A candidate's sex may affect how accessible elective state judgeships are to women attorneys. Research has offered conflicting findings on the competitiveness of women congressional candidates: Several studies argue that women compete equally with men, while others conclude that women are at a disadvantage. This study draws upon the North Carolina trial court elections and provides one of the few empirical analyses of the women and men who compete in judicial elections. Its findings coincide with both conclusions of the research on women congressional candidates. Although the women candidates in this study were more adept than men at financing their campaigns, they still had to overcome challenges to win seats on the trial court.
A more women join the legal profession, their chances of becoming judges are affected by their ability to compete effectively in state judicial elections. Several states have replaced (or are considering replacing) elections with merit selection; nevertheless, a majority of states still rely upon contested partisan or nonpartisan elections to choose at least some of their judges (Carp, Stidham, and Manning, 2004; Smith, 1997).1 In these states, women attorneys must compete in contested judicial elections in much the same way as candidates campaign for nonjudicial polit' ical office.
Although numerous studies have examined the impact of gender upon political campaigns and elections, research upon the women who campaign in judicial elections has been strikingly limited. (Reid, 2000). Though important and helpful, the existing literature exploring the impact of gender upon political campaigns and elections offers differing views on how competitive women candidates are when they pursue elective political office. Several studies have suggested that women compete equally with men in large part because they can finance their campaigns as effectively as men (Hernnson, Lay, and Stokes, 2001; Smith and Fox, 2001; Burrell, 1998, 1994, 1985; Uhlaner and Schlozman, 1986). Yet, a number of studies have concluded that women nevertheless are at an electoral disadvantage, particularly when they compete directly against men (Herrick, 1996, 1995; Fox, 1997).
This study fills a void in the existing literature. Specifically, it examines the impact of a candidate's sex upon judicial campaigns and judicial elections. By providing one of the few empirical comparisons of the campaigns of the women and men who competed for state judgeships, this study complements and supplements the prevailing research on judicial selection and gender politics. The findings illuminate the intricacies of court elections, highlight the similarities and differences between judicial elections and political elections, and show broadly how women candidates fared in electoral politics. Therefore, this study's empirical analysis of the men and women who participated in judicial elections promises to add to our understanding of gender politics, judicial selection, and electoral politics.
This study adopts the methodology found in existing political science research. It uses campaign contribution totals and percentage of the vote as indicators of electoral competitiveness. Drawing upon the research on women political candidates and judicial selection, this study also assumes that money plays an important, though not determinative, role in mounting viable campaigns for political and judicial office (Richey, 2000; Burrell, 1998; Selzer, Newman, and Leighton, 1997; Herrick, 1995; Nicholson and Nicholson, 1994; Moog, 1992; Westlye, 1991; Schotland, 1985). Similarly, percentage of the vote provides a very good measure of electoral competitiveness because it more accurately than win/loss verdicts indicates how competitive candidates are in comparison to their opponents and to other candidates. For example, a candidate who wins by 60 percentage points is thought to be electorally stronger (more competitive) than a candidate who wins by 51 percentage points (Hall, 2001; Herrick, 1996, 1995). Furthermore, the scope of this study's analysis includes comparisons of men and women as a group: comparisons of men and women when they ran as incumbents, as challengers, and as open-seat candidates, and comparisons of men and women when they competed in same-sex and different-sex races. Finally, this study employs aggregate comparisons and multivariate analysis to examine the judicial candidacies of men and women.
This study draws upon the North Carolina District Court elections, held between 1994 and 1998, to examine the impact of a candidate's sex upon judicial campaigns and judicial elections. It provides an empirical and comparative analysis of the campaigns of the men and women who participated in North Carolina trial court races and an opportunity to assess whether the prevailing assumptions about women political candidates similarly characterized how women fared in these district court races. The article begins by discussing the opposing viewpoints on how competitive women candidates are in the electoral process. The next section explains why the trial court elections in North Carolina represent a good set of races to examine whether and how the sex of a candidate affects judicial elections. The third section presents the findings from aggregate comparisons and multivariate analysis and, thus, provides an empirical comparison of the campaigns and electoral results of the women and men who vied for seats on the North Carolina District Court between 1994 and 1998. The article concludes with a summary of the findings and final observations.
