Women in Politics
The Gender Gap in Political Representation and Ambition in the United States
and the Netherlands
Graduate School of Humanities MA History - American Studies Iris Bos
11108932
Thesis supervisor: dr. M.S. Parry June 2017
“Don’t take yourself out of the race before you even start running.”
Table of Contents
CHAPTER I – INTRODUCTION 1
Political Representation & Ambition in the United States 1
The Dutch Case 3
Chapter Outline & Sources 4
CHAPTER II - THE UNITED STATES 8
A Short History of Women in American Politics 8
The Voter 9
Incumbency & Geography 10
Partisanship 11
Fundraising 12
The Candidate Emergence Phase 13
Political Knowledge & Participation 13
Recruitment 14
Competition 15
The Role Model Effect 15
Conclusion 16
CHAPTER III - POLITICAL AMBITION 18
Family Socialization 19
School Environment, Peer Associations & Media Habits 21
Competition 23
Encouragement 24
Self-Perceived Qualifications & Self-Confidence 24
Career Aspirations 26
Conclusion 26
CHAPTER IV – THE NETHERLANDS 29
A Short History of Women in Dutch Politics 29
Gender Equality in the Workplace 31
Women’s First Steps on the Labor Market 32
Education 33
Proportional Representation 34
Women’s Political Knowledge & Participation in the Netherlands 36
Dutch Adolescents 38
Conclusion 39
CHAPTER V – CONCLUSION 40
Chapter I – Introduction
Political Representation & Ambition in the United States
Feminism suffered a tough blow in the United States 2016 election. Not only did the defeat of Democratic presidential nominee and former U.S. Senator and Secretary of State Hillary Clinton mean that women have yet to shatter the highest political glass ceiling in the United States, the number of women in Congress and state legislatures failed to increase as well. The number of women elected to the U.S. Senate expanded from 20 (20%) to 21 (21%), while the gender gap in the U.S. House of Representative has grown as the number of women decreased from 85 (19.5%) to 83 (19.1%)1. On a state level, women managed to make only marginal gains. Women now make up 24.9 percent of state legislature, as opposed to 24.5 percent before the election – an increase of just 26 women (Center for American Women in Politics, 2016b; 2017a) . This minor increase of elected women is especially striking when placed in context. While the latter three decades of the twentieth century were characterized by a steady growth and later surge in female representation, the number of women in Congress has stayed relatively constant over the last election cycles. On March 1, 2017, 100 countries succeeded the United States in their percentages of women in national legislature (Inter-Parliamentary Union, May 26, 2017).
This gender gap in political representation is troublesome for several reasons. Paxton and Hughes (2007) summarize the arguments for women’s equal participation in politics into three types of justice: formal, descriptive and substantive representation. The first has already been achieved, as legal barriers to women’s entrance into the political sphere have been eliminated in 99 percent of countries worldwide. The U.S. Congress ratified the 19th
amendment to the Constitution granting women the right to vote in 1920 and Jeanine Rankin (R-MT) became the first women ever elected to Congress four years earlier (Center for American Women in Politics, 2016a). Women obtained the same rights as men and became equal for the law, eliminating direct discrimination against women in politics. However, formal representation did not and still does not lead to gender parity. Political theorists turned to the concept of descriptive representation, in which a descriptive similarity between representatives and constituents exists – a constituency of which 50 percent are women should have 50 percent female legislators. “Rights alone do not remedy the substantial social and economic inequalities that prevent women from taking advantage of their political
1 On June 20, 2017, Republican candidate Karen Handel won a special election in Georgia’s sixth district,
opportunities” (Paxton & Hughes, 2007, p. 9) . Arguments for this type of representation are based on the belief that groups are best suited to represent their own interests. However, gender parity in government does not necessarily signify that women’s interests are served. Substantive representation means that politicians speak for and act to these interests, but politicians have to be willing and able to do so.
Obtaining gender equality in political representation is not only just, but also useful. Women’s participation in politics changes deliberation for the better, as more attention is directed towards social and communal policies. Compared to their male counterparts, female legislators more often sponsor ‘women’s issue’ bills such as education, child-care, and family health legislation (Volden, Wiseman, & Wittmer, 2013) . Moreover, women manifest more leadership styles associated with effective performance as leaders (Eagly, 2007). For example, congresswomen secure 9 percent more spending from federal discretionary programs and sponsor and cosponsor significantly more bills than congressmen (Anzia & Berry, 2011) . However, other scholars find conflicting results. Analyzing current research and taking into account party status, Volden et al. (2013) found that exerting high effort, maintaining a policy focus, and engaging in consensus building helps minority party women achieve increased legislative effectiveness. Majority party women, however, seem to be less successful because of the highly partisan and polarizing strategies adopted by majority party men.
If gender equality in political representation is so important, why has progress been so slow? Scholars identified a myriad of explanations. Women’s chances of getting elected are no longer impeded by voter bias2, but are strongly influenced by incumbency, geography and partisanship. As these reasons do not account for the entire gender gap, scholars recognized the obstacles during the phase in which candidates come forward as a key argument for the lack of women in politics. They found that political knowledge greatly influences a person’s level of political participation and that women participate differently than men. Moreover, women are less likely to perceive themselves as qualified to seek office and are less likely to be actively recruited by political parties.
In order to increase the number of women in politics, enough women need to have the ambition to run for office. Political scientists Richard Fox and Jennifer Lawless analyzed political ambition in their Citizen Political Ambition Study and found a gender gap in political ambition among potential candidates. However, the potential candidates they
2 Significant gender disparities in vote totals have been eliminated (Anastaspoulos, 2016). Overall, women
surveyed, mostly men and women already successful in the ‘pipeline professions’, might just be less used to seeing women in office than are younger generations. In “Girls Just Wanna Not Run: The Gender Gap in Young Americans’ Political Ambition”, Lawless and Fox examined whether young women in high school and college have similar levels of political ambition as their male counterparts. They looked at several aspects of teenage- and adolescents’ life that might influence their desire to run for office in the future. Disappointingly, they found that the gender gap was already well in place by the time they graduated college.
This thesis will present a triptych about the lack of women in electoral politics in the United States and the Netherlands. In the first part of this thesis, I will analyze the current research on the origins of the gender gap in political representation in the United States. The second part of this thesis will focus on political ambition as a fundamental influence upholding the gender gap in electoral politics. I will analyze the sources used in and conclusions drawn by Fox and Lawless in their leading study on the gender gap in political ambition among college students. The factors that influence female representation, found in the first and second chapter, will be used to analyze the current research on the gender gap in political representation in the Netherlands.
The Dutch Case
Even though The Netherlands scores significantly better on the Inter-Parliamentary Union’s World Classification than the United States and is currently ranked 26th, the latest Dutch election for the House of Representatives on March 15, 2017 can be interpreted as a blow for gender equality as women lost three seats – now occupying 54 (36%) of the 150 seats. The inability to make any progress also resulted in a Dutch ‘plateau’. Since 2007, female representation in the Dutch Senate has remained constant at 34,7 percent – holding 26 out of 75 seats. Large gender disparities are also evident at the provincial and state level, as only 198 (34.7%) members of the Provincial-States and 28 percent of city council members are women (Atria, 2016; Prodemos, 2015).
However, another gender gap is studied more intensely and frequently. Scholars and government policies are more often focused on the gender gap in employment, with 73 percent of employed women working part-time. In no other EU-member country do women work fewer hours. This has important implications for the lack of women in top-level positions, as women who work part-time are less likely to advance into these positions (Portegijs & van den Brakel, 2016) . Paradoxically, Dutch girls and young women have been
outperforming their male counterparts in education for several years. This advancement is essential when looking to close the gender gap, as education is a good predictor of political participation (Bovens & Wille, 2016) . However, young women have thus not yet managed to translate that head start into better positions in the labor market and politics.
The third part of the triptych will study the current research on women’s political knowledge, interest and participation in the Netherlands. Even though the Dutch are acclaimed and praise themselves to be an extremely progressive country3, the facts in this particular area indicate otherwise. As American scholars have already extensively researched the origins of the gender gap in political representation, the analysis from the first two chapters will be used to study the reasons underlying the lack of women in electoral politics.
Chapter Outline & Sources
The interdisciplinary field of women and gender studies has developed greatly in the United States over the last fifty years. Moreover, political scientists, sociologists and historians have focused more intensely on the issues of gender and the position of women in their respective fields. The second chapter of this thesis will start by presenting a brief overview of the history and development of women in electoral politics in the United States. Research from the Center for American Women and Politics (CAWP), a unit of the Eagleton Institute of Politics at Rutgers University will be used as the key source for the history of women’s political participation, on a local, state and national level. Subsequently, data from the Pew Research Center will be used to complement the short historical overview with statistics of public opinion to assess whether the advancements made with gender equality run parallel to Americans’ attitudes towards women in politics.
The third section will study various explanations for the gender gap in American electoral politics. Women have already managed to level the playing field in one critical area, as significant gender disparities in vote totals have been eliminated. Several scholars have demonstrated that women’s overall chances of getting elected are not longer impeded by voter bias (Anastasopoulos, 2016), but are strongly influenced by incumbency (Palmer & Simon, 2005) geography (Ondercin & Welch, 2009) and partisanship (Dolan, 2008; Fox & Lawless, 2010; Sanbonmatsu, 2002) . Following up on the perspective of the voter, the influence of gender stereotyping on the public willingness to vote for female candidates will
3 The Netherlands is currently ranked 7th on the Social Progress Index, which measures the extent to which
countries provide for the social and environmental needs of their citizens. They adopt 54 indictors in the areas of basic human needs, foundations of well-being, and opportunity.
then be discussed (Dolan & Lynch, 2016) . Subsequently, fundraising will be discussed as one of the most crucial components of a viable candidacy. In general, women raise the same amounts of money as their male opponents (Burrell, 1985; Hogan, 2007; Kitchens & Swers, 2016; Uhlaner & Schlozman, 1986) . However, female candidates spend more time on this task and view it more negatively than do men (Jenkins, 2007). Incumbency, (Fouirnaies & Hall, 2014; Kitchens & Swers, 2016) and a candidate’s party identification (Francia, 2001; Kitchens & Swers, 2016; Thomsen & Swers, 2017) appear to influence a female candidate’s fundraising efforts and effects.
Even though the above-mentioned factors all influence a female candidate’s chances of getting elected, they do not account for the entire gender gap in political representation. Scholars turned to obstacles to women’s numerical representation prior to having made the decision to run for office and identified the ‘candidate emergence phase’ as the most substantial impediment. The ‘candidate emergence phase’ encompasses several concepts and factors. This section will first focus on political knowledge and participation. Political knowledge is an important factor to look at when studying the gender gap, as it effects a person’s political participation (Delli Carpini & Keeter, 1993) . People with higher levels of political knowledge are more likely to participate in politics (Burns, Schlozman, & Verba, 2001) ; this effect is even stronger for women than men (Ondercin & Jones-White, 2011) . Moreover, women and men participate differently in politics. Women tend to participate more in personal acts of engagement, while men engage more in collective and public action (Coffé & Bolzendahl, 2010) .
Other factors that appear to affect the emergence of female candidates, which will be discussed in the third chapter, are disparities in candidate recruitment, the competitive nature of elections and the role model effect. Fox and Lawless find that women are less likely to be recruited to run for office than men (2004; 2010). Research also shows that women are more averse to the strategic and competitive nature of elections (Kanthak & Woon, 2015) . Moreover, women’s political engagement increases when female candidates are visible and competitive (Campbell & Wolbrecht, 2006) . These factors will be elaborated on in the second chapter.
The third chapter will study the gender gap in political ambition as one of the most significant factors of the ‘candidate emergence phase’ by critically analyzing “Girls Just Wanna Not Run: The Gender Gap in Young Americans’ Political Ambition” (2014; 2013) by Fox and Lawless, the leading study of political ambition among adolescent Americans. In their study, Fox and Lawless discuss several elements that together influence young men and
women’s political ambition. This chapter will look at these separate elements, analyze the relevance and applicability of the cited sources per element, and evaluate the methods adopted and conclusions drawn by Fox and Lawless in their study. The elements constituting political ambition named in the study relate to adolescents’ career aspirations, family socialization, a political context in the classroom and media, the competitive nature of extracurricular activities, and self-perceived qualifications.
