Political Experience and the Success of Female Gubernatorial Candidates

This research examines the effect that political experience has on the success of female gubernatorial candidates by analyzing the female vs. male gubernatorial elections from 1976–2014. The study questions whether prior statewide political office experience is advantageous for female candidates and is this consistent across party lines. This research builds on the political pipeline theory, which suggests that the shortage of women holding higher level political office, such as governor, is due to the lack of qualified and motivated women running for political office. We argue that women who hold lower level statewide office develop the necessary experience and name recognition that provides them with the qualifications and motivation, both within themselves and from other sources such as political parties, to run for and win the office of governor. Our findings suggest that statewide office experience is important for the success of Democratic women gubernatorial candidates but not for Republican women gubernatorial candidates.


Introduction
In 1925, the first female governor was elected and since then, thirty-seven women have governed their states [1] The growing number of women governors has enabled scholars to study the reasons for these women's success.Currently, there are six women governors leading their states (NH, NM, OK, OR, RI and SC).One of the most recent additions to this group of female leaders is Rhode Island governor, Democrat Gina M. Raimondo, who won the open seat with 40.7% of the vote [2].Governor Raimondo is the first female governor for her state, although she is not the first female gubernatorial candidate for Rhode Island.Democrat Myrth York ran unsuccessfully for the governor's seat in 1994, 1998 and 2002, and prior to York's three gubernatorial campaigns, Republican Elizabeth Ann Leonard was unsuccessful in her attempt to unseat the incumbent governor in 1992 1 .Even though residents of Rhode Island had five opportunities to elect a female governor, only in 2014 was the female gubernatorial candidate successful.
One of the reasons why York and Leonard were unsuccessful in their gubernatorial quests while Raimondo is now leading her state as governor may be the type of political experience the candidates have as they run for the top executive office in their state.Elizabeth Leonard, a small business owner, was considered a political newcomer when she became the Republican gubernatorial candidate in 1992, and Myrth York's political experience was in the state senate.However, Gina Raimondo came into the 2014 gubernatorial race as the current General Treasurer of the state.In 2010 she had been elected to this statewide position and as a result, had the experience and name recognition that came with that 2010 election and her years of service as the "custodian of State funds, charged with the safe and prudent management of the State's finances" [3].
This study examines the effect that previous political experience can have on the success of female gubernatorial candidates.In other words, are women more likely to be successful in getting elected as governor if they have held an elected political office previously?We base our expectations on the political pipeline theory, that the reasons there are fewer female candidates and officeholders are because there are less women who have the qualifications and encouragement to run for political office.The pipeline theory is best summarized by Mariani's work which states, "The pipeline theory predicts that women serving in lower levels of political office will leverage political resources and experience gained at those levels to advance to higher office" ( [4], p. 285).Research shows gaining experience through the pipeline is more important for women than for men.Baer and Hartmann ( [5], p. 12) state that, "Female candidates often experience greater questioning of their qualifications than their male counterparts".Therefore, women need success at a lower level to establish credibility for higher office while men are more likely to be perceived as qualified even with less political experience.In addition, even when women have the qualifications to run for political office, research tells us that there is a gender gap in political ambition.Fox and Lawless [6,7] conclude that women are more likely than men to question their credentials when determining whether to run for political office.In other words, despite women and men's similar personal and professional characteristics, women have lower levels of political ambition than men do.Therefore, it is even more important for women to have success at a lower level of political office because of the gender disparity in how their qualifications are viewed both by themselves personally and by others.We expect as more women develop the necessary qualifications and realize they are viable candidates, we will see more female candidates and officeholders.
Furthermore, we question if certain levels of political office experience are more advantageous for female candidates.For example, are female candidates who have lower level statewide office experience, such as Lieutenant Governor or Attorney General, more successful in their quest for the governor's office than female candidates with other types of experience, such as holding a local office, state legislative office, or other roles, such as in business?Finally, because past research suggests that for female gubernatorial candidates, the path to success reflects a different candidate pool for Democrats than for Republicans [8], we examine any differences in paths to success by political party.
We begin with a discussion of past research on the factors that explain the success and failure of female candidates before focusing more specifically on gubernatorial elections.We then examine the female vs. male gubernatorial elections from 1976-2014 to identify if a certain career path is associated with the success of female gubernatorial candidates measured by the percentage of the vote received by the female candidates.We conclude with suggestions for future areas of research in this field.