CONFLICTING VIEWS ON THE COMPETITIVENESS OF WOMEN CANDIDATES
The existing literature exploring the impact of gender upon political campaigns and elections offers conflicting conclusions on the electoral competitiveness of women candidates. Influenced largely by Barbara Burrell's research on women congressional candidates, several studies have found that women competed equally with men when pursuing elective political office (Hernnson, Lay, and Stokes, 2001; Smith and Fox, 2001; Burrell, 1998, 1994, 1985; Werner, 1997; Uhlaner and Schlozman, 1986). Burrell and others have found that women political candidates competed equally with their male counterparts because women experienced little or no discrimination in financing their campaigns. They have assumed that although fund-raising prowess hardly guarantees electoral success, it nevertheless provides a very good indicator of electoral competitiveness (Hojnacki and Kimball, 2001; Burrell, 1994; Sorauf, 1992, 1988; Sabato, 1990, 1984; Eismeier and Pollock, 1984). Indeed, access to a varied and large network of funding sources suggests that a candidate has the financial resources to mount a viable political campaign, that a candidate has a broad base of electoral support, and that a candidate has become an effective player in the electoral process (Fox, 1997).
Burrell's findings were further affirmed when researchers showed the importance of candidate status in assessing the impact of a candidate's sex on political campaigns. When women incumbents were compared with men incumbents, when women challengers were compared with men challengers, and when women and men in open-seat races were compared with each other, no significant sex differences in the candidates' reported receipts were discovered (Selzer, Newman, and Leighton, 1997; Werner, 1997; Darcy, Welch, and Clark, 1994). Finally, women congressional candidates were found to have established a broad fund-raising network by attracting financial support from political action committees (PACs), from women's groups, from political parties, and from large donors (Burrell, 1994; Uhlaner and Schlozman, 1986). According to Burrell: "Whether we look at totals, sources or timing, female candidates in similar situations as male candidates generally do as well and sometimes even better in financing their campaigns for national office" (Burrell, 1994:128).
Women Face Disadvantages in Political Races
An alternative view argues that such broad comparisons of men and women may ignore electoral realities and underplay the disadvantages that women confront when competing in political contests. Perhaps the scholar whose research has mounted the most substantial challenge to Burrell's conclusions has been Rebekah Herrick. Herrick has questioned the assumption that men and women compete equally in political elections by pointing to the underrepresentation of women in Congress and in state offices; if campaign funding is one of most important keys to electoral success and if women are superb fund raisers, then, as several scholars have asked, why are there so few congresswomen and so few women in state legislatures and governor's mansions? These studies have questioned whether comparisons of contribution totals reported by men and women candidates can discern the subtle types of discrimination that women candidates are likely to confront (Herrick, 1996; Fox, 1997; Leeper, 1991; Sapiro, 1981-82).
In her studies of women candidates for the United States House of Representatives, Herrick argued that the scope of the analysis must take into account the realities of the electoral process. Aggregate comparisons of men's and women's campaign finances present an interesting yet partial picture of political races because candidates do not campaign in isolation; instead, they compete against other candidates. To assess the importance of a candidate's sex upon his or her electoral success, one should compare men and women when they compete for the same seat or, at least, when they compete in similar contests. In addition to studying men and women as competitors in political races, Herrick's indicator of electoral competitiveness was percentage of the vote, and not win/loss outcomes. Percentage of the vote has the advantage not only of providing a precise measure of a candidate's electoral strength but also of showing one candidate's electoral strength in direct comparison with an opponent (Herrick, 1996).
Accordingly, by revising the methodology used by Burrell and others, Herrick concluded that women do indeed face gender-based barriers in political elections. Women candidates running in the congressional races held between 1988 and 1992 were at an electoral disadvantage because "campaign spending helps male candidates more than female candidates" (Herrick, 1996:73). Since the value of a dollar spent by women candidates was less than that spent by men candidates, women have to spend more money to be as competitive in their races as their similarly situated male counterparts (Herrick, 1996).
Women in Judicial Elections
Since women judicial candidates have received scant scholarly attention, the extent to which their sex has helped or hurt their electoral chances is largely unknown. However, the research on judicial elections and gender politics allows us to develop certain assumptions or hypotheses about how a candidate's sex is likely or unlikely to affect the judicial electoral process.
Identifiable differences between political and judicial elections as well as between women political and women judicial candidates may underplay the importance of a candidate's sex in judicial races. Women attorneys may be more insulated than women political candidates from the effects of adverse gender stereotypes. The prevailing research has shown that women who pursue congressional and statewide office tend to be much more politically experienced and politically sophisticated than most women attorneys who participate in judicial elections (Duersti-Lahti, 1998; Thielemann, 1993). Women political candidates establish impressive political resumes in large part to offset voter concerns about their competency and qualifications (Leeper, 1991; Sapiro, 1981). These concerns about competency might be less important in judicial elections because most states statutorily mandate that all judicial candidates must meet stipulated educational requirements and must have specific professional credentials. By meeting state eligibility requirements for judicial office, women attorneys have already insulated themselves against the charge that they are not as competent to become state judges as their male counterparts. Women political candidates, who do not have to meet similarly rigorous eligibility requirements, must prove through other methods that they are as capable as men.
Moreover, women judicial candidates may be better positioned to fund their campaigns than women political candidates and are probably positioned as well as men judicial candidates. One of the most serious problems confronting women who run for political office is developing strong and reliable contacts with traditional funding sources. Scholars have argued that the creation of women's organizations, such as Emily's List, indicated that women were not accepted by traditional funding sources, such as business PACs and political parties (Fox, 1997). Like political candidates, judicial candidates must aggressively pursue funding opportunities to offset the mounting costs of campaigning for judgeships (Wohl, 2000; Reid, 2000, 1996; Mahtesian, 1998; Friedman, 1998; Uelmen, 1997; Champagne and Cheek, 1996; Dubois, 1986a, b). However, access to funding networks might be less of a problem for women pursuing judicial office, though. Women judicial candidates, who are all attorneys, are likely to have developed through their law practices and professional affiliations strong working and personal relationships with the individuals likely to be the primary contributors to judicial campaigns. Since individual attorneys, law firms, bar associations, and attorney groups are the main financiers for judicial candidates (Champagne and Cheek, 1996; Reid, 1996), women attorneys should have numerous professional and personal contacts within this lucrative fund-raising network.
Although these professional differences may blunt the impact of being a woman in a judicial race, different campaign styles could also exacerbate or minimize sex/gender differences between men and women candidates. Political campaigns are often characterized by their combative and, at times, vitriolic style. Such an intense, competitive, and confrontational environment may elevate the relevancy of gender stereotypes for candidates when developing campaign strategies and for voters when appraising candidates. Unlike their male counterparts, for example, women political candidates must be mindful of cultural perceptions that condemn women for being viewed as too aggressive and, yet, dismiss women for being viewed as too demure. Indeed, women political candidates have adopted deliberate campaign strategies to downplay unfavorable sex stereotypes that hurt their viability and to emphasize favorable sex stereotypes that boost their viability (Fox, 1997). Women judicial candidates may not have to overcome or, at least, address a similar electoral burden because, until recently, state canons of judicial ethics and state statutes have restricted the manner in which candidates can campaign for state judgeships. These canons and statutes might have leveled the electoral playing field for men and women because they ensure that the confrontational and vitriolic campaigns that characterized political contests are less likely to characterize judicial campaigns.2 Women judicial candidates may be less concerned about addressing cultural perceptions in their campaigns for judgeships than are women seeking political office.
Despite the above factors that would minimize the importance of a candidate's sex in judicial elections, the low salience of judicial races conceivably has the opposite effect. In comparison to political races, judicial contests tend to be low-salience elections in which voters have very little information about the candidates. Unlike their political counterparts, women attorneys can point to state statutory requirements as an indicator of their competence to serve on state courts. However, such an argument is unlikely to resonate with voters, who have little or no knowledge of state eligibility statutes. Since voters are largely unaware of the personalities and platforms of judicial candidates, they are likely to use a voting cue, such as a candidate's sex or a sex stereotype, in judicial races. Accordingly, the role of "judge" may be perceived as a "man's job," and voters might be inclined to favor male candidates in men v. women contests. Conversely, voters may also use sex or a sex stereotype as a cue to vote in favor of women candidates (Leeper, 1991; Conover, 1981). From this perspective, the sex of the candidate may become the most important factor in determining whether women can win elected judgeships.
Therefore, it is unclear whether the conclusions presented by Burrell or Herrick best reflect the electoral environment facing women judicial candidates. In many ways, women judicial candidates are less likely to be affected negatively in their campaigns for judgeships by sex stereotypes. According to Burrell, women are as competitive as men because women are as proficient in financing their campaigns as men and because they are especially prudent in expending their funds (Burrell, 1994). There is no reason to believe that women judicial candidates would be less effective fund raisers than women congressional candidates; in fact, given their professional and personal associations, women attorneys might have better access to potential contributors than many women seeking political office. It is reasonable to hypothesize or to assume that women judicial candidates will encounter little or no discrimination in financing their campaigns and that, like their political counterparts, they will raise and spend more money than men. It follows that, as BurrelPs research suggests, women should be perceived as being as electorally competitive as men when they run for elected state judgeships.
In contrast, the electoral climate that characterizes most judicial elections makes Herrick's findings applicable to women judicial candidates. The low salience of judicial elections creates a climate whereby voters are likely to be influenced by or to use sex perceptions or sex stereotypes as voting cues. If the position of being a judge is perceived to be a "man's job," then women must do more in their campaigns (for example, raise and spend more money) than men to convince voters of their competence or to overcome voters' perceptions about the competence of men. Herrick's research suggests that, as they have in congressional races, voters might be inclined to support male candidates in men v. women races. (In other words, women were electorally disadvantaged particularly when they competed against men.) Under this scenario, Herrick's findings conceivably are as applicable to judicial contests as they are to political contests. In this context, it is reasonable to hypothesize that, unlike their male counterparts, women will encounter significant challenges in winning judicial elections and that, like women congressional candidates, they will be particularly challenged when competing directly against men.
Recognizing the methodological limits of generalizing from a small number of races held in a single state, this study is unable to test whether Burrell's or Herrick's arguments explain how all or most women compete in judicial elections. What this study can and will do, however, is ask whether the women who pursued seats on the North Carolina trial court confronted an electoral climate that, as Burrell argues, allowed women to fund successful campaigns or an electoral climate that, as Herrick argues, forced women to overcome challenges that their male counterparts did not have to face.
NORTH CAROLINA DISTRICT COURT RACES
The general elections for the North Carolina District Court held between 1994 and 1998 provide a good set of races to explore how women compete for elected state judgeships. The North Carolina District Court is the state's minor trial court. As such, it hears misdemeanors, civil suits in which less than $10,000 is in jeopardy, and selected family cases. Judicial candidates run for district court seats within one of forty geographical districts (judicial districts) dispersed throughout North Carolina.3 Interestingly, these judicial districts do not correspond closely with other political districts; they can cross county lines or encompass one county, and they do not have the same geographic parameters as state legislative or congressional districts. District court judges are the most electorally accountable judges in North Carolina because, in contrast to higher state court judges who serve eight-year terms, their term of office is four years. During the 1994 to 1998 election cycles, North Carolina used partisan elections to select district court judges.4
Focusing upon the North Carolina District Courts allows us to examine judicial elections in the courts where elections are most likely to be used to select state judges. It is curiously interesting that the judicial selection literature largely focuses upon elections for state supreme court justices (Champagne and Creek, 1996; Hall, 1992; Baum, 1987) because, in practice, more trial court judges attain their seats via partisan or nonpartisan elections than do appellate judges. Of the approximately 8,500 state trial judges in the country, 76 percent are elected; in contrast, about 53 percent of the 1,200 appellate judges compete in state judicial elections (Smith, 2002; Solimine, 2002).
Furthermore, campaigns for district court seats typify trial court campaigns. Although candidates run as nominees of the state political parties, they are not formally chosen by the political parties; the decision to run is essentially an individual decision, and the candidate's party affiliation is determined by the candidates. Like most judicial elections, most district court races are largely low-salience races characterized by candidate-centered and very personal campaigns. However, there have been district court races that resemble the professionally managed, very competitive, and very costly trial races that take place in urban areas in New York or in politically charged states such as Texas (Newfield and Miner, 2003; Bernstein, 1998; Borreson, 1998). Generally, judicial campaigns are likely to be very expensive and likely to involve professional managers and consultants in North Carolina's metropolitan areas of Raleigh and Charlotte. In addition, like most states that elect their trial judges, North Carolina has adopted contribution limits but has failed to place any ceilings on campaign spending. Judicial candidates are free to raise and spend as much money as they can in their efforts to win a seat on a district court.5
Moreover, trial courts, like the North Carolina District Courts, provide a very good venue to examine the campaigns of women candidates. Occupying the lowest rung in the state court system, trial courts are probably the most convenient and least threatening courts for attorneys who have judicial aspirations (Martin, 1999). These seats are very accessible for a number of factors: more seats available, short terms of office, and elections are local rather than statewide. Unsurprisingly, women serve in higher numbers at the trial court level than in state appellate courts. This is true in North Carolina as well as in several states that elect trial court judges.
In addition to structural features of its district court, North Carolina's political climate recommends this study's focus upon the trial court races. Despite the visibility of women in several statewide electoral posts, women have fared less well in their pursuit of seats at the district level. The contrast is indeed notable. As a result of statewide elections, North Carolinians have elected women as their lieutenant governor, commissioner of labor, commissioner of agriculture, secretary of state, and, most recently, United States senator. There is one woman sitting on the state supreme court, and women judges fill a third of the seats on the state's intermediate appellate court. A different electoral picture emerges at the state legislative/congressional district level, however. Only three women have been elected to the United States Congress in the state's history. Based upon the number and percentage of women who have served in the North Carolina General Assembly, the Center for Women and Politics ranks the state as thirty-third out of fifty states in electing women to legislative office. Currently, only thirty-two, or about 19 percent, of the seats in the North Carolina General Assembly are held by women.6 Given the different electoral climates that characterize statewide and district elections, it seems as though being a woman is more of a electoral liability in district races than in statewide contests. Therefore, this study appropriately selected the North Carolina District Court races to assess whether women, running at the judicial district level, fared as well as their political counterparts.
The district court races held between 1994 and 1998 netted impressive results. Sixty-eight contested races occurred over these three general election cycles. Fortythree women competed for thirty-four seats; these women ran as incumbents, as challengers, and as open-seat candidates; and they comprised approximately 50 percent, 37 percent, and 63 percent of the district court candidates who competed in the 1994, 1996, and 1998 general elections, respectively. Moreover, these elections permit comparative analyses of races involving women candidates only, of races involving men candidates only, and of races involving contests between men and women. Sixteen women ran in women v. women races; sixty-six men competed in men v. men races; and twenty-seven men competed against twenty-seven women in women v. men races. Another benefit accrued from these district court races was that, like their male counterparts, women candidates ran as nominees of both the Democratic Party as well as the Republican Party. Twenty-three women candidates ran as Democrats, and twenty women ran as Republicans; forty-five Democrat men and forty-eight Republican men ran in these contests.
Therefore, the women who vied for seats on the North Carolina District Court represent a good pool of candidates to explore whether a candidate's sex has a significant impact on judicial campaigns and elections. Despite the methodological concerns of generalizing from the races held in a single state, this study nevertheless can complement and supplement existing judicial selection, gender politics, and political elections research by exploring if the findings of Burrell or Herrick best characterize the competitiveness of this group of women judicial candidates. In other words, for the first time, this study examines empirically whether women judicial candidates faced no discernible discrimination or whether they encountered significant discrimination in the pursuit of seats on the North Carolina District Court.
CANDIDATE'S SEX, MONEY, AND ELECTORAL RESULTS
An overview of the aggregate data reported (see Table 1) corresponds to Burrell's findings on women congressional candidates. As a group, the women who ran in the 1994 to 1998 general elections for the district court reported more money in total receipts that men district court candidates, and these women used their collected funds to outspend the men in promoting their candidacies.7 There was an approximately $6,500 and $6,800 difference between the amount of money that women and men raised and spent during these three election cycles. Furthermore, women in every candidate category were more adept at funding their district court campaigns than men. Women incumbents outpaced men incumbents; women challengers surpassed men challengers; and women open-seat candidates outpaced men open-seat candidates. Therefore, if aggregate data of contributions and expenditure indicate electoral competitiveness, then women, as a group or as incumbents, as challengers or open-seat candidates, were as competitive as their male counterparts in pursuing seats on the North Carolina District Court.
In addition, women were able to attract a broad array of political actors to fund their campaigns. Although PAC participation varied over the three general elections, their contributions largely went into the coffers of women candidates, who, as a group, received more than 55 percent of all PAC dollars. Indeed, approximately one-half of the women candidates received financial support from at least one PAC, while almost three-quarters of the men running for the district court reported no PAC contribution. Most of the men who were PAC recipients were Democrats, while PACs contributed fairly equally to Democratic and Republican women. Unsurprisingly, incumbents received the lion's share of PAC money, and yet, somewhat surprisingly, women incumbents were the main recipients of PAC contributions. Interestingly, contrary to some research on women political candidates (Fox, 1997), women in these races seemed to have access to traditional funding networks because PACs created by businesses, labor unions, and issue groups all contributed to women candidates. For example, although business PACs favored men over women, they nevertheless gave twenty-one men an average contribution of approximately $215 and gave ten women an average contribution of approximately $395. Finally, women had funding opportunities to which men had little or no access because groups such as the North Carolina Women Attorneys Association contributed almost exclusively to women candidates.8
State political parties also supported, to varying degrees, their male and female nominees for district court seats.9 Men received the largest percentage of all party money distributed to district court candidates. Yet, women, as a group, were more likely to get financial help from their political parties than men; 55 percent of the women running in these district court races got party money and less than 30 percent of the men candidates were similarly funded. Interestingly, the state Democratic Party and state Republican Party adopted different funding strategies. Whereas the Democrats financially supported roughly the same number of men and women and provided them, on average, with approximately the same size contribution, the Republicans were less equitable. Only seven of the twenty Republican women reported at least one party contribution, but twenty of the forty-five Republican men were beneficiaries of party support; however, the men and women recipients, on average, reported similar contributions. Democrats also provided slightly more generous contributions than Republicans. On average, Democratic men and women outpaced their Republican counterparts by $115 and $106, respectively. Approximately 65 percent of the funded Democrats won their seats, while approximately 60 percent of the Republican recipients lost their contests.
The aggregate data support Burrell's findings. The findings cited here suggest that, as a group, women who pursued district court seats between 1994 and 1998 seemed to experience no notable discrimination in garnering financial support for their campaigns. Given the exclusivity of some funding sources, the proclivity of PACs to contribute to women and the availability of party funds, women candidates appeared to be in a very good position to parlay their financial gains into electoral gains.
Yet the electoral results, using the percentage of the vote garnered by candidates as the measure,10 hardly comport with the large differences in total receipts, total expenditures, PAC contributions, and party contributions reported by men and women. Marginal differences in average vote percentages separated men and women as groups, as well as men and women who ran in different candidate categories. Considering the funding disparities between men and women, however, women would be expected to garner more impressive vote tallies than they did. The significance of the disparity between women's funds and their vote percentage becomes more pronounced over the three election cycles. In the 1994 election, women outspend men by an average of over $5,600, and they garnered, on average, 48.5 percent of the vote; in the 1996 election, men were outspent by almost $3,000, but their average vote percentage was 50.17 percent; and in the 1998 election, women's vote percentage exceeded men's by 1.88 percentage points perhaps because they outspent men by more than $12,000. But, to be honest, close to 54 percent of the women candidates won their races, while about 52 percent of the men were losers between 1994 and 1998. Therefore, though not as impressive as campaign finance data, these electoral results alone fail to show, as Herrick's findings would argue, that, like women congressional candidates, women district court candidates were confronted with insurmountable or significant electoral barriers when they ran in the North Carolina District Court races.
Candidate's Sex in District Court Races
Aggregate comparisons of all men and women candidates present an interesting yet partial picture of the North Carolina District Court races. As Herrick argued, the preceding findings fail to take into account the realities of pursuing elective office: candidates do not campaign in isolation; instead, they compete against other candidates. In the three general elections held between 1994 and 1998, twenty-seven district court races pitted women candidates against men candidates and forty-one races involved two candidates of the same sex. Sixteen women judges were challenged by twelve men and four women while twenty-five men and eight women tried to unseat thirty-three men judges. There were nineteen open district court seats: seven involved men v. women races, four involved women v. women races, and eight involved men v. men races.
Recognizing that every election is different and that each election is likely affected by a wide range of variables, Table 2 nevertheless shows the distinctiveness of men v. women district court races. First, women, as a group, lost some of their financial dominance. Women raised and spent less money when they competed against men, and men raised and spent more money when they competed against women. Political parties and PACs did not defect from women candidates because women received substantial support from these actors. Yet individual contributors seemed to be more reluctant to fund women when they were opposed by men. Women, on average, received more than $6,000 in individual contributions than men, but, in men v. women contests, only $1,200 separated women from their male opponents.
Second, a mixed picture emerges from the electoral results of these races. The average vote percentages for men and women who competed in men v. women races were essentially equal; men, as a group, garnered 49.94 percent and women, as a group, received 49.76 percent. These aggregate numbers mask an interesting picture. Over the four-year period, women have not fared as well as the aggregate numbers might suggest. In 1994 women lost to their male competitors by an average vote margin of 4.5 percent; yet they outspent their opponents by more than $2,600 (average). The funding disparity between men and women essentially disappeared in the 1996 election. Men maintained their apparent vote-getting ability because they, as a group, received close to 51 percent of the vote. The 1998 election revealed an electoral victory for women because they, as a group, beat their male competitors by an average vote margin of 3.6 percent. However, the men were outspent by over $6,000.
Therefore, these findings parallel Herrick's conclusions about women congressional candidates. When comparing women with their male opponents, women lost their financial advantage because they were unable to attract significantly more contributions than their male competitors. If campaign finance is an indicator of electoral competitiveness, as Burrell argues, then women were notably less competitive when they ran against men. Disparities between women's finances and their vote percentages largely favored their male competitors. Though not definitive, the aggregate findings suggest that, in these district court races, women fared less well in campaign finance and vote acquisition when they competed against men.11
Multivariate analysis was used here to determine whether the findings presented in the research of Burrell (women face no discrimination) or Herrick (women are disadvantaged) best characterized the impact of a candidate's sex in the North Carolina District Court races. Past research on political elections was particularly helpful in identifying variables that have been found to have a significant impact on campaign funding and votes. Several were used as independent variables here: gender (candidate's sex), incumbency, political party affiliation, candidate's expenditures, and opponent's expenditures.12
Party strength, electoral contest, and primary opposition were included as independent variables. Party strength was used as a refinement of the political party variable. The impact of a voter's perception of a candidate's sex or a sex stereotype might be offset if the man or woman is affiliated with a political party that has strong electoral support within the state (Herrick, 1996). Party strength constituted the average vote percentages of all Democratic and Republican candidates competing in the 1994, 1996, and 1998 statewide elections (United States president, United States Senate, North Carolina governor, North Carolina Court of Appeals and North Carolina Supreme Court).13 The inclusion of electoral contest reflects the preceding findings as well as prevailing research. It was used to assess whether candidate's contribution totals and vote percentage were significantly affected by the type of race. Studies have shown that the sex of one's opponent affects how campaigns are mounted and how campaign strategies are formulated and implemented. Similarly, races pitting men against women force voters to choose between candidates of different sexes, thereby increasing the likelihood that voters might deliberately or unwittingly draw upon sex perceptions in their voting.14 Primary opposition was also included in the regression model because this study assumes that winning a primary might accrue several benefits to candidates competing in general elections. One, candidates are likely to be very aggressive fund raisers because they recognize that they need large campaign coffers to compete effectively in two races. Two, candidates acquire electoral experience during a primary contest that gives them an advantage over candidates who never sought elective office. Three, candidates who win primaries increase their name recognition that, consequently, improves their electoral chances in the general election. Of course, primary opposition can undermine a general election campaign if the winning candidate is unable to reenergize his or her campaign to confront a general election challenge.15
In addition, this study further adopts Herrick's methodology by creating three interactive variables (Herrick, 1996:72). These interactive variables were created by multiplying the gender variable by each of the independent variables of incumbency, electoral contest, and primary opposition. Incumbency is believed to correlate positively and significantly with both the acquisition of campaign funds and votes; the advantage of having an interactive variable of gender and incumbency is that it allows us to test the combined effect of a candidate's sex and incumbency upon contribution total and vote percentages. The second interactive variable resulted from combining electoral contest with gender. Since Herrick concluded that the competitiveness of women diminished in different sex races, her conclusion is tested by creating a variable that focuses upon a candidate's sex within the context of a particular set of electoral contests (men v. women races). If Herrick's findings are applicable to these district court races, then the effect will negatively and significantly correlate with the dependent variables. Finally, the third interactive variable measures the combined effects of gender and primary opposition. This interactive variable permits us to see if the combination of being a woman and having primary opposition affects a candidate's success in acquiring financial and electoral support during the general election.16
Table 3 presents a conflicting picture of the impact of a candidate's sex in the North Carolina District Court races. The findings show that gender correlates significantly and positively with campaign contribution totals. This means that women candidates, indeed, had a financial advantage over men in running for the district court. Yet the findings also show that women candidates were unable to parlay that financial advantage into tangible electoral results because candidate sex had no statistically significant impact upon the percentage of the vote candidates acquired in these elections. The percentage of votes held a positive and statistically significant correlation with incumbency, as well as with candidate spending. Overall, the statistically significant findings highlighted in Table 3 support Burrell's conclusion about the impact of gender on campaign contributions. Just as women congressional candidates, women district court candidates experienced no apparent discrimination in finding financial support for their campaigns. At the same time, findings in Table 3 also provide some evidence to back Herrick's supposition that, despite their competitiveness in campaign finance, women candidates lost their competitive edge in convincing voters to support them in elections. The same phenomenon characterized these trial court races because these women candidates, too, failed to produce statistically identifