A Woman's Work Is Never Done? Fundraising Perception and Effort Among Female State Legislative Candidates

Transcription

1 University of Massachusetts at Dartmouth From the SelectedWorks of Shannon Jenkins June, 2007 A Woman's Work Is Never Done? Fundraising Perception and Effort Among Female State Legislative Candidates Shannon Jenkins Available at:

2 A Woman s Work is Never Done? Fundraising Perception and Effort Among Female State Legislative Candidates Shannon Jenkins University of Massachusetts Dartmouth Department of Political Science 285 Old Westport Road North Dartmouth, MA sjenkins@umassd.edu Shannon Jenkins is an Assistant Professor of Political Science at the University of Massachusetts Dartmouth. Her research interests focus on state legislatures, including state legislative elections and roll call voting and the differences between female and male state legislative candidates and legislators. Her research has appeared in journals such as Legislative Studies Quarterly, Social Science Quarterly, and Urban Affairs Review.

3 A Woman s Work is Never Done? Fundraising Perception and Effort Among Female State Legislative Candidates Abstract: The lack of female politicians has been attributed to a lack of female candidates for office. However, the reason why there are so few female candidates is not clear. I examine whether differences in fundraising perceptions and effort between female and male state legislative candidates contribute to the lack of female candidates. The results indicate that women do tend to be more concerned about fundraising as is evidenced by greater effort devoted to this campaign function as compared to their male counterparts. Women use more techniques and rely on more sources to secure funds for their campaigns. This suggests that part of the reason women are reluctant to run for office may be due to the fact that they will have to devote more effort to a task candidates generally find distasteful. Despite research showing women are not at a competitive disadvantage when running for office or raising funds for such an endeavor, there still continues to be a dearth of women who do actually run for office. In fact, the number of women seeking office in state legislatures actually declined from 2,375 in 1992 to 2,220 in the 2004 election cycle (CAWP 2004a). The lack of qualified female candidates is of concern to many, as research has shown the presence (or absence) of women in legislative institutions has important implications for both descriptive and proscriptive representation. Part of the reason there are fewer female candidates is that women are less likely to think they are qualified to run or that they are going to win if they do run. This raises the question as to why women think this is the case. Perhaps part of the answer is concerns about fundraising. Voters seem to believe female candidates will have greater difficulty securing funds for their campaigns (Ford 2002). Female candidates may feel the same way. Ironically, the emergence of women s political action committees (PACs) that stress the importance of early money may reinforce these fears (that female candidates need yeast to make dough rise whereas male candidates do not). Thus, it may be that women are more concerned about their ability to 1

4 gain adequate financial backing to mount a successful campaign than are men. Such concerns may also lead women to organize their campaigns differently, with women casting a wider fundraising net. Thus, while women may be just as (or more) successful than men when it comes to the amount of funds they raise, they may have to devote more effort to reach this parity. Several questions are suggested by this line of thought. First, are women more concerned about their ability to fundraise? Second, do such concerns translate into the utilization of more fundraising techniques? Finally, do female candidates rely on more sources for raising campaign funds? This article examines these questions by comparing male and female state legislative candidates perceptions about and effort devoted to fundraising. Surveys of state legislative candidates from nine states in the 1996 election cycle are utilized to examine the extent to which women are more concerned about their ability to raise funds. Additionally, this analysis examines whether male and female candidates rate the importance of various fundraising techniques and sources differently. In general, the analyses show that women do tend to be more concerned than men about fundraising, although the differences are not statistically significant. Additionally, women are more likely to rate as important more fundraising techniques and sources in their campaign. This suggests that while women raise as much money as men in state legislative campaigns, they must worker harder to achieve this parity, relying on more techniques and hitting up more people and groups for money. Potential female candidates may be aware of these differences, contributing to their reluctance to run for office and thus to gender imbalances in the number of candidates for legislative office and, ultimately, legislators. 2

