Women hold fewer than 25% of all American elected offices despite being over 50% of the population. This paper explores this incongruity and explains why women feel that they must reach a higher threshold than men in order to run for office. Consequently, this work demonstrates that we should expect female candidates to be of higher quality than male candidates.
Kendall Tucker is an undergraduate at Columbia College working toward her political science bachelor’s degree in May 2014. She is currently studying politics and history at the University of Oxford and she is excited to once again work as a research assistant for Professor James Snyder in Harvard University’s Department of Government this summer. Her areas of interest include the measure of quality candidates, women in politics and social/educational policy reform.
Women make up 51% of the United States population, but only 18.3% of Congress.1 According to recent studies, these low rates cannot be attributed to discrimination at the polls. Why then aren’t women proportionally represented in politics?2 Each year, far fewer women run for office than men.3 Jennifer Lawless and Richard Fox conducted a study for the Brookings Institute in which they asked, “Why Are Women Still Not Running for Public Office?”4 They concluded that because “women are significantly less likely than men to view themselves as qualified to run for office,” women need to reach a higher threshold before they decide to enter a race.5 This threshold, whether it be monetary, familial, or fame-based, would make a candidate higher quality, because it would indicate that his/her probability of winning is higher than his/her competitors. Lawless and Fox’s conclusion was that this threshold is the leading cause of the disparity between the number of men and women who decide to try for elected office.6 In their article, “Entering the Arena? Gender and the Decision to Run for Office”, Lawless and Fox wrote, “it stands to reason that women who think they are ‘qualified’ are actually more qualified than men who self-assess this way.”7 In this paper, I will expand their conclusion by arguing that because women feel that they must reach a higher threshold than men in order to run for office, we should expect female candidates to be of higher quality (i.e. more likely to raise more money and more likely to win) than male candidates.
Lawless and Fox concluded that women feel less qualified than men to run for office because “deeply embedded patterns of traditional gender socialization” cause women to receive less encouragement than men to run for office.8 Additionally, the lack of an incumbency advantage for most women likely contributes to their decisions not to run for office.9 Both of these reasons lead women to feel less qualified than their male counterparts. Often they don’t overcome the threshold needed to enter an electoral race. This threshold depends on many factors. In Cherie D. Maestas, Sarah Fulton, L. Sandy Maisel and Walter Stone’s article called “When to Risk It? Institutions, Ambitions, and the Decision to Run for the U.S. House”, they write that “ambition, in conjunction with lower office conditions [i.e. being a lower office holder], create incentives” to run for office. They also highlight the monetary and time commitments that might dissuade a potential candidate from entering a race.10 These measures of fundraising ability, prior office holdings, political connections and/or incumbency all determine the relative quality of a candidate and lower the threshold that a potential candidate might feel that he/she would need to overcome before running for office.11
My hypothesis going into this study was that females would be higher quality candidates than male candidates, because they needed to reach a higher threshold before they decided to run for office. Because quality is difficult to measure empirically, I used fundraising as a proxy for quality. It seems likely that higher quality candidates would be more effective at raising money than lower quality candidates. Research shows that legislative candidates who raise more money than their opponents are successful about 76% of the time.12 For this analysis I gathered Federal Election Commission records for the 2010 House races and compared them to information provided by Opensecrets.org. These records gave me information about the amount of money each candidate raised and the candidate’s status as an incumbent, challenger, or candidate for an open seat. Using Google and other online resources, I determined the gender of each candidate. In an Excel spreadsheet, I compiled all this information and graphed the mean and median fundraising levels for different groups. I used regression analysis to determine whether the relationship between money and gender was statistically significant. For the gender variable, I coded female candidates as 1 and male candidates as 2.13
I discovered that women raised more money on average than men both in total and as incumbents and challengers, but less when they were competing for open seats. Conversely, men’s median values were higher than women’s both in total and as incumbents and candidates for open seats, but they were lower as challengers. Ultimately, the t statistic for the coefficient on gender was not high enough to indicate that gender had a statistically significant effect on levels of fundraising. This is true whether we look at all candidates or subsets of incumbents, challengers, or candidates for open seats. This complicates my hypothesis, because even though women did not raise statistically higher values of money than men, the question arises whether fundraising ability was the perfect proxy for the quality of candidates.
The main limitation of this study was the amount of data that I used. The information that I compiled was only from the 2010 House election, meaning that I was comparing data from 816 candidates who ran in that cycle. Compared to the total number of candidates who have run for federal office in recent years, this was a small sample size. This limitation may have contributed to the mean and medians’ lack of statistical significance. Also, the sample size of women who ran for office was incredibly small. Compared to the 683 men who ran for the House of Representatives in 2010, only 133 women ran. In terms of contenders for open seats, only 11 women ran. This probably decreased the t statistic for the coefficient on gender, because it further decreased a sample size that was already small.
Another limitation of this study was that because I researched only one election, I was assuming that there was nothing special about 2010 that may have caused female candidates to raise more or less money than in other election cycles. However, the 2010 election was a midterm election. Midterm elections are historically different than elections during presidential years, because voter turn-out is lower and the majority party (for the most part) loses seats in both houses.14 These trends hurt incumbents and may have changed the fundraising strategies of the candidates, but presumably these trends to some extent helped female candidates because they are less likely to be incumbents.
Despite these limitations and the lack of a statistically significant difference in the fundraising levels of men and women, the analysis of the data provided compelling evidence about the quality of female candidates. I predicted when starting this analysis that female candidates would be better candidates than males and that this could be measured by their fundraising ability. My data from the 2010 election provides mixed evidence for this hypothesis. Even though the differences in the data are not statistically significant, the data can still be examined for patterns that may arise if the sample size were expanded. Also, after sorting the data by gender and money raised, I was able to find other characteristics that may have made the women higher quality candidates.
The mean amount of money female candidates raised was higher than the mean amount men raised. Female candidates raised on average $1,201,104 and male candidates raised just $1,162,560. On the other hand, males’ median value of money raised ($997, 508) was higher than female’s median value of money raised ($853,579). This raises the question of why the mean and median values of female candidates were different and how this affects my hypothesis. The amount of money raised by female incumbents versus male incumbents follows this trend. Female incumbents raised on average $1,606,149, while male incumbents raised just $1,599,218 on average. In terms of the money raised at the median, male incumbents beat female incumbents with $1,304,217 compared to $1,243,441. This means that for incumbents, women raised more money than men on average, but women’s median value was lower. This again raises the question of whether the amount of money women raised was a trend or an artificially high value.
Challengers and candidates for open seats provide the most direct test for my hypothesis, because presumably more of them have never held office. This means they relied more on their ability to fundraise than incumbents did, because they couldn’t count on as much name recognition. The average ($672,435) and median value ($216,980) of money that female challengers raised was higher than the corresponding values ($666,051, $213,449) that male challengers raised. Unfortunately, there was much less data on the amount of money that candidates for open seats raised, but female candidates competing for open seats had lower average ($1,122,713) and median values ($965,853) of money raised than male candidates.
I believe there was a contradiction between the mean and median values of money that female candidates raised because the top female candidates raised more money than the top male candidates. The top 10% of men raised $3,630,812 on average while the top 10% of women raised $3,922,668. Female candidates’ mean was bolstered by women like Michelle Bachmann (R-MN) who was planning on running for president and Tarryl Clark (D-MN) who Democrats had recruited to defeat Bachmann.15 Bachmann raised $13,567,811, while Clark raised $4,718,912. This group of top female fundraisers also included Nancy Pelosi (D-CA) who as Speaker of the House raised immense amounts of money to fend off Republican challengers and Carolyn Maloney (D-NY) who represented the expensive media market of New York City.16 The reasons that these women raised so much money indicate that their personal stories, rather than their gender, contributed to the inflated female mean.
What does this difference between female candidates’ mean and median values mean for my hypothesis? Even without the top 10% of female candidates, the middle 80% of women still raised more money on average than the middle 80% of men. The place where women stopped raising more money on average than men was in the middle 50% of female and male candidates. This explains male candidates’ higher median. The middle 50% of female candidates raised $954,953 on average versus the middle 50% of male candidates who raised $974,583. This indicates that either my hypothesis that female candidates are of higher quality than male candidates is incorrect, or fundraising ability is not an effective proxy for the quality of candidates.
Of the middle 50% of female candidates (in terms of fundraising), 72% of this group were incumbents and 82% of them won their races. Many women in this group, like Doris Matsui (D-CA) and Donna Edwards (D-MD) faced weak challengers who did not come close to raising ¼ of what they raised. Barbara Lee (D-CA), Jackie Speier (D-CA) and Candice Miller (R-MI) faced no challengers at all. Karen Bass (D-CA) ran for an open seat. Because her district was so heavily democratic and she was a fixture in the community, she was able to raise 9 times the amount of money her Republican challenger did and she won the district.17 Tea party candidate Vicky Hartzler (R-MO) who raised ½ of what incumbent Ike Skelton did, beat him by connecting his policies to those of the locally unpopular President Barack Obama.18 All of these women raised money in the middle 50% range of female candidates, but the case study evidence and their high win percentage show that they were high quality candidates.