The fourth chapter of this thesis will focus on the gender gap in political representation and ambition in the Netherlands. Similar to the second chapter, this section will start by presenting an overview of the history and development of women in electoral politics in the Netherlands. Data from Parlement & Politiek, an initiative from the Parlementair Documentatie Centrum at Leiden University and studies by political scientist Monique Leyenaar will be used as key sources for the history of women’s political participation on a national level. Data on provincial and local level offices will be collected from Atria, institute on gender equality and women’s history and ProDemos. Subsequently, data from the Netherlands Institute for Social Research (Sociaal en Cultureel Planbureau) will be used to complement the short historical overview with statistics of public opinion to assess whether the advancements made with gender equality run parallel to attitudes towards women in politics in the Netherlands.
The next section will study the discussion and findings on women’s participation in the workplace and its implications for obtaining higher and political office. Primary source for this paragraph will be the Emancipation Monitor (2016), which is commissioned by the Directorate of Emancipation of the Ministry of Education, Culture and Science every two years. The project is carried out by the Netherlands Institute for Social Research and Statistics Netherlands (Centraal Bureau voor de Statistiek). These findings will introduce strong views of gender role socialization.
Subsequently, the third section will focus on gender gaps in secondary and higher education in the Netherlands. Research findings commissioned by the Ministry of Education, Culture and Science will be used to discuss the gender gap in study success and will highlight the inability to translate the head start of young women in education into better positions in the labor market and politics (Portegijs & van den Brakel, 2016) . Additionally, the importance of higher education will be studied by examining the composition of the Dutch House of Representatives and other levels of political office (Bovens & Wille, 2016) .
As most research on the reasons for women’s underrepresentation in politics has been conducted in the United States, it is important to consider the different electoral system in the
Netherlands. The United States uses a single-member-district system to elect its representatives, which, as the name suggests, elect only one politician per district. The Netherlands uses a system of proportional representation, in which parties gain seats in proportion to the number of votes cast for them. On average, countries using a system of proportional representation elect substantially more women to their national parliaments than single-member-district systems (Darcy, Welch, & Clark, 1985) .
The fifth section will study political knowledge, interest and participation in the Netherlands. Political knowledge and interest are essential factors to look at when studying gender disparities in electoral office, as they effect a person’s political participation. In the absence of Dutch research on these concepts, data and findings from international research will be used. Not only do Dutch women know less about politics than men, they are also less interested in politics and less likely to have engaged in at least one political or campaign activity (Fraile, 2014; Fraile & Gomez, 2017; Karp & Banducci, 2008) . Data from the World Values Study (2014) and Statistics Netherlands will be used to complement their findings.
The last section will focus on the political knowledge, interest and participation among young men and women in the Netherlands. For lack of scholarship on adolescents’ political behavior, I will only use the Emancipation Monitor and Statistics Netherlands’ Youth Monitor.
Chapter II – The United States
A Short History of Women in American Politics
The Seneca Falls Convention is often considered to be the beginning of the women’s rights movement in the United States. This meeting, convened by Elizabeth Cady Stanton and others active in the anti-slavery movement in 1848, produced a Declaration of Sentiments demanding several rights for women. Even though the resolution was heavily argued during the convention, women’s suffrage was eventually included in the document. Eighteen years later, Stanton became the first woman to run for president. The event serves as the first milestone for women in American politics. Jeannette Rankin (R-MT) became the first woman elected to Congress in 1916. Only four years later, 72 years after the Seneca Falls Convention, the 19th Amendment to the Constitution was ratified by Congress granting women in all states the right to vote. The latest addition to such milestones took place last year, when former Senator and Secretary of State Hillary Clinton became the first woman to be a major party’s presumptive presidential nominee (Center for American Women in Politics, 2016a). The 96 years between the two events are characterized by enormous progress in women’s political participation, both as voters and candidates. Although women have had the right to vote for less than a century, since the early 1980’s they have exercised it in greater percentages than men. In the 2012 presidential election, 63.7 percent of the female eligible adult population reported voting. This exceeds the proportion of eligible male adult voters by 3.9 percent. This gender gap is also evident in midterm elections, as the gender gap in voter turnout in the 2014 election was 2.2 percent. The absolute number of female voters has outpaced the number of male voters since the early 1960’s (Center for American Women in Politics, 2015). In the early 1980’s, another gender gap became apparent. Women are more likely to identify themselves as Democrat, while men are more likely to support the Republican Party. Women also view the job performance of Democratic Presidents more favorably than men (Center for American Women in Politics, 2014).
The progress women have made as voters has not yet translated into advances of women on the ballot. The rate of female representation has stagnated during the last decade. The 1980’s saw a surge of women elected to Congress and the number of Congresswomen more than doubled from 30 in 1992 – the ‘Year of the Women’ – to 62 in 2012. As a result of the 2016 election, the number of women elected to the U.S. Senate expanded from 20 (20%) to 21 (21%), while the gender gap in the U.S. House of Representative has grown as the number of women decreased from 85 (19.5%) to 83 (19.1%). On a state level, women
managed to make only marginal gains. Women are now in possession of 1840 (24.9%) state legislative seats, as opposed to 1814 (24,5%) before the election. Moreover, women hold 24.0 percent of statewide elective executive office. As of February 2017, twenty of the 100 biggest cities have a female mayor. These women are not equally divided by party; Democratic seats are more often filled by women. This holds true for every level of office, except statewide elective executive office. Democratic women have 75 percent of the female seats in Congress, 60.4 percent of state legislature and 42.7 percent of statewide executive office (Center for American Women in Politics, 2016b; 2017a) .
Perhaps one reason for the underrepresentation of women in American politics is that the American public does not (yet) value women to be equally qualified as men, although the Pew Research Center (2015) finds that women are perceived to be as competent as men by most Americans. Additionally, women are even viewed as being more compassionate and organized. In their survey, Pew asked respondents what they thought were significant reasons for the lack of women in high political offices. About four-in-ten Americans think women are held to a higher standard than men and that the electorate is not ready to elect more female leaders. Even though the progress towards gender parity is slow, 73 percent of American believes they will have a female president in their lifetime. When asked about leadership characteristics, the majority of Americans find women identical to men on features typically attributed to leadership, namely intelligence and capacity for innovation. Even though most do not distinguish between the sexes, some say that women excel at working out compromises and being honest and ethical. Almost 40 percent of American women say that having more women in top leadership positions would do a lot to improve all women’s lives, compared to only 19 percent of men. Gallup (June 22, 2015) found that 92.3 percent of American said they would be willing to vote for a female presidential nominee. These findings appear hopeful, though these attitudes have not yet translated in gender-balanced representative bodies.
The Voter
‘When women run, women win’ is an often-heard phrase when discussing women’s involvement in electoral politics. Women have managed to level the playing field at the ballot box, as significant gender disparities in vote totals have been eliminated. Voters no longer discriminate against female candidates as female candidates receive similar amounts of votes as their male counterparts (Burrell, Carroll, & Sanbonmatsu, 2014; Seltzer, Newman, & Leighton, 1997; Smith & Fox, 2001) . For example, current research by Anastasopoulos
(2016) found no evidence of a gender penalty in elections for the House of Representatives between 1982 and 2012.
Following up on the perspective of the voter, gender stereotyping influences the public willingness to vote for female candidates. Political scientist Kathleen Dolan has extensively researched and published on the role of people’s evaluations of women’s abilities in shaping their support. She argues that the assumption of the influence of gender stereotyping on women’s electoral chances seems obvious, but recent elections do not offer empirical support. She argues that the influence of party stereotypes appears stronger than that of gender stereotypes (2014). In their most recent article looking at the effect by level and type of office, Dolan and Lynch (2016) again conclude that gender stereotypes are not a strong or consistent influence on vote choice. They find that gender stereotypes have distinct effects on vote choice decisions for female Republican and Democratic candidates. Additionally, they conclude that gender stereotypes are more often used in voters’ evaluation of female candidates in congressional than in gubernatorial races. Contrary to what earlier research in the 1990s argued, when scholars contended that voters attribute male traits to executive office more than to legislative position, Dolan and Lynch find no relationship between male trait stereotypes and executive office. They find that analysis from real-world elections, instead of laboratory or public opinion studies, “suggests a disconnect between what people might say they value in abstract situations and what they actually employ in making real voting decisions (2016, p. 590).’ However, earlier research by Dolan (2010, p. 85) did find that, while stereotypes apparently have little effect on the lack of women in politics, views on stereotypically male issues are important. ‘People who see women as competent to deal with things like the economy and terrorism are dramatically more likely to voice a willingness to support them for office and a desire for greater gender balance in government.’
Incumbency & Geography
However, scholars found that women candidates’ chances of getting elected are strongly influenced by incumbency, mainly enjoyed by men. Seltzer et al. (1997, p. 79) state that ‘winning elections has nothing to do with the sex of the candidate and everything to do with incumbency’. In elections for the House, female incumbents win at similar rates as their male counterparts, but face more competition (Palmer & Simon, 2005) . The ideal situations for female candidates are open-seat races. As support from female voters is stronger in these races, women have a greater advantage (Ondercin & Welch, 2009; Smith & Fox, 2001) .
Another explanation is given by Lawless and Pearson (2008), who identify the congressional primary process as another explanation for women’s underrepresentation. They find similar results in primaries as Palmer and Simon did in general elections. Female candidates have similar chances of winning primaries as men, but face more competition than their male counterparts. As Lawless and Pearson (2008, p. 67) comment, ‘gender-neutral victory rates, then, are not the result of a gender-neutral primary process. Women have to be “better” than their male counterparts in order to fare equally well’.
Not every district offers female candidates similar environments and chances to men prior to, and on, Election Day. Ondercin and Welch (2009) extensively studied congressional races between 1992 and 2000. They found that electoral circumstance and the women currently in elected office influence the emergence and success of female candidates. In addition to the benefits of open-seat races, they conclude that Democratic-leaning districts see more women running and winning primaries and elections. They also looked at the “women-friendliness”4 of the district and conclude that it only has a significant impact in open-seat races. Another influence is the electoral nature of districts, as multi-member districts tend to deliver a more balanced representation of the community. More women run in and are elected by multi-member districts than by single-member districts5 (Darcy et al.,
1985; Sanbonmatsu, 2002). Religion also influences women’s candidacies. Seltzer (2016) found that women are more likely to run for office in less religious districts. District in highly religious communities are about fifty percent less likely to elect female candidates, compared to less religious but otherwise comparable districts.
Partisanship
Partisanship also influences the gender gap in political representation. The number of elected women is not equally divided by party. Democratic women have 75 percent of the female seats in Congress, 60.4 percent of state legislature and 42.7 percent of statewide executive office (Center for American Women in Politics, 2016b; 2017a) . First, the Pew Research Center (2016) states that women also identify more often with or lean toward the Democratic
4 Palmer and Simon (2006) argue female candidates—Democratic and Republican—are more likely to win in
districts that are urban, non-Southern, Democratic, geographically small, racially or ethnically diverse, affluent, highly educated, and middle class. They refer to districts that fit this profile as “womenfriendly districts” because they contain the political, demographic, and geographic characteristics common to districts where female candidates have been elected.
5 Single-member district are electoral districts that send only member to a legislative chamber, multi-member
districts send two or more. Ten American states have at least one legislative chamber with multi-member districts. Multi-member electoral systems are more often referred to as systems of proportional representation.
party (54%) compared to men (42%). Scholars offer several additional explanations. Lawless and Fox (2010) find that Republican and Democratic women are not recruited to run for office in similar rates, as Democratic women are 7 percent more likely to receive encouragement from elected officials and 12 percent from local political activists. They also conclude that Republican women are not equally represented in the ‘pipeline professions’. Political scientist Kira Sanbonmatsu supports this finding and states that the social eligibility pool, legislative professionalism and partisan composition of the legislature affect women’s representation in the lower houses of state legislature differently by party (2002). Moreover, women’s organizations, of which EMILY’s List is the most influential, focus to a greater extent on the recruitment and financial support of pro-choice, Democratic women. Research from Political Parity (2015), a program from Hunt Alternatives, found that the primary is the highest hurdle for Republican women. Republican women have greater difficulty raising money and receiving similar amounts of attention by their party as their democratic counterparts. Moreover, women are still perceived to hold more moderate views than men. As the most conservative candidate usually wins Republican primaries, Republican women have to combat the perception that they are less conservative than their male counterparts. As a result, female Republican candidates are less likely to enter or win a primary than female Democrats. Those who do often lack coaching and support and do not benefit from a ‘gender advantage’, such as specific financial or organizational support from women’s organizations or a female voter’s preference to vote for a woman. Voters also appear to have some influence. Dolan (2008) concludes that women voters feel more positively towards female Democratic candidates than do men; they are drawn by both political party and candidate sex. The effect does not apply to female Republican women, as they might experience conflicting pressures. The presidential election of 2016 serves as a good example, where Republican female voters had to choose between electing a Democratic woman or a Republican man.
Fundraising
In order to win an election, candidates increasingly need more money to run a successful campaign. In general, women raise the same amounts of money as their male opponents (Burrell, 1985; Uhlaner & Schlozman, 1986) , on state and national levels (Hogan, 2007), and in primaries as well as general elections (Kitchens & Swers, 2016) . This would imply that female candidates’ campaigns are not financially disadvantaged. However, female candidates have to spend more time on fundraising efforts to raise similar amounts of money and view this task more negatively than do men (Jenkins, 2007). This thus suggests that fundraising is
another barrier for women to decide to run for office. If that decision has been made, two factors influence a candidate’s fundraising efforts and effects. Incumbency and partisanship are not only the two most significant influences on women candidates’ chances of getting elected; they also greatly impact a candidate’s fundraising. Fouirnaises and Hall (2014) find that incumbents in the state legislatures and the House of Representatives receive 25 percent more in donations than their challengers. This is detrimental to women’s chances, as incumbents are exceedingly more often male than female. Secondly, Kitchens and Swers (2016) found that in the elections for the House of Representatives in 2010 en 2012, contrary to female Republicans, Democratic women outperformed the fundraising totals of their male counterparts in their primaries. This is not the only advantage female Democratic candidates enjoy. Thomsen and Swers (2017) use data from the same races and conclude that female Democratic donors value the election of liberal Democratic women, who benefit from this ‘gender advantage’. However, female Republican donors promote ideology over gender and prefer conservative candidates. As mentioned earlier, Republican women have to combat the perception that they are less conservative than their male counterparts. This thus not only has consequences for vote totals, but also for money received from donors from their own party.
The Candidate Emergence Phase
Even though incumbency, district characteristics, partisanship and fundraising all influence the underrepresentation of women in politics, they do not account for the entire gender gap. Scholars identified another obstacle to women’s political participation prior to the campaign and named the ‘candidate emergence phase’ as a crucial impediment to political gender parity. This phase is not strictly demarcated in current research and encompasses several different concepts and factors. I discuss those that are relevant to the subject of gender disparities in electoral politics below.
Political Knowledge and Participation
Firstly, political knowledge and participation are considered to be central parts of an active democracy. Therefore, gender disparities in political knowledge and participation are not only features of unequal participation in citizenship; they will also reinforce the current gender gap in political representation. Political knowledge is an essential factor to look at when studying the gender gap, as it effects a person’s political participation (Delli Carpini & Keeter, 1993) . People with higher levels of political knowledge are more inclined to engage in political activities (Burns et al., 2001). Women are found to already have less political
knowledge than men (Ondercin & Jones-White, 2011) and moreover, women are particularly disadvantaged, as this effect is stronger for them than for men. This finding has serious consequences for political discussion, as people turn to those they regard as knowledgeable to collect political information. Both men and women perceive women to have less political knowledge than men (Morehouse Mendez & Osborn, 2010) . However, serious controversy exists about the content and structure used to measure political knowledge and the ways in which it influences the gender gap. Tourangeau, Maitland and Yan (2016) found that encouraging people to use a ‘don’t know’ option instead of guessing improved the question’s the reliability and validity. This is important, as women more often than men use this option. Dolan (2011) focused on the content used to measure political knowledge and concluded that the often observed gender gap dissolves when respondents are asked about the levels of women’s representation in Congress. As mentioned earlier, when women participate in politics they engage in different activities. Women tend to participate more in personal acts of engagement, while men engage more in collective and public action (Coffé & Bolzendahl, 2010) . This could be troublesome, as these forms of public action – party membership, collective activism and political contact – create political networks and might better prepare a person to eventually run for office.
Recruitment
Even though the practice of candidate recruitment by political parties has not always been an open process, research from the last two decades has shown that a persistent gender gap exists. From research conducted in their Citizen Political Ambition Study, Fox and Lawless (2004; 2010) conclude that only 32 percent of women, compared to 43 percent of men, from the candidate eligibility pool6 have received encouragement to run for office from an elected official, party leader or political activist. This is detrimental, as people are more likely to consider running for office when they are recruited. The likelihood a woman considers running for office increases with 37% if she is recruited. The effect of encouragement to run increases as the number of recruitment contacts goes up (Moncrief, Squire, & Jewell, 2001) . As previously mentioned, Republican and Democratic women are not recruited in similar rates. Democratic women are 23 percent more likely to receive encouragement from political activists (Fox & Lawless, 2010) .
6 The candidate eligibility pool from the Citizen Political Ambition Study is comprised of men and women from
the four professions that are most likely to yield political candidacies for both state legislative and congressional offices in the United States: laws, business, education and politics.
Competition
Another factor that keeps women from stepping into the political arena is the nature of the game. Research purportedly shows that women are more averse than men to the strategic and competitive nature of elections. In their study, Kathak en Woon (2015) find that men and women are equally inclined to serve as representatives of their groups. The gender gap appears only when selection involves an election, as women’s willingness decreases substantially. Moreover, men perform better in competitive environments than women (Gneezy, Niederle, & Rustichini, 2003) . If they have a choice, men are more likely than women to opt for competitive situations7 (Niederle & Vesterlund, 2007) . Additionally, Gneezy and Croson (2009) conclude that women are more averse to risk than men. They assume that men assess their chances in risky situations differently than women, as men are more confident about their own qualifications than are women. Since participating in an election could be perceived as a high-risk activity, men might be more prone to participating than women.
The Role Model Effect
Lastly, scholars point to role model theory as another explanation for the gender gap in political representation. They assert that the presence of women in politics can encourage women to express higher levels of political participation. Burns, Schlozman and Verba (2001) found that living in a state with a statewide female politician has a positive influence on psychological political engagement, such as political information, interest and efficacy. However, the mere presence of women in political office does not appear to be enough. Campbell and Wolbecht (2006, p. 235) summarize multiple research findings and conclude that ‘the degree to which the candidacy is visible […] creates a context in which women’s presence as politicians can affect the political engagement of women’ and refer to three conditions. First, several scholars draw conclusions from research conducted in 1992. During the ‘Year of the Woman’, gender issues were notable and female candidates were unusually competitive. That year, the presence of female politicians was connected to higher levels of political involvement, efficacy and media use by women and men (Burns et al., 2001; Hansen, 1997; Koch, 1997). Secondly, the level of office also affects the strength of the role-model effect. For example, Burns and her colleagues (2001) focus on candidates for Congress
7 Participants in a laboratory experiment had to solve tasks under noncompetitive piece rate and a competitive
and Mariani et al. (2015) focus on Hillary Clinton, Nancy Pelosi and Sarah Palin as highly visible candidates. Third, the viability of the office appears to be an important part of this function. Atkeson (2003) concludes that female voters engage more when they observe a competitive race. Token symbolic representation does not enhance political engagement, as female politicians have to be viable and competitive in order to have a positive effect. Reingold and Harrell (2010) add another element to this function and argue that party mediates the role-model effect.
Conclusion
In conclusion, not just one reason can be given for the gender gap in political representation. After analyzing the most-often cited impediments to gender parity, I would argue that the current electoral environment is not the fundamental reason for women’s underrepresentation in American politics. Voters and party donors are no longer overtly biased against female candidates, but subtle hurdles still exist for some women in certain races. For example, female Republican candidates are less likely to receive early fundraising and support from women’s organizations than their Democratic counterparts. Moreover, women less often enjoy the benefits of incumbency, as men maintain the majority of local, state and national offices.
These environmental hurdles interact with other obstacles perceived by women prior to having made the decision the run for office. Fox and Lawless (2004, p. 275) articulate it accurately when they say that ‘even though women who run for office are just as likely as men to emerge victorious, the substantial winnowing process in candidate emergence yields a smaller ratio of women than men. The pool of candidates who run for office, therefore, looks quite different than the eligibility pool of potential candidates with whom we began. This finding reveals the danger of honing in on electoral performance as a gauge for gender neutrality.’ This evident conclusion is supported by the numerous other studies I cite in the second part of this chapter. However, this does not mean that the gender gap can now easily be resolved. If there is one thing that the current research shows, is that many variables influence a woman’s decision to run for office and affect her electoral chances.
As many studies focus on only one piece of the puzzle, it is hard to construct a comprehensive analysis. Moreover, much of the research is flawed. Not all research findings are generalizable across all levels of office and election years. And while some studies draw conclusions from laboratory settings, others base their findings on dated election results. But most important, ‘the female candidate’ does not exist. Party affiliation, race, education and
religion all influence a woman’s political socialization and the possibility she will one day decide to run for political office. Only a small minority of studies takes into account these demographic variables and discusses their implications.
By collecting the findings from the various studies cited in this chapter, I am able to create a preliminary analysis of the factors hindering women’s representation. Consequently, scholars’ focus on adolescents’ political socialization becomes evident. Even though further research is needed to analyze the various factors in greater detail, the search for the origin of the gap becomes more substantiated and structured. In the second part of this triptych, I will critically analyze “Girls Just Wanna Not Run: The Gender Gap in Young Americans’ Political Ambition” by Fox and Lawless as a case study, to consider the issues in the research more broadly by examining the leading study on political ambition as the principle impediment the gender gap in political representation in detail.
Chapter III – Political Ambition
In their research, Fox and Lawless often refer to the eligibility pool of potential candidates and identify the ‘candidate emergence phase’ as a crucial impediment to gender equality in political representation. In their Citizen Political Ambition Studies, they found that similarly situated men and women do not show equal levels of political aspiration. As illustrated in the previous chapter, women have lower levels of political knowledge and are less likely to engage in the political process. Moreover, women are discouraged by the competitive nature of the electoral process. Also, they more often perceive their own qualifications as not sufficient to run for political office and as such are less likely to consider declaring a candidacy. Party officials and politicians more often encourage men to run for office.
These studies show that the gender gap in political ambition is already well in place by the time women reach ‘pipeline professions’, occupations often held prior to running for political office. Fox and Lawless concluded that they still encounter many limitations in the conclusions they can draw about the roots of the gap. As such, they turned their focus towards adolescence, a crucial phase in which attitudes towards political and democratic responsibilities are formed, gender role socialization plays an important role and young men and women are developing and working towards their career aspirations.
In 2012, Fox and Lawless surveyed 1818 high school and 2117 college students to determine if young women already showed lower levels of political ambition than their male counterparts. In this chapter, the report “Girls Just Wanna Not Run: The Gender Gap in Young Americans’ Political Ambition” (2013) will be reviewed as the leading study on the origins of the gender gap in political ambition. One year after the publication of the report, Fox and Lawless published the journal article “Uncovering the Origins of the Gender Gap in Political Ambition” (2014) to elaborate on the theoretical framework, research design and conclusions of their study. This chapter will look at both the report and article, but will focus on the results gathered from respondents in college, as the gender differences are more evident among college than high school students. Results from respondents in high school presented only one significant gender gap. The five factors that, according to Fox and Lawless, contribute to the gap among college studies will be discussed by reviewing the survey questions, their foundation in previous scholarship and the conclusions drawn from the gathered data.
Family Socialization
The first of five factors that Fox and Lawless identify is family socialization. They argue that the development of political attitudes and behavior is greatly influenced by the political socialization a person receives at home during their childhood years. They look at the presence of politics in the household when growing up and the engagement in political activities with parents to evaluate politicized home environment. Fox and Lawless find a strong foundation for the inclusion of family socialization in previous research. The theoretical framework on family socialization focuses on three concepts: party affiliation, political attitudes, and political knowledge and participation.
The first concept Fox and Lawless refer to argues that political party affiliation is often handed down from parent to child. This specific transfer has been the subject of many political and sociological studies over the last forty years. As such, three out of the four sources Fox and Lawless cite date back to the late 1970’s and early 1980’s. The earliest cited source finds a strong relationship between parental political orientation and children’s responses (Acock & Bengtson, 1978) . Niemi, Ross and Alexander (1978) focused on the intergenerational continuity of political issues, and only found a strong correlation between parents and their children for partisanship. The most recently published study cited uses longitudinal data tracking three generations. The study confirms the conclusions drawn from previous research and concludes that the effect of political socialization is stronger in situations in which the family is highly politicized and children receive regular clues (Jennings, Stoker, & Bowers, 2009). Surprisingly, Fox and Lawless only asked the respondents about their own party affiliation and did not include a question about the respondents’ parents’ affiliation. The literature was thus used to point to the effect of parental political socialization in a broader sense. The questions do ask for regular cues by including the word ‘often’ in three questions.
The second concept of the section on family socialization argues that attitudes about good citizenship, political activism and interest originate partially in family upbringing. The oldest study Fox and Lawless refer to already showed that political behavior is strongly influenced by parents’ socioeconomic status, political activity and civic orientations (Beck & Jennings, 1982) . Other scholars contributed to and elaborated on this early finding. In their longitudinal study, Jennings et al. (2009) also conclude that the effect of parental socialization is influenced by early acquisition. Verba, Schlozman and Burns (2005) also focus on the role of parents’ socioeconomic status and found two mechanisms by which high-educated parents generate politically active children. First, they conclude that because
well-educated parents more often take part in politics, they are more likely to have political discussion at home. Consequently, their children assimilate these practices and develop orientations that predispose them to take part in politics themselves. Secondly, they point to the fact that well-educated parents are reasoned to have well-educated children. An important notion, as educational attainment is the most dominant predictor of future political behavior.. The articles selected by Fox and Lawless are mostly focused on the effect of parents’ socioeconomic status. In its section on sample demographics, this study asks about estimated household income, but in their discussion of the data, Fox and Lawless do not go into any detail about the effects of socioeconomic status relating to gender. Moreover, no questions on parental educational attainment were included. The first and second concept lay the theoretical foundation for the four questions on the presence of politics in the household growing up, which focus on the level of politics children encounter at home when growing up.
Where parental party affiliation and civic orientations influence adolescents’ political behavior indirectly, the third part of family socialization deals with active interactions between parents and their children. Scholars found that political discussion between parents and adolescents positively influences adolescents’ political knowledge and participation (Andolina, Jenkins, Zukin, & Keeter, 2003; McIntosh, Hart, & Youniss, 2007) . Fox and Lawless give much weight to this finding, as they included multiple questions on discussions with parents on political subjects.
In conclusion, the ten questions asked by Fox and Lawless about political participation at home form a complete illustration of the concept socialization by the family. All questions, with the exception of the fourth inquiring about yelling at the TV, seem to result logically from the cited sources. Surprisingly, parental party affiliation and an analysis of the gender differences by socioeconomic status were absent. Also, the cited scholarship only included ‘private’ forms of political participation, such as discussion and watching TV. Public forms of participation, such as voting or attending political events as a family, were not included.
Only four questions resulted in a statistically significant gender gap, of which three work to women’s advantage: women were 6 percentage points more likely to have followed the 2012 election with their parents, 5 percentage points more likely to have discussed same-sex marriage with their parents and 6 percentage points more likely to have shared news through social media. The fourth significant gap works to the advantage of young men, as 24 percent of male respondents indicated to talk about politics at mealtime, compared to only 19
percent of female respondents. In conclusion, young women are generally as likely as their male counterparts to grow in in politicized households. Even though they found some notable statistically significant gender gap, the gender gap as defined by Fox and Lawless does not originate mainly in these concepts of family socialization.
School Environment, Peer Associations and Media Habits
The second of five factors identified by Fox and Lawless shows that adolescent women are less likely than their male counterparts to be exposed to politics in the classroom, during extracurricular activities, with friends and the media. Similar to the first factor, survey questions result logically from the cited scholarship.
The first area Fox and Lawless point to is exposure to politics in classrooms. The first survey question inquiring about political science or government classes finds it foundation in two studies. Using data from a longitudinal study, Hilligys (2005) concluded that social science curriculums are positively related to future political engagement. Moreover, one study on a supplementary civics program in Philadelphia found that these programs could strengthen future political participation through increased political efficacy (Pasek, Feldman, Romer, & Jamieson, 2008) . Results from this first question show a significant gender gap, as male college students are 6 percent more likely to have taken a political science or government class. Fox and Lawless also cite Campbell (2008), who found that an open classroom environment in which students feel free to express their opinions, openly disagree with their teachers and are encouraged to discuss political and social issues, increases high school students’ levels of political knowledge. Fox and Lawless translated this finding into their second questions, which inquires about classroom discussions about politics and current events. As Campbell (2008) indicates, the nature of the discussion could prove more influential than the frequency of formal social studies instruction. However, the second question still inquires about the frequency instead of the nature of the discussions. Another significant gender gap becomes apparent, as female college students are 5 percentage points less likely to discuss politics and current events in class than male students.
Secondly, Fox and Lawless included questions about the political nature of extra-curricular activities, as Glanville (1999) already found that instrumental extraextra-curricular activities have a positive effect on future political participation. These types of activities include student government, student newspaper, yearbook, debate club and political clubs. Students involved in other activities, such as academic clubs, cheerleading, band and athletics, did not express higher levels of political participation than those who did not
participate in any activity at all. Volunteering for politicized groups and club is also found to positively influence future political involvement (McFarland & Thomas, 2006) . Involvement is high school community service is found to predict adult voting and volunteering too (Hart, Donnelly, Youniss, & Atkins, 2007) . This finding was used for the questions for high school respondents. The three questions Fox and Lawless included for college students, all result logically from the above mention studies. However, student government is included as the only extracurricular activity in the survey. As Glanville argues, other institutional extracurricular activities also predict later political involvement. Even though student government might appear most relevant in this case, no study has found student government to be a better predictor of future political or electoral participation than other instrument extracurricular activities. Fox and Lawless find statistically significant gender gaps, as young women are 7 percentage points less likely to have participated in College Democrats or Republics, 4 percentage points to have run for and 3 percentage points to have held a student government position.
College students do not spend all of their on campus in classrooms or the library. Fox and Lawless do not cite separate sources on the effects of socialization with friends on future political involvement. However, data on the questions on discussions about politics or current events with friends show a significant gender gap. Female college students are 7 percentage points less likely to talk about politics with friends, and 6 percentage points less likely to discuss current events. As no scholarly foundation is provided, we cannot draw any conclusion from the data regarding future political participation. This finding contributes to the notion that young women’s environments, both in and outside of school, are less politicized.
Lastly, the survey asks respondents about the sources they use to obtain political information. Fox and Lawless find that young men and women are equally inclined to watch cable news or read a hard copy newspaper. However, female colleges students are 12 percentage points less likely to visit news websites, 13 percentage points less likely visit political websites or blogs and 10 percentage points less likely to watch The Daily Show with
Jon Stewart and/or The Colbert Report. Surprisingly, the theoretical framework for this
section is comprised of only three sources published in 2003 and 2004. As the media, especially the online media has developed rapidly between 2003 and the publication of this study, the selection of media sources seems rather outdated. Even though social media was not used as regularly as today, they should have been included in the survey. Moreover, the three sources cited by Fox and Lawless are experimental and/or inadequate. The cited pilot
study by Iyengar and Jackman (2003) was conducted during a single campaign in California and its low response rate of 22 resulted in a small sample size of 152. Its conclusions about the use of CD’s to boost voter turnout also seems dated. The study by Fuller (2004) was used to point to a gender gap in internet habits regarding news and political information. However, Fuller concluded that few gender differences exist in online civic participation; social status was found to better explain civic participation than gender. In conclusion, the selection of sources seems outdated and incomplete.
Competition
As established in the previous chapter, men perform better in competitive environments than women (Gneezy et al., 2003). If they have a choice, men are more likely than women to opt for competitive situations8 (Niederle & Vesterlund, 2007) . Fox and Lawless looked for competitive activities in college students’ environments and included competition in organized sports in their study. They find that male college students are more likely to have played organized sports and be competitive when playing. Moreover, they conclude that those college students playing organized sports are more likely to have ever thought about running for office.
Fox and Lawless cite multiple yet conflicting sources when introducing the notion of competitive experiences. They claim that playing organized sports enhances several traits critical to political success. They contend that participation in athletic activities is found to enhance young girls’ self-regard, including motivation, self-esteem and confidence (Erkut, Fields, Sing, & Marx, 1996) . The second study they cite finds that sport activities have a positive effect on future earnings, health and subjective wellbeing (Lechner, 2009). However, these studies might not prove representative for American college students, as different sample groups were used. The study by Erkut et al. (1996) was conducted using a sample of American girl scouts9; the study by Lechner (2009) was conducted using a sample of adults in Germany.
Subsequently, Fox and Lawless identify some limitations to their inclusion of athletic participation. The previous section on extracurricular activities already showed that instrumental activities10 have a more positive effect on future political participation than
8 Participants in a laboratory experiment had to solve tasks under noncompetitive piece rate and a competitive
tournament incentive scheme.
9 Girls Scouts are usually between five and twelve years old.
10 Instrumental activities include student government, student newspaper, yearbook, debate club and political
playing sports (Glanville, 1999). Fox and Lawless recognize that research finds that playing sports is not related to civic engagement (Kahne & Sporte, 2008) . However, they explain that they include athletic participation because of its competitive character – and its possible correlation with interest in competing in competitive elections – not because of its weak relationship with civic participation. Surprisingly, Fox and Lawless did not include any previous studies looking into the effects of the competitive nature of organized sports played in high school or college of its relationship with competitive environments pursued later in life. Even though the theoretical framework is incomplete, the results show a gender gap and indicate that playing organized sports has a positive effect on considering to run for political office.
Encouragement
The fourth conclusion Fox and Lawless draw from the data is that young women receive less encouragement to run for office than their male counterparts. This is troublesome, as people are more likely to consider running for office when they are recruited (Fox & Lawless, 2010) . This factor is not introduced by citing separate sources. Rather, Fox and Lawless refer to previous work in which they extensively researched the effect of encouragement from elected officials, party leaders and political activists on a potential decision to run for office.
As such, they include questions about encouragement from those in adolescents’ environments in the sections on family socialization, school environment and peer associations. They focus on encouragement to run in two elections: student government and political office later in life. Fox and Lawless found no statistically significant gender gap in encouragement given to run for student government. However, male college students were more likely to receive encouragement to run for office later in life than their female counterparts from all parties asked about. Parents were 11 percentage points less likely to encourage their daughter to run for office, teachers 7 percentage points and friends 9 percentage points. Moreover, women were 6 percentage points less likely to have received suggestions from three or more sources. This is alarming, as the effect of encouragement strengthens as the number of recruitment contacts increases (Moncrief et al., 2001).
Self-Perceived Qualifications and Self-Confidence
The results from Fox and Lawless’ Citizen Political Ambition Study from 2001 and 2011 already indicated that self-perceptions of qualifications proved a central barrier to women’s decision to run for office. In this study, Fox and Lawless provide a broad and consistent
theoretical framework for their inclusion of self-perceived qualifications in their survey and their resulting claim that female college students are less likely to think they posses the right qualification to run for office in the future.
They refer to studies that show women tend to underestimate their qualifications and show lower levels of confidence. Regardless of their actual levels of political knowledge, both men and women perceive women to be less informed about politics than men (Morehouse Mendez & Osborn, 2010) . Women are also more likely to perceive themselves as less qualified to seek office (Fox & Lawless, 2004) and to undervalue their intelligence, while men are more likely to overestimate theirs (Furnham & Rawles, 1995) . Another study found a small gender gap among young Americans for self-esteem (Kling, Shibley Hyde, Showers, & Buswell, 1999) and confidence(Elder, 2004). Most studies on self-perceived qualifications and self-confidence cited by Fox and Lawless were conducted more than ten to twenty ago. Nonetheless, studies conducted after the publication of this study in 2013 confirm these findings and show that men consistently report higher levels of self-esteem than women (Bleidorn et al., 2016; Orth, 2017).
Similar results can be found in the data. The survey’s section on self-confidence includes three sets of variables: self-assessment of qualifications to run and questions about political traits and skills. Male college students were twice as likely as female college students to indicate that, after they had finished school and had been working for a while, they would know enough to run for political office. ‘Among those who thought they might be qualified to run for office, 53 percent of women, compared to 66 percent of men, had considered politics as a viable option for the future’ (Lawless & Fox, 2013, p. 13) .
Surprisingly, the results of the questions on political traits and skills are not elaborated on in the report, in either a figure or the text. Only the second appendix of the article gives insight into all the skills and traits included in the survey. Respondents were asked about the possession of four traits: confidence, assertiveness, friendliness and knowledge and five political skills: good writer, good at public speaking, knows a lot about politics, usually good at most things, willing to try new things. The only results reported on noted that man were 6 percentage points more likely to assert that they were good at public speaking and 8 percentage points more likely to likely to contend they knew a lot about politics. None of the other skills or traits are reported on. The lack of discussion of the data of the separate variables might point to statistically insignificant results. Moreover, even though the selection of these traits and skills seems logical, no theoretical foundation was provided for the selection of these ‘politically-relevant traits skills’.