5 Explaining the Presence and Absence of Female Candidates for Office Currently in the United States, female representatives, at all levels, make up a smaller percentage of legislative bodies than they do of the population at large. For example, in 2003, only 13.6 percentage of Congresspersons were female and only 22.5 percent of state legislators were female (CAWP 2004a). Due to the fact that the representation of women in legislative bodies has important implications for the representation of women s policy concerns and how decisions are actually made in these legislative bodies (see Dolan and Ford 1998 for a more complete discussion of this topic), much research has focused on the lack of female representation in legislatures in the United States. Initially, it was thought that women must be facing discrimination in the electoral system. However, research has demonstrated that such discrimination is increasingly rare in terms of electoral outcomes. When women run for office, both at the national and state levels, they are just as successful as men when controlling for incumbency status (Burrell 1992, 1994; Darcy, Welch and Clark 1994; Dolan 1998; Duehrst-Lahti 1998; Huddy and Terklidsen 1993; Seltzer, Newman and Leighton 1997; Smith and Fox 2001). In fact, in comparing success rates for open seat candidates, women typically fare as well as if not better than men (Burrell 1992, 1994; Duehrst-Lahti 1998; Newman 1994; Thompson and Steckenrider 1997). Additionally, female candidates tend to fare just as well as their male counterparts when it comes to financing their campaigns. Again, women consistently do as well as if not better than men when it comes to attracting funds for their candidacies (Burrell 1998, 3

6 1994; Darcy, Welch and Clark 1994; Fox 2000;). Research has shown women fare as well as men in soliciting large donations (Burrell 1994), from a variety of PACs (Theilmann and Wilhite 1991), early donations (Burrell 1994), and from political parties (Burrell 1994). Thus, the key to explaining why there are not more women serving in the legislature, both at the national and sub-national level, is not that women have a difficult time in the electoral arena. Rather, few women serve in the legislature because fewer women run for elective office than men. 1 As a result, research has begun to focus on explaining why fewer women emerge as candidates. It seems that generally speaking, women are more reluctant to run for Note: I would like to thank John Frendreis and Alan Gitelson for the use of their data as well as Doug Roscoe for his comments and Matt Sylvain for his research assistance. Data for replication of the analysis are available from the author. 1 Of course, the fact that incumbents are more likely to win and are more likely to be male hurts women generally. In other words, part of the reason legislatures are so male dominated is because men win seats and then keep them for a long time thus leading to a lack of opportunities for women. However, if this were the sole reason for the lack of female representation, then women should be running for open seats just a frequently as men. This is not the case though; men and women do not run at equal rates in open seat races. For instance, in the period from , only about 24% of state legislative candidates in open seat races were female (Duerst-Lahti 1998). More recently, 21.7% of all female candidates for state legislative seats were candidates in open seat races in 2006 (CAWP, personal correspondence). Thus, emergence appears to be the critical variable in explaining the lack of female representation. 4

7 office than their male counterparts. Research has revealed eligible women (or women in positions that make them most likely to run for office) are less likely to consider running for office, less likely to see themselves as qualified to run for office and less confident in their ability to win if they do run than are men with similar backgrounds (Fox and Lawless 2004; NWPC 1994). But while we do know that women are less confident and therefore less likely to run, we do not know why this is so. As Fox and Lawless (2004, 275) note, we have little data to help pinpoint the source of women and men s different beliefs about their own qualifications. One potential explanation for this hesitation on the part of women is concerns about fundraising. Despite the fact that female candidates are not disadvantaged when it comes to raising money, this does not mean that potential female candidates know this. Or it may be that they do know that they will have to work long and hard (and longer and harder than men do) in order to raise enough funds to mount a competitive candidacy. In fact, Witt, Paget and Matthews (1994, 132-3) assert that many believe potential donors are less likely to contribute to women s campaigns, women are less psychologically disposed to ask for money, and women are less likely to have developed financial networks to help their campaigns. They argue women are not accustomed to asking for things for themselves nor do they feel comfortable with the self-promotion fundraising requires. Thus, women may dislike the idea of spending large amounts of time asking people for money and may think they will have to do a lot of this. This, in turn, may lead them to decide not to run for office in order to avoid having to do something they find 5