These instances indicate that women in the middle 50% may have raised less money in their races because they had other reasons to believe that they were high quality candidates. Incumbency, strong community ties and national trends are all powerful factors that contributed to the strength of these women’s candidacies. There are many other similar examples where factors other than the ability to fundraise made females high quality candidates. These results indicate a gap between the strength of the candidate and the amount of money that they were able to fundraise. This means that if this study were to be conducted again, other measures of quality aside from fundraising abilities should be incorporated into measures of quality as well.
Earlier in this analysis, I said that challengers and open seat candidates would be the best measures of the quality of female candidates, because they have fewer attributes other than their ability to fundraise that contribute to their decision to run. Of course, they may have held prior political office or may have family members involved in politics, but none of them are able to claim the incumbency advantage, which greatly helps candidates win political office.
As was mentioned above, female challengers raised more money on average than male candidates did. This was the one category where female candidates’ median values were also higher than male candidates’ median values. Open seat candidates’ values were the opposite, but because there were so few women who ran for open seats, I am disregarding this data. Even though the data on female challengers was not statistically significant, these patterns were compelling and they beg further investigation. It is possible that the amount of money raised was not a good proxy for the quality of a candidate if the candidate was an incumbent, but if the candidate was a challenger or a candidate for an open seat, fundraising ability would be much more indicative of their quality. If this is true, I predict that further investigation into female challengers will show that they consistently raise more money than male challengers because they are higher quality candidates.
Even though the data that I collected was not statistically significant for the differences in money raised by male and female candidates, other factors led me to confirm my hypothesis that females are higher quality candidates than males. When the middle 50% of women fundraisers was closely examined, the evidence suggested that they were high quality candidates who, because of their incumbency or another factor, didn’t raise as much money as other candidates, but still raised enough to win their races. 82% of female candidates in the middle 50% of female fundraisers won their races in comparison to 78% of males in the middle 50% of male fundraisers.
If I were to repeat this study in the future, I would improve my analysis by increasing my sample size to include at least all the federal races in the last 15 years. With this larger sample size, I would likely be able to get statistically significant results on the different amounts of money raised by male and female candidates. I could also get a better feeling about which women were outlier candidates.
The conclusion that I have drawn from this study is that females who decide to run for office are more likely to be higher quality candidates than males, but their quality cannot be directly connected to their fundraising abilities. Even though there was no statistically significant correlation between women’s fundraising abilities and their decisions to run, when I examined case studies, it seemed like women had stronger backgrounds than men. There are many attributes that raise a candidate’s level of quality, so by accounting for just one I got statistically inconclusive results. Through my closer examination of individual races, I realized that incumbency advantages, familial connections and anything else that garnered name recognition raised female candidates’ perceptions of their own quality so that they would be willing to run for office.
1 “Current Numbers of Women Officeholders.” Center for American Women and Politics. Rutgers, Web. 17 Apr. 2013.
2 Darcy, R., and Sarah Slavin Schramm. 1977. “When Women Run Against Men.” The Public Opinion Quarterly 41 (1) (April 1): 1-12.
3 Leeper, Mark Stephen. 1991. “The Impact of Prejudice On Female Candidates:” American Politics Research 19 (2) (April 1): 248 -261. doi:10.1177/1532673X9101900206.
4 Lawless, Jennifer L., and Richard L. Fox. 2008. Why Are Women Still Not Running for Public Office? Governance Studies. Washing: Brookings, May.
5 Lawless, Jennifer L. 2010. It Still Takes a Candidate: Why Women Don’t Run for Office. Rev. ed. New York: Cambridge University Press.
6 Lawless and Fox. 2008.
7 Fox, Richard L, and Jennifer L Lawless. 2004. “Entering the Arena? Gender and the Decision to Run for Office.” American Journal of Political Science 48 (2) (April 1): 264-280. doi:10.1111/j.0092-5853.2004.00069.x.
8 Lawless and Fox. 2008.
9 Erikson, Robert S. 1971. “The Advantage of Incumbency in Congressional Elections.” Polity 3 (3) (April 1): 395-405. doi:10.2307/3234117.
10 Maestas, Cherie D., Sarah Fulton, L. Sandy Maisel, and Walter J. Stone. 2006. “When to Risk It? Institutions, Ambitions, and the Decision to Run for the U.S. House.” American Political Science Review 100 (02): 195-208. doi:10.1017/S0003055406062101.
11 Krasno, Jonathan S., and Donald Philip Green. 1988. “Preempting Quality Challengers in House Elections.” The Journal of Politics 50 (04): 920-936. doi:10.2307/2131385.
12 Casey, Linda. “The Role of Money & Incumbency in 2009-2010 State Elections.”FollowTheMoney. National Institute on Money in State Politics, 3 July 2012. Web. 17 Apr. 2013.
13 The graphs, charts and t statistics are included as an addendum at the end of this paper.
14 Tufte, Edward R. 1975. “Determinants of the Outcomes of Midterm Congressional Elections.” The American Political Science Review 69 (3): 812-826. doi:10.2307/1958391.
15 Rothenberg, Stuart. 2009. All Signs Point to Another Top-Notch Election Cycle in 2010. The Rothenberg Political Report. September 10. http://rothenbergpoliticalreport.com/news/article/all-signs-point-to-another-top-notch-election-cycle-in-2010.
16 Johnson, Dennis W. 2001. No place for amateurs: how political consultants are reshaping American democracy. Psychology Press.
17 Felde, Kitty, and Frank Stoltze. 2010. Congresswoman Diane Watson to retire; Assembly Speaker Karen Bass to run. 89.3 KPCC. February 10. http://www.scpr.org/news/2010/02/10/11697/congresswoman-diane-watson-retire-assembly-speaker/.
18 Anon. 2010. Ike Skelton Loses, Vicky Hartzler Wins Missouri House Seat. Huffington Post. November 3. http://www.huffingtonpost.com/2010/11/03/ike-skelton-loses-vicky-hartzler-_n_778054.html.
Mean amount of money raised for all candidates: $1,168,842 Median amount of money raised for all candidates: $984,052
t = -0.33 (t statistic for the coefficient on gender)
t = -0.04 (t statistic for the coefficient on gender)
Anon. 2010. Ike Skelton Loses, Vicky Hartzler Wins Missouri House Seat. Huffington Post. November 3. <http://www.huffingtonpost.com/2010/11/03/ike-skelton-loses-vicky-hartzler-_n_778054.html>
Casey, Linda. “The Role of Money & Incumbency in 2009-2010 State Elections.”FollowTheMoney. National Institute on Money in State Politics, 3 July 2012. Web. 17 Apr. 2013.
“Current Numbers of Women Officeholders.” Center for American Women and Politics. Rutgers, Web. 17 Apr. 2013.
Darcy, R., and Sarah Slavin Schramm. 1977. “When Women Run Against Men.” The Public Opinion Quarterly 41 (1) (April 1): 1-12.
Erikson, Robert S. 1971. “The Advantage of Incumbency in Congressional Elections.” Polity 3 (3) (April 1): 395-405. doi:10.2307/3234117.
Felde, Kitty, and Frank Stoltze. 2010. Congresswoman Diane Watson to retire; Assembly Speaker Karen Bass to run. 89.3 KPCC. February 10. <http://www.scpr.org/news/2010/02/10/11697/congresswoman-diane-watson-retire-assembly- speaker/>
Fox, Richard L, and Jennifer L Lawless. 2004. “Entering the Arena? Gender and the Decision to Run for Office.” American Journal of Political Science 48 (2) (April 1): 264-280. doi:10.1111/j.0092- 5853.2004.00069.x.
Johnson, Dennis W. 2001. No place for amateurs: how political consultants are reshaping American democracy. Psychology Press.
Krasno, Jonathan S., and Donald Philip Green. 1988. “Preempting Quality Challengers in House Elections.” The Journal of Politics 50 (04): 920-936. doi:10.2307/2131385.
Lawless, Jennifer L. 2010. It Still Takes a Candidate: Why Women Don’t Run for Office. Rev. ed. New York: Cambridge University Press.
Lawless, Jennifer L., and Richard L. Fox. 2008. Why Are Women Still Not Running for Public Office? Governance Studies. Washing: Brookings, May.
Leeper, Mark Stephen. 1991. “The Impact of Prejudice On Female Candidates:” American Politics Research 19 (2) (April 1): 248 -261. doi:10.1177/1532673X9101900206.
Maestas, Cherie D., Sarah Fulton, L. Sandy Maisel, and Walter J. Stone. 2006. “When to Risk It? Institutions, Ambitions, and the Decision to Run for the U.S. House.” American Political Science Review 100 (02): 195-208. doi:10.1017/S0003055406062101.
Manning, Jennifer, and Colleen Shogan. 2011. Women in the United States Congress: 1917-2011. CRS Report for Congress. Congressional Research Service, September 27.
Rothenberg, Stuart. 2009. All Signs Point to Another Top-Notch Election Cycle in 2010. The Rothenberg Political Report. September 10. <http://rothenbergpoliticalreport.com/news/article/all-signs- point-to-another-top-notch-election-cycle-in-2010>
Tufte, Edward R. 1975. “Determinants of the Outcomes of Midterm Congressional Elections.” The American Political Science Review 69 (3): 812-826. doi:10.2307/1958391.