle we have been describing is that education may not have much of an effect on anyone, man or woman. Instead, the causal arrow may run the other way; it may be that people already inclined to participate, engage with, and know about politics are also the people who tend to choose higher education (Highton 2009). That might explain why getting educated does not drive women very much—perhaps, while men who choose to obtain more education are also more active in politics to start with, women who choose more education are not more politically active than women who forego education. Regardless, this leaves us back where we started: why don’t women participate as much as men despite having similar resources? And why do we see “continued and unabated differences between men and women in their willingness to openly express political attitudes” (Atkeson and Rapoport 2003, 495)?
19 One factor that does not affect women’s participation, surprisingly, is having children, even young children, and time spent on household chores also has no effect (Burns, Schlozman, and Verba 2001, 320, 331).
20 Results in this paragraph come from Marcelo, Lopez, and Kirby 2007 and Zukin et al. 2006.
21 That is the single biggest gender difference across the nineteen forms of civic or political engagement in the survey. One other explanation that may at first seem decisive also falls flat. Perhaps women believe that government is not a likely, or appropriate, venue to address the needs of the polity. Perhaps women are more cynical or distrusting of government, and that is why they are less interested in what it does and less involved in its activities. This hypothesis turns out to have no support. Young women actually exceed young men in their desire to see the government do more to address problems and are more likely than men to reject the notion that government does “too many things” (Marcelo, Lopez, and Kirby 2007). (In both 2002 and 2006, CPHN survey of fifteen-to twenty-five-year-olds conducted by CIRCLE.) Young women are no more likely than young men to endorse the notion that “government is not responsive to the genuine needs of the public” (Marcelo, Lopez, and Kirby 2007, 15). They are slightly more likely than men to indicate that “government often does a better job than people give it credit for,” as opposed to “government is almost always wasteful and inefficient” (13).
22 His subsequent study of “Don’t Know” (DK) responses found significant sex differences in DK response rates as well (Rapoport 1982).
23 Burns, Schlozman, and Verba 2001, 268, table 10.10.
24 Reader comment by Taffy C, September 19, 2012. http://jezebel.com/5944642/women-speak-75-less-when-theyre-surrounded-by-dudes-and-thats-bad?post=52822328.
25 The study observed participation in 113 sessions of twenty-one Law School courses, and surveyed 62% of the Law School students. “Yale Law School Faculty and Students Speak Up about Gender: Ten Years Later,” by Yale Law Women, cited in http://www.yaledailynews.com/news/2012/apr/20/gender-imbalances-found-in-law-school/ and in Carol Bass, Yale Alumni Magazine, July/August 2012, p. 26.
26 Women went from 0% in 1968 to 50% in the class of 2013.
27 Some of the men holding the top posts joked in focus groups that while they take the credit, they rely on women holding the supporting positions below them to ensure that “work gets done” (Princeton University 2011, 68).
28 That is 14% of 70 (the 7 most important offices, assuming one person per office per year, times 10 years, equals 70) (Princeton University 2011, 27, 67).
29 Women were slightly less likely than men to “strongly agree” that “I consider myself to be a leader,” but only by 5 percentage points (Princeton University 2011, 40).
30 We develop this point in chapter 3.
31 Professors, who were also surveyed by Princeton, responded that some of their most quiescent students are female. Shy women may compensate by going to office hours and doing their work more diligently (Princeton University 2011, 60).
32 Devitt 2002; Kahn and Goldenberg 1991; Kahn 1992, 1994a, 1994b, 1996; Heldman, Carroll, and Olson 2005.
33 Kahn 1992, 1994a, 1994b, 1996; Kahn and Goldenberg 1991.
34 Burrell 1994; Dolan 2004; Heldman, Carroll, and Olson 2005; Huddy and Terkildsen 1993a, 1993b; Kahn 1996; Norris 1997b; Sapiro 1983; Watson and Gordon 2003.
35 Murphy 1998, 167, quoting Wood 1994, 232.
36 Progress can be gleaned, with greater diversity of female characters and plot lines, but even when women are portrayed as powerful, they are also portrayed as conforming to feminine traits of beauty and sexuality (Roessner 2012, 331).
37 The study found that the effect of visible women does not work by elevating women’s endorsement of equal female participation in politics. However, that simply means that girls no longer adhere to the notions of explicit sexism (4% of girls endorse such views). It does not test our argument that girls are likely to view themselves as less influential when women are portrayed that way.