8 distasteful. 2 Witt, Piaget and Matthews (1994) note that money as a deterrent to seeking higher office was a common concern cited by potential female candidates. Thus, it seems reasonable to hypothesize that money may be a deterrent to an initial decision to enter into the political arena and seek a lower level political office. Furthermore, if women generally are more concerned about their ability raise funds, then it may also be that women are more aggressive in working to acquire these funds. They may devote more effort to chasing money, relying on multiple fundraising techniques and focusing on a wide variety of sources for these campaign funds. If this is the case, this additional effort may help explain why woman raise as much money as men despite the fact that they are more concerned about their ability to do so. Women may raise as much money because they cast a wider net when appealing for campaign funds. Thus, understanding how women feel about fundraising may be an important initial step in determining why women are more reluctant to run for office. Data and Methods The data for this analysis come from a survey of major party state legislative candidates from nine U.S. states (Arizona, Colorado, Florida, Illinois, Missouri, Ohio, South Carolina, Washington, and Wisconsin) taken in the 1996 election cycle (see 2 This opinion seems to be shared by most politicians, regardless of gender, although it may be the case that, as Witt, Paget and Matthews claim, woman are more uncomfortable than men in doing this. See Open Secrets ( for examples of how current politicians indicate that they too generally find this task distasteful. 6

9 Appendix A for information on survey methods). A number of questions from this survey are of interest for this analysis. First, candidates were asked whether they had concerns about their ability to raise funds, the importance of a variety of groups in their fundraising efforts, and the importance of various fundraising techniques (see Appendix A for the exact wording of these questions). The candidates were also asked a number of demographic questions, including gender and occupation, as well as political questions, such as party affiliation, position in the general election (incumbent, challenger or open seat), competitiveness of the election, how many times prior to this election they had run for this office, and previous political experience. Thus, the responses of male and female candidates to the fundraising questions can be compared while controlling for a number of other factors that may impact responses to these questions. Since this research is focused on the impact of fundraising perceptions on the initial decision to run, only challengers and open seat candidates were included in this analysis. Presumably, since incumbents have already run for office and won, they should be quite aware of the amount of fundraising they would have to undertake in this endeavor. 3 It might be argued that looking only at state legislative candidates does not provide an accurate picture of why women choose not to run for office since these people have decided to become candidates. However, if there are gender differences across the board, as previous research suggests about the decision to run for office generally, then we would expect these differences to remain even for those who have made the decision 3 Incumbent candidates did not differ significantly from open seat candidates and challengers in terms of responses to these questions, so the exclusion of incumbent candidates in the analysis did not change the results presented here. 7

10 to run for office. Furthermore, analyzing differences between candidates is a more stringent test of these hypotheses than examining differences between potential candidates as these people have overcome any reluctance to run. Presumably, these women (as well as the men examined here) have decided to run despite any trepidations about fundraising. Thus, if differences do exist between those who have decided to run, it seems reasonable to suggest that such differences also exist for potential candidates. It may be that differences for potential candidates are even larger given there are theoretical reasons to believe women will have greater concerns about raising money. However, care must be taken in extrapolating these results to potential candidates. In looking at the data, an initial bivariate analysis was run to see whether there were significant differences between the sexes on the dependent variables from above: fundraising concerns, the importance of various groups for fundraising and the importance of various fundraising techniques. When significant differences were found in these bivariate analyses, a multivariate analysis was run to control for the potential impact of other factors. The ten variables where significant differences emerged were ordinal variables; therefore, an ordered regression model was used as it is the appropriate method for examining the relationships between the independent and dependent variables(borooah 2002). The multivariate model includes gender (0 for female, 1 for male) as well political party (1 for Democrat, 2 for Republican), open seat status (0 for challengers, 1 for open seat), whether the candidate had previously run for office (0 for no previous times and 1 8

11 for one or more previous attempts), and how competitive the election was (0 for low competition, 1 for some competition). 4 Results Generally speaking, female candidates for the state legislature do seem to be slightly more concerned about their ability to raise the funds necessary to mount an effective campaign as is shown in Table 1. This is true for all the subsets of candidates examined: female challengers, open seat candidates, candidates in competitive races, Republicans, Democrats, those with no experience running for office or in holding elective office and in a variety of occupations. All were more concerned about fundraising than similar male candidates. However, none of these differences were 4 The first two variables are dichotomous, but the latter two variables had multiple categories. Given that the ordered logit essentially creates a controlled crosstabulation, the large number of categories created when all of these variables were included as originally coded caused problems. Therefore, the categories in these variables were collapsed. For the competitive variable, no opposition, no competition, and slight competition were included in the low competition category while moderately, very and extremely competitive were included in the some competition category. The models were also estimated controlling for whether or not the candidate had a background in business or law, to see if having an occupation with potential connections to financial networks made a difference. This variable was not significant in any of the models and its inclusion did not alter the results in any meaningful way, so it is not presented in the results that follow. 9

12 statistically significant. In some cases, this is probably due to the fact that the differences are fairly small. For instance, only 1.5% more female challengers were concerned than male challengers. However, in other cases, the lack of significance is probably due in part to the small number of cases in a particular category. Comparing open seat candidates, 19.9% more women were concerned than men, but there were only 29 open seat female candidates. Republican women were also more likely to be concerned about fundraising with a difference between men and women in this category of 11.9%. Once again though, there were only 28 women in this category, so this result was not significant. Despite the lack of significance, the consistency of these findings is striking; in every category examined, women were more likely to be concerned than their male counterparts. Table 1 about here So, the next question that emerges is what impact do these concerns have on the fundraising activities of male and female state legislative candidates? As Table 2 indicates, women are more likely to rate the importance of a variety of groups as being extremely important to their fundraising efforts than are men. 5 For every group except 5 It might be argued that these differences are merely a reflection of different evaluative schemes used by men and women. In other words, women rate these groups as more important because they have a tendency to assign higher ratings generally. However, in looking at a number of variables in the survey not related to fundraising, men rated these actors as more important than women in a variety of cases. Thus, these differences seem to reflect actual differences in how men and women s campaigns were organized rather than being an artifact of different rating techniques. 10

13 the national party committee, women were more likely to rate the group as being extremely important to their fundraising efforts than were men. Women were also less likely to rate the group as being not important for every group examined here. For example, 9.6% more female candidates rated the state party as being extremely important as compared to male candidates while 10.8% fewer female candidates said the state party was not at all important. A number of these differences are statistically significant, too. The difference in ratings for the state party, the state legislative campaign committee, labor unions, interest groups and PACs are all significant. As the chi-square and Cramer s V indicate, the largest differences fall into the last two categories. This reinforces the conventional wisdom that women s groups and PACs are important players in the electoral arena, particularly when it comes to funding female candidates. Table 2 about here A similar pattern emerges when the ratings of the importance of various fundraising techniques by female and male candidates are compared. Table 3 reveals women were more likely to rate each technique as extremely important and less likely to rate the technique as not important. For example, 8.5% more women than men rated direct mail as extremely important and 12.4% fewer rated it not important. Most of these differences are statistically significant, too; the differences in ratings for direct mail, television advertising, large fundraising functions, requesting funds from other party committees, and contact with political action committees are all significant. Once again, the largest differences relate to female candidates dealings with political action committees. 18% more women rated contact with PACs as extremely important while 17.4% fewer women rated this technique as not important. 11

14 Table 3 about here So, the picture that emerges in the bivariate analysis is one where women are much more likely to rely on PACs in procuring campaign financing than are men. However, the differences between men and women do not stop here. Women are also more likely to rate a variety of fundraising techniques and the fundraising assistance of a number of groups as being extremely important than are men. Summary variables were created that summed the importance ratings of each technique and each group. 6 The average women s ratings (21.19) for these groups were 3 points higher than the average ratings for men (18.01); again, these differences were statistically significant. Women s ratings (19.47) of the various fundraising techniques were approximately 2.5 points higher, on average, than were men s (17.08); this difference is also significant. Finally, women are also slightly more likely to be concerned about fundraising than are men, although these differences were not statistically significant. All in all though, the data suggests that while female state legislative candidates do raise as much money as male state legislative candidates, they must work more aggressively to do so, by employing multiple methods and targeting multiple sources. Of course, it could be argued that such connections are spurious and it is necessary to control for other factors before determining that women truly are more 6 For the groups, there were eight listed; therefore, scores could range from 8 (all groups were not important) to 40 (all groups were extremely important). For the techniques, there were seven techniques, so scores could range from 7 to

15 concerned about fundraising. 7 Thus, multivariate models were estimated for all of the significant relationships discovered in the bivariate analysis. What is striking across all of these multivariate models is that in almost every situation, there are significant differences between the genders in terms of fundraising activities even after controlling for a number of factors. 8 Looking at table 4, which reports the results for the models estimating the importance of various groups to candidates fundraising efforts, gender is significant in 7 However, it should be noted that bivariate analysis reveals that few of the variables included in the multivariate models are significantly related to gender or each other, and those that are have extremely weak relationships. 8 The models were also estimated using the difference in vote share between the two major party candidates as a proportion of the total vote for the competition variable. In all cases, the model performed less well as measured by the R 2. Gender was no longer significant in the labor union model, and it was only significant at the p<.10 model for the TV model. In most cases, the effect for gender was unchanged although for several models, the effect was larger with the dummy variable. On balance then, regardless of which competition variable was included, there was not much difference in terms of the general conclusions drawn about gender. However, I believe the perception of competition measure is a better gauge of competition for the purposes here. If a candidate felt the race was competitive, then presumably he or she would act as if it were regardless of the actual level of competition as measured by vote totals. Therefore, I have presented the models with the collapsed competition variable. 13

16 every model. 9 Women are more likely to rate the state party, the state legislative campaign committee, labor unions, interest groups and political actions committees as extremely important than are men. Thus, the bivariate relationships remain even after controlling for party identification, open seat status, previous attempts to run for office and the competitiveness of the election. Additionally, gender is the only variable that is significant in all of the models. Table 4 about here From the ordered logit analysis, the predicted probabilities of male and female candidates rating these groups at a given level of importance can be calculated. 10 These predicted probabilities are presented in Table 5. In every case, men were more likely to rate the given actor as not important while women were more likely to rate that actor as very or extremely important. For example, men had a predicted probability of.316 of rating PACs as not important while women had a predicted probability of.208, indicating men were 50% more likely to rate PACs as not important. Conversely, women had a.149 predicted probability of rating PACs as extremely important while men had a predicted probability of.090, indicating women were over 65% more likely to rate PACs 9 Rather than looking at the impact of a given variable on the dependent variable as a regression does, the ordered logit compares the value of the dependent variable for the different categories of a given independent variable. Thus, the positive coefficients for gender indicate that women, when compared to men, give higher rankings on the dependent variables. 10 In calculating predicted probabilities, all other variables were held at their mode. 14

17 as extremely important. Thus, the data reveal that women are far more likely to rely on a variety of groups in their fundraising efforts than are men. Table 5 about here Gender is also a significant predictor of whether a candidate rated direct mail solicitation, television advertising, and contact with political action committees are being extremely important as Table 6 shows. In each of these three cases, women were significantly more likely to report these techniques as being extremely important than were men. Table 6 about here Once again, the predicted probabilities, shown in Table 7, indicate that for these techniques, men were always more likely to rate the technique as not important while women were always more likely to rate it as very or extremely important. For instance, men had a predicted.227 probability while women had a predicted.160 probability of indicating direct mail was not important, revealing men were almost 30% more likely to rate this technique as not important. Conversely, women had a.251 predicted probability while men had a.179 predicted probability of rating direct mail extremely important, a difference of over 40%. Clearly then, women also rely more heavily on a wider variety of techniques in their fundraising efforts than do men. Table 7 about here Discussion The results presented here support the hypotheses about female state legislative candidates and fundraising. Women are more concerned than men about their ability to raise funds. While these differences were not statistically significant (probably due to the 15

18 small number of cases being examined), the results were quite consistent. When looking at a variety of different types of candidates, females were always more concerned than males. These concerns translate into a reliance on more sources and more techniques for female candidates. Of the fifteen different sources and techniques examined here, there were statistically significant differences between men and women for eight in the multivariate analysis. Women were more likely to rate the state party, the legislative campaign committee, labor unions, interest groups and PACs as extremely important to their fundraising efforts than were men. Finally, women were more likely to rate direct mail, television ads and contact with PACs as extremely important techniques in their fundraising than were men. These results present a clear and consistent finding: while women may raise as much money as men, they must work harder to do so by asking more sources and using a wider variety of techniques and services. Thus, contrary to the findings of Dabelko and Herrnson (1997) who observe men and women assemble similar organizations and run similar campaigns, female candidates clearly assemble more extensive fundraising operations than male candidates. 11 Thus, while women s PACs have certainly led to parity in fundraising and so have helped increase the number of women serving in elected positions, women still face difficulties in the electoral arena that may influence their initial decision to run for office. 11 This is probably due to two factors; the first is that they examine Congressional candidates who may have more experience in assembling and running campaigns and the second is that they rely on campaign expenditures rather than specifically comparing the fundraising operations of female and male candidates. 16

19 It is fairly well accepted that most candidates dislike fundraising, and women, it turns out, have to do more of it. Thus, it is probably not surprising that fewer women run. When considering whether or not to run, women may see that they will have to work long and hard (and longer and harder than men) doing something they find distasteful and so may decide that the expected benefits are not worth the effort. While women s PACs have certainly been beneficial to female candidates, their mantra that women must focus on fundraising may discourage potential female candidates from entering into the electoral arena. Perhaps these groups, among others, must do more to reach out not only to those women who are running but to those are who qualified to run in order to assure them that they will be able to mount well-financed and successful campaigns. Finally, more research is needed to understand why women don t run, particularly since the number of women running for office in recent years has been declining from an all time high in 1992 (CAWP 2004b). Fox and Lawless (2004), among others, have shown women are less likely to run. Future research should replicate their methods and survey potential candidates but should particularly focus on those who have contemplated running but who decided not to do so. Determining why these women chose not to run is critical to increasing the representation of women and women s interests in the political arena. 17

20 Appendix A As part of the Election Dynamics Project (EDP) survey, state legislative candidates and county party chairs were surveyed during the 1992, 1994 and 1996 election cycles in Arizona, Colorado, Florida, Illinois, Missouri, Ohio, South Carolina, Washington and Wisconsin. However, candidates were only asked about fundraising in the 1996 cycle. While these states do not represent a random sample of US states, they were chosen because there is a good deal of variation on key state political characteristics, such as political competition, party control of the legislature, governmental capability and legislative professionalization. In the 1996 election cycle, a total of 1,686 mail surveys were sent out to legislative candidates. A total of three waves of the survey were sent out; in each wave, candidates were sent a copy of the survey along with a return envelope with pre-paid postage. The first and second waves were sent to all candidates while the third wave was sent to selected candidates in order to maximize the number of responses in specific states and chambers. In all, a total of 887 surveys were returned for a response rate of approximately 53%. Responses were relatively evenly distributed among Democrats and Republicans and between those who won and lost. While the full survey contained questions on a number of topics, the questions that are of interest here are those dealing with fundraising. The wording of the questions utilized in this analysis are as follows: 18

21 15. How important were the following groups in providing your campaign with assistance in fundraising? (Circle the number associated with the most appropriate choice for each group) Note: Emphasis in original. Respondents were asked to rank the importance of your family and friends, county party, state party, your party s state legislative campaign committee, national party committee (DNC or RNC), labor unions, other interest groups, and political action committees on a scale of 1 (Not Important) to 5 (Extremely Important) 20. How important were the following techniques to your fundraising efforts? (Circle the appropriate choice for each technique.) Note: Emphasis in original. Respondents were asked to rank the importance of direct mail, television advertising, asking individuals or small groups for money, large fundraising functions, requesting funds from your party s state legislative campaign committee, requesting funds from other party committees, and contact with political action committees on a scale of 1 (Not Important) to 5 (Extremely Important). 22. When you decided to run for the state legislature in this election, did you have concerns about your ability to raise the funds which you thought would be necessary to run an effective campaign? Note: This was an open-ended question. Respondents who indicated a concern were coded as having concerns, no response (or a response of no in the comments section) were coded as not having concerns. 19

22 WORKS CITED Borooah, Vani K Logit and Probit: Ordered and MultinomialModels. Thousand Oaks: Sage Publications. Burrell, Barbara Campaign Finance: Women s Experience in the Modern Era. In Women and Elective Office, ed. Sue Thomas and Clyde Wilcox. Oxford University Press: New York. Burrell, Barbara A Women s Place is in the House: Campaigning for Congress in the Feminist Era. Ann Arbor: University of Michigan Press. Burrell, Barbara The Presence and Performance of Women Candidates in Open- Seat Primaries for the U.S. House of Representatives: Legislative Studies Quarterly 17: Center for American Women and Politics (CAWP). 2004a. Facts on Women Candidates and Elected Officials. Retrieved on 1/14/2005 from /Facts/Officeholders/cawpfs.html Center for American Women and Politics (CAWP). 2004b. Record Number of Women Seek Seats in U.S. House; Candidate Numbers at Other Levels Don t Match Record Highs. Retrieved on 3/10/2005 from Facts/Elections/Post%20primary%202004%20press(01-05).pdf. Dabelko, Kristen La Cour and Paul S. Herrnson "Women's and Men's Campaigns for the U.S. House of Representatives." Political Research Quarterly 50: Darcy, R., Susan Welch and Janet Clark Women, Elections and Representation. Lincoln: University of Nebraska Press. 20

23 Dolan, Kathleen Voting for Women in the Year of the Woman. American Journal of Political Science 42: Dolan, Kathleen and Lynne E. Ford Are All Women State Legislators Alike? In Women and Elective Office, ed. Sue Thomas and Clyde Wilcox. New York: Oxford University Press. Duehrst-Lahti, Georgia The Bottleneck: Women s Experience in the Modern Era, in Women and Elective Office: Past, Present and Future. Thomas, Sue and Clyde Wilcox, eds. New York: Oxford University Press. Ford, Lynne E. Women and Politics The Pursuit of Equality. Boston: Houghton Mifflin Company. Fox, Richard Gender and Congressional Elections. In Gender and American Politics, ed. Sue Tolleson-Rinehart and Jill Josephson. Armonk: M.E. Sharpe. Fox, Richard L. and Jennifer L. Lawless Entering the Arena? Gender and the Decision to Run for Office. American Journal of Political Science 48: Huddy, Leonie and Nayda Terkildsen The Consequences of Gender Stereotypes for Female candidates at Different Levels and Types of Office. Political Research Quarterly 46: National Women s Political Caucus (NWPC) Why Don t More Women Run? A study prepared by Mellman, Lazarus, and Lake. Washington: National Women s Political Caucus. Newman, Jody Perception and Reality: A Study Comparing the Success of Men and Women Candidates. Washington, DC: National Women s Political Caucus. 21

24 Seltzer, Richard, Jody Newman and Melissa Voorhees Leighton Sex as a Political Variable: Women as Candidates and Voters in U.S. Elections. Boulder: Lynn Reinner Publishers. Smith, Eric R.A.N., and Richard L. Fox A Research Note: The Electoral Fortunes of Women Candidates for Congress. Political Research Quarterly 54: Theilman, John and Al Wilhite Discrimination and Congressional Campaign Contributions. New York: Praeger. Thompson, Seth and Janie Steckenrider Gender Stereotypes and Decision Context in the Evaluation of Political Candidates. Women and Politics 17: Witt, Linda, Karen M. Paget and Glenna Matthews Running As A Woman: Gender and Power in American Politics. New York: The Free Press. 22

25 TABLE 1: Differences in Concerns about Fundraising % Women % Men Chi-Square Concerned Concerned Open Seat & Challengers 78.6 (84) 71.5 (260) (.205) Challengers 78.2 (55) 76.7 (172).049 (.825) Open Seat 79.3 (29) 61.4 (88) (.077) Competitive Seats 79.2 (72) 72.7 (176) (.290) Republicans 78.6 (28) 66.7 (129) (.218) Democrats 78.6 (56) 76.3 (131).111 (.739) Never Run Before 78.1 (64) 74.6 (201).321 (.571) No Prior Elective Office 82.2 (45) 73.4 (173) (.223) Lawyers 87.5% (7) 78.1% (25).352 (.553) Business/Sales 78.9% (15) 65.9% (54) (.269) Education 100% (9) 77.8% (21) (.121) Note: Cell entries are the percent of candidates in each category who indicated they had concerns about their ability to raise funds. Numbers in parentheses are the number of cases included in the cell except for the final column where the numbers in parentheses is the p value. For the final eight rows, only open seat candidates and challengers are included in the comparison. 23

26 TABLE 2: Differences in the Importance of Groups in Assisting with Fundraising Women Men Importance Not Slight Moderate Very Extreme N Not Slight Moderate Very Extreme N Chi- Square Family and Friends (.318) County Party (.104) State Party (.012)* State LCC National Party Committee Labor Unions Interest Groups (.003)** (.673) (.014)* (.000)** PACs (.001)** Note: Cell entries are percent of candidates in each category indicating the level of importance for the specified group in providing assistance with fundraising. * p <.05, ** p <.01 Cramer s V

27 TABLE 3: Differences in the Importance of Fundraising Techniques Women Men Importance Not Slight Moderate Very Extreme N Not Slight Moderate Very Extreme N Chi- Square Direct Mail (.028)* Television Advertising (.001)** Asking Individuals Large Fundraising Functions Requesting from State LCC Requesting Funds from Other Party Committees (.683) (.003)** (.054) (.024)* Contact with PACs (.000)**.233 Note: Cell entries are percent of candidates in each category indicating the level of importance for the specified given technique in fundraising efforts. * p <.05, ** p <.01 Cramer s V

28 TABLE 4: Predicting the Importance of Various Groups in Fundraising Ordered Logit State LCC Labor Interest PACs Party Unions Groups Gender * * * ** ** Open Seat Candidate ** ** Party Previous Times Run Competitive N Chi-Square Nagelkerke R Note: Entries in cells are unstandardized logit coefficients; p values are in parantheses ( * p <.05, ** p <.01) 26

29 TABLE 5: Predicted Probabilities for Men and Women of Rating Various Groups At a Given Level of Importance in Fundraising Value Not Slight Moderate Very Extremely State Party Male Female LCC Male Female Labor Unions Male Female Interest Male Groups Female PACs Male Female Note: Predicted probabilities are calculated by holding the other variables at their mode. 27

30 TABLE 6: Predicting the Importance of Various Fundraising Techniques Ordered Logit Direct TV Ads Large Other PACs Mail Groups Party Committees Gender * ** ** Open Seat Candidate * Party Previous Times Run Competitive * N Chi-Square ** * Nagelkerke R Note: Entries in cells are unstandardized logit coefficients; p values are in parantheses ( * p <.05, ** p <.01) 28

31 TABLE 7: Predicted Probabilities for Men and Women of Rating Various Fundraising Techniques at a Given Level of Importance Value Not Slight Moderate Very Extremely Direct Mail Male Female TV Ads Male Female PACs Male Female Note: Predicted probabilities are calculated by holding the other variables at their mode. 29