Literature Review
Over the years, research has addressed the factors that affect the electability of female candidates.One reason that is used to explain the lack of women in elected office is the incumbency advantage.The benefit of incumbency is the name recognition and resources that are vital for a successful campaign [9].Because, over the years, most of the incumbents have been men, they are the candidates who have benefited from the advantage [10][11][12].As more women get elected and run for re-election, the advantage of incumbency will benefit the women.The anticipated solution to this barrier for women is for women to run for open-seats where they stand an equal chance as men of getting elected [13,14].
Another reason for the low level of female elected officials is the lack of party support that female candidates receive.Studies contend that party elites may not view women as viable candidates and therefore are less likely to provide organizational and monetary support for the women [11,12,15].There are indications that as more women have run and been successful in their bids for political office, party support has increased [16,17].Along with this party support, interest groups are realizing the viability of female candidates and are providing the necessary financial support for a female candidate to be successful.Evidence of this is in the similar amounts of Political Action Committee money that female and male candidates are receiving [18].Furthermore, the establishment of various organizations, such as EMILY's List and Wish List, which support female candidates have also provided the necessary encouragement and resources for a candidate to become an officeholder [11,19,20].
Researchers also suggest that the types of issues focused on during the campaign can influence the support that women candidates receive [12,21,22].When elections revolve around security and crime issues, voters tend to view women as ill-equipped to deal with such issues and will not vote for the women.These assumptions are based on long-established stereotypes that are difficult to overcome.However, when the issues are gender-related, such as reproductive freedom, or social concerns like education, health care and assistance for the disadvantaged, voters assume that female candidates are more competent in dealing with these issues and will therefore vote for the women.
Finally, the lack of qualified female candidates has slowed attempts to increase the number of female officeholders at the higher levels.Scholars have found that women are less likely than men to run for political office [23,24] and this results in a limited number of viable female candidates that voters can choose from.In their comprehensive analyses of the political ambition of women, Fox and Lawless [6,7] found that even when women have the same personal characteristics and qualifications as men, the women are less likely to seek elective office.This is due to several factors including the lack of encouragement by political and non-political actors to run for office.Another consideration is the fact that women tend to underestimate their qualifications while men are inclined to overestimate their qualifications.In addition, women are more likely to perceive the political process as biased against them resulting in their decision not to run for political office.Further, women are more limited by their family obligations than men are; research suggests that household and childcare responsibilities are still performed more by women than by men [24].Whether it is due to the lack of encouragement for women to run for office, women's personal views that they are unqualified, or women's perceptions of their family responsibilities, the pool of viable female candidates remains shallow.
Although this may seem discouraging, this condition may be changing.Literature has addressed the development of a political pipeline that is providing a pool of experienced female candidates for various political offices [25][26][27].Here we find more women working their way up the political ladder and developing the credentials and name recognition that are necessary for a candidate to get elected to a higher office.Advocates for expanding the pool of experienced women candidates expect that as more women get elected and serve in lower political offices, we will see increases in the number of these experienced female candidates who run for higher offices [10,11,23].However, others [4] are less optimistic and contend that gender-based differences complicate the political pipeline for women resulting in women being less likely than men to get elected to a higher level political office.
While most of this literature focuses on female candidates for state and national legislative seats, more recently we find research on the obstacles that female gubernatorial candidates deal with.A significant amount of this research addresses the media coverage of the women's campaigns.Scholars note that due to differences in media coverage, women are at a disadvantage when compared to their male rivals [28][29][30][31][32]; however, some argue that media coverage is becoming more gender balanced [33] and is more reflective of other aspects such as partisanship and incumbency [34].
Additionally, studies examine pre-election and exit polls to determine voter support of female gubernatorial candidates [35,36].Although finding different results, these studies show that the gender, age, partisanship and region of residence of voters affect their level of support for female gubernatorial candidates.Further research demonstrates that the novelty of being the first female gubernatorial candidate in a state's history decreases voter support for the female gubernatorial candidate and their likelihood of winning [37]; when the novelty is gone, female gubernatorial candidates are more successful.In addition, research tells us that there are differences in the parties' nomination of female candidates for governor and these differences can negatively affect the chances of female gubernatorial candidates [8].The Republican Party is more likely to nominate a female candidate as a sacrificial lamb to run against an incumbent than the Democratic Party is.Furthermore, the Democratic Party is more likely than the Republican Party to have developed a political pipeline of qualified female candidates to run for governor.As a result, female Republicans have a more difficult time of getting elected to the governor's office.
Moreover, the literature demonstrates that the society and culture of each state determines the likelihood of women entering and winning their respective races.Jason Windett [38] concludes that a favorable female sociopolitical subculture results in a stronger probability of success for female gubernatorial candidates.This study also found that female gubernatorial candidates who held elective office previously had an increased likelihood of winning their respective race.
Windett's findings prompt us to ask if experience as a statewide executive office-holder increases the likelihood of success for a female gubernatorial candidate, which is the core focus of this study.Our study relies upon the theoretical approach of the political pipeline theory discussed above.Experience in lower-level statewide political offices should produce greater success for those same women when they run for governor.Specifically, we argue that women with statewide experience will do better than those without it because of the benefits in name recognition, experience with running a statewide campaign, and the credibility that comes with holding these types of offices.Therefore, we hypothesize that female gubernatorial candidates who have lower level statewide executive office experience are more likely to get elected than female gubernatorial candidates with other types of experience such as local political office, state legislative, or even from the world of business.Furthermore, because the traditional pipeline model applies to Democratic women candidates and not to Republican women candidates [8], we hypothesize that this finding will be stronger for Democratic women than for Republican women candidates.

Data and Analysis
To investigate the impact of prior statewide elective office on the success of female gubernatorial candidates, we examine the elections in which a female gubernatorial candidate ran against a male gubernatorial candidate from 1976 to 2014.We follow the approach of Oxley and Fox [39] and use Ordinary Least Squares (OLS) regression with robust standard errors of the pooled data to account for issues related to states appearing multiple times in the data.In addition, we use logistic regression to test the same model on a dichotomous dependent variable indicating whether the female candidate won the election 2 .
Because we are interested in whether statewide elective office is an advantage for women seeking the governorship, we limited the cases to only those in which a female candidate was seeking election and not women incumbents seeking re-election to the position of governor.Therefore, the data includes all contests in which a female challenger faced a male incumbent or when a female candidate faced a male candidate in an open-seat election.
Furthermore, we restrict the analysis to only those races in which the two major candidates in the general election were of different sexes.This allows us to perform the most direct examination of the impact of candidate sex on electoral success.Elections with two men or two women running against each other are excluded because they do not provide a scenario in which candidate sex can be treated as a variable within an electoral context because the sex of the candidates is the same for both opponents.Because of the possibility of the results being impacted by outlier cases, we examined the variance, min/max, and histogram of the dependent variable.There is no evidence suggesting a particular problem.The variance is 78 with a maximum value of 59.2 and a minimum of 20.The histogram shows a somewhat normal distribution with a slight right skew but without obvious outliers that would impact the results.To stay consistent with the Oxley and Fox [39] article about women in elective office across the states, we use the same OLS approach.The concern about outliers was more of a concern for us when we conducted the logistic analysis.Therefore we also estimated it using exact logistic regression and rare event logit (which is more common in political science).There was no substantive difference in the results suggesting that there is no problem with outlier cases.

Dependent Variables
The primary dependent variable, Female Vote Share, is the overall percentage of the vote received by the female candidate.Therefore, in races with a Democratic female gubernatorial candidate the dependent variable is the percentage of the vote received by the Democratic candidate.For contests with a Republican female gubernatorial candidate, the dependent variable is the percentage of the vote for the Republican candidate.For the logistic analysis, we use a dichotomous variable that indicates if the female candidate won the election.

Independent Variables
Our primary independent variable of interest is Female Candidate with Statewide Office Experience.This variable captures whether or not the female gubernatorial candidate held statewide elective office.As mentioned earlier, the political pipeline theory contends that female candidates need to get elected to lower level political offices to gain the credentials to be successful in getting elected to higher level political offices such as the governor's office.Literature tells us that experience in some offices, such as lower level statewide offices, may be more beneficial than other lower level political office experience.In a 1992 study, Squire [40] argued statewide office holders might make stronger candidates than those with state legislative or local office experience due to statewide name recognition and the experience of running a successful statewide campaign.Additionally, the Barbara Lee Family Foundation [41] found that voters viewed lower level statewide office experience, such as the offices of Attorney General or Lieutenant Governor, as more beneficial for gubernatorial candidates than state legislative, civic or corporate leadership positions.
Moreover, focus groups done by the Barbara Lee Foundation [41] revealed that voters want more information about female gubernatorial candidates than male gubernatorial candidates to determine if they are qualified to be governor.Evidence in crisis management experience and the ability to deal with financial issues was required of the women candidates.Men, on the other hand, were viewed as qualified to be governor if they had some form of leadership experience.Furthermore, men were given more credit for their private sector experience than women were given for comparable experience.In other words, women have to prove themselves more to get voters' support.Based on these findings, lower level statewide experience appears to be more important for women than for men.
This variable is measured as a dichotomous variable indicating whether the candidate's most recent level of political office was a statewide position below the office of governor.We use the dichotomous measure because of both the nature of what the variable is expected to measure and based upon past research.This type of experience is expected to capture the importance of a prior statewide election for the female candidates.This importance is based on the experience and credibility that is gained by mounting a successful statewide campaign in the past.Candidates who have done this have shown the ability to raise money that is needed for such a contest, have been able to construct a winning coalition in the exact political district (the entire state) that is needed for a successful gubernatorial campaign, and have worked with party activists and media throughout the state during the past campaign.In addition, the previous work by Squire [40] which developed a more comprehensive measure, in fact, treated all lower statewide offices as a single value 3 .
3 Using Squire's [40] work as a model, we examined the possibility of a more comprehensive measure of candidate experience.He develops a more comprehensive measure for US gubernatorial elections which ranks different types of offices in an ordinal measure and then multiples that value by the percentage of the state population in the official's constituency.However, that measure is not applicable to our analysis for several reasons.First, his highest category includes former governors and U.S. senators; however, there are no women with such experience who have run for governor in the general election.Therefore, that part of the measurement is not applicable.Second, as mentioned in the text, even his more comprehensive measure assigns a constant value to all lower statewide officials.They are given a score of 4, which is then multiplied by 100 since they all represent 100% of their states' population.Third, the data include two female U.S. House members who ran for governor.Although it is problematic to include a dummy variable with only two positive cases, we estimated the equation with a separate variable for congressional experience, with congressional experience included The list of the most frequent prior levels of political experience by gubernatorial candidates in the study is provided in Table 1.Among the most commonly held lower level statewide offices for female gubernatorial candidates are Lieutenant Governor (14.8%),State Treasurer and other similar positions (18.5%), and Attorney General (8.6%).In all, 41 of the 81 women candidates (50.6%) in these races held a statewide position as their most recent political position.Of the remaining cases with either no prior political experience or experience at some other political level, the most common office is that of a state legislator.Twenty-four (29.6%) of the candidates in this sample held state legislative office as their most recent political experience.As discussed above, it is expected that holding a statewide political position should be positively related with the percentage of the vote received by the female candidate and that this effect should be greater for Democratic women than for Republican women.In addition to the primary independent variable of interest, we include several additional variables to account for any influence they might have on the dependent variables.First, we include Male Candidate with Statewide Office Experience, a variable that measures the prior political experience of the male opponent.This is measured the same way as it is for women candidates.This variable is included because experience level of the female candidate does not exist in a vacuum.Each female candidate for governor in the dataset competed against a male candidate for governor with each asking voters for their support based, in part, on both candidates' political experience.
It is difficult, however, to know exactly what to expect from this variable.As discussed earlier, past research suggests that men tend to be more aggressively ambitious about their political careers [6].The common way this has been described is that women need to be asked while men do not.Within this study and research question, aggressive ambition suggests that men may be less likely to move up the traditional political ladder and, instead, cut in line or move from an entirely different candidate pool.In fact, unlike the political experience levels we found among the female candidates, the male candidates in the dataset are less likely to have statewide experience.While just over half of the women in the dataset had lower level statewide experience, just under 30% of the men in the dataset had that same level of experience.This is consistent with Squire's [40] study that found a higher level of political experience for female gubernatorial candidates than for male gubernatorial candidates.
Because the dataset includes those races in which a female candidate is seeking election as governor and not those seeking reelection, it includes races with a male incumbent.To account for any difference between running as a challenger against a male incumbent and as a candidate in an open-seat election, we also include the dichotomous variable, Open-seat Elections, indicating whether the contest is an open-seat election (1) or one in which a female challenger faces a male incumbent (0).with statewide experience based on Squire's claim that it was similar, and with those two cases excluded from the analysis.Congressional experience was not significant as a stand alone variable and none of the results for the other variables were significantly different under any of those scenarios.Therefore, to isolate the impact of statewide experience versus all other types of experiences, we present the analysis with the simple dichotomous measure.
Because it is expected that candidates do better for open-seats than as challengers, we expect a positive relationship between this variable and the dependent variables.
In addition to the above election specific factors, we also include a variable, Novelty, which indicates if the female gubernatorial candidate is the first female gubernatorial candidate in that state's history.The novelty factor has been shown to have a negative impact on the success of women gubernatorial candidates [37].
Besides these election specific factors, we incorporate two state specific features into the analyses as independent variables.The first of these is Citizen Ideology.This is measured using the citizen ideology data from Berry et al. [42].For a given election, we used the value of the ideology of the state for the period prior to the election.Although it is anticipated that more ideologically liberal states will be a more supportive environment for women gubernatorial candidates, any impact should be greatest for Democratic women.Additionally, the success of a new generation of more distinctly conservative Republican gubernatorial candidates [43] makes any particular expectation for state ideology more difficult to anticipate for Republican women 4 .
The second state specific variable is Women in the State Legislature.Here we include the percentage of the state legislature that is female prior to the gubernatorial election.Conventional wisdom suggests that success at one level eventually leads to success at another level.Greater percentages of women in the state legislature also suggest that the state's electorate and other political actors are more willing to elect women to higher levels of political office.

Findings
The following equation was estimated for this study of all female vs. male gubernatorial campaigns with the results presented in Table 2  Our primary hypothesis is that women with prior statewide office experience will experience greater electoral success than women with other types of experience.The findings support our hypothesis and, in total, suggest that three of the four election specific variables impact the success 4 The Berry et al. [42] measure of ideology is only one of those used in the literature.In addition, we estimated the equations using the Enns and Koch [44] and Windett [38] state public opinion and female sociopolitical subculture measures respectively.The results were the same.Like the Berry ideology variable, neither the Enns and Koch nor the Windett measure achieved statistical significance and neither the significance nor the coefficients of the other variables were substantially impacted.Because both of those measures are only available until 2010 and the Windett data does not begin until 1978, we show the results that include the Berry measure so that we do not lose the 1976, 2012, and 2014 elections from our analysis.of female gubernatorial candidates.Women who have held lower level statewide elective office receive approximately 3 more points on their vote totals than women who have not.It appears that women gubernatorial candidates enjoy a modest but potentially important benefit from prior statewide experience, which is consistent with our expectations.This suggests that to be successful it helps to have experience running and winning statewide office prior to the gubernatorial election.The impact may be due to the experience of running a statewide campaign or even the credibility that comes with holding that level of political office.
It is interesting to note the effect of the variable, Male Candidate with Statewide Office Experience.There is a statistically significant and positive relationship between the male opponent holding statewide elective office and the share of the popular vote for the female candidate.In other words, female candidates do slightly better against male opponents who hold this level of elective office than against those with other types of experience such as local level office, state legislative office, or from non-political recruiting pools such as the business community.This finding is consistent with Squire's contention [40] that high levels of political experience are more important for women than for men.
As expected, the variable Open-Seat Elections attains traditional levels of statistical significance.For women candidates, running in an open-seat election adds almost 5 points to their vote total compared to those running against male incumbents.
Neither of the state specific factors nor the Novelty variable were statistically significant.Women candidates did not do better in liberal states than they did in conservative states.Although this finding might be slightly surprising, as previously noted the recent success of a new generation of particularly conservative female governors in decidedly conservative states, such as Governor Haley of South Carolina [43], may indicate that state conservatism is declining as a barrier to women gubernatorial candidates' success.
Because elections are ultimately about winning, we also used logistic regression to estimate an equation predicting whether the female candidate won the election.Our model produced similar findings for our primary hypothesis.The results are presented as odds ratios in Table 2.This analysis produced similar outcomes for our primary hypothesis.The findings indicate that women candidates with prior statewide experience are more likely to win elections than those without that type of experience.These results suggest that women candidates with this type of experience are approximately four times more likely to win election than those without it.The findings are also as expected for women running in open seat elections compared to those running against male incumbents.Women in open seats are thirteen times more likely to win than those running as challengers showing the power of incumbency.The impact of statewide experience for men no longer reports as statistically significant suggesting that for male candidates this experience can impact vote margins but not necessarily their overall chances of winning.
The findings presented in Table 2 are for all female candidates.Because past research suggests that there is a substantial difference in both recruiting and the opportunity structure for women in the Democratic and Republican parties [8,45], we conducted separate analyses for Democratic and Republican female candidates.This research suggests that the pipeline theory applies to Democratic women gubernatorial candidates but not for Republican women gubernatorial candidates.This partisan difference is based upon the different cultures of the two parties and their approaches to recruiting and supporting women politicians.Freeman [46] explains that the differences in the party cultures and structures lead to a partisan difference in the success of people within each party who express a high level of group salience such as women.Therefore, we hypothesize that the impact of lower level statewide office experience on the success of women candidates is more important for Democratic women than for Republican women.To test for this partisan difference we conduct separate analyses for Democratic and Republican female candidates.One challenge for this research is the sample size.The data include 55 Democratic women candidates and only 26 Republican women candidates.Because of the small sample size for Republican women, we began by estimating the full model for only Democratic women candidates.These results are reported in Table 3.The findings for the analysis of Democratic female candidates suggest an even larger impact of prior statewide elective office experience.Like the analysis of the entire dataset, the variable Female Candidate with Statewide Elective Office Experience is statistically significant but the coefficient indicates that this level of experience adds approximately 5½ points to the overall point total instead of over 3 points as suggested by the other findings.The findings also indicate a slightly higher odds of winning than the other findings based upon the logistic analysis.This result suggests that the impact of prior political experience is particularly important for Democratic female candidates and is consistent with past work that indicates the presence of a more traditional political pipeline within the Democratic Party for women as they move from one level of elective office to higher levels of elective office.
The findings in Table 3 suggest a similar effect of the prior political office of the male candidate as found in the analysis of the entire dataset while the variable, Open-seat Elections, is no longer statistically significant.
Because of the small number of Republican women in the data set, we were hesitant to estimate the full model in our analysis of Republican women.Instead, we estimated a reduced model that only included the three election specific variables that attained statistical significance as reported in the OLS model in Table 2.These results are presented in Table 4.The findings for the analysis of Democratic female candidates is almost identical to those found in Table 3.The findings for Republican female candidates indicate that none of the variables attain statistical significance.These findings must be viewed cautiously because there are so few Republican women.The results suggest that the success and failure of Republican women candidates for governor cannot be explained by the same factors used to explain the success and failure of Democratic women.More research on the particular question about Republican women running for governor is needed to develop a more comprehensive understanding for these contests.

Conclusions and Directions for Future Research
In this research, we examine the unique situations of female vs. male gubernatorial campaigns.While much of the early research was limited to single or comparative case studies, the increasing frequency of these types of races permits us to begin conducting larger n quantitative studies.Our research examined the impact of contest specific and state specific factors on the vote totals and the likelihood of winning for women running against men for governor.
Of primary interest for this research is whether the expectations from the pipeline theory hold when women politicians seek the highest executive office in their respective states by testing the effect of prior political experience on the success of female gubernatorial candidates.Past research suggests that prior political experience is important and, perhaps, is more important for women than for men.Consistent with pipeline theory, our findings provide evidence that a certain type of political experience is beneficial for Democratic women seeking election to the position of governor.As hypothesized, it is advantageous for Democratic women to have the experience of a prior lower level statewide campaign and the credibility from holding such an office to use as a springboard for a run for governor.The same advantage is not found for Republican women, which supports research that suggests different career paths for candidates in the Democratic and Republican parties.However, because of the small sample size it is too early to make definitive conclusions for Republican women.
By focusing on only female vs. male gubernatorial campaigns, we are able to isolate gender dynamics for a given election for the under-researched area of gubernatorial campaigns.This focus also presents challenges based upon the limited number of cases.Early research such as ours can help guide future endeavors as the number of cases increases with time.Of particular interest for future research might be a better understanding of the career paths of male gubernatorial candidates relative to their female counterparts.Due to differences in party support and personal ambition, male candidates with strong potential and appeal as gubernatorial candidates may skip the political middle-ground of lower level statewide positions.Men who move more slowly up the political ladder may be those whose political career paths have been blocked or who have more limited political ceilings.These types of male candidates may be safe picks when the party's prospects are less promising but could be passed over for nomination by less experienced rising stars when the party's prospects are strong.
Finally, another important question for future research concerns the nomination stage.While this research focused on the general election, the impact of lower level statewide elective experience may also impact the likelihood of winning a party's nomination for governor.Specifically, do women with this level of experience face tougher primary opposition than their male counterparts and do we see differences in this effect based upon partisanship?As more women run for governor over time, we will be able to gain a broader and more comprehensive understanding of the challenges and opportunities for women as they seek the highest political office in their state.

Table 1 .
Prior political office of non-incumbent gubernatorial candidates.
Numbers indicate percentage of all non-incumbent candidates with the indicated level of prior political experience by gender.The Total Lower Level Statewide category includes offices other than the top three that are listed for comparison purposes.

.
Female Electoral Success = a + Female Candidate with Statewide Office Experience + Male Candidate with Statewide Office Experience + Open-seat Elections + Novelty + Citizen Ideology + Women in the State Legislature.

Table 2 .
Vote for female gubernatorial candidates running as challengers or in open-seat elections, 1976-2014.

Table 3 .
Vote for Democratic female gubernatorial candidates running as challengers or in open-seat elections, 1976-2014.

Table 4 .
Reduced model explaining vote for female gubernatorial candidates running as challengers or in open-seat elections by party, 1976-2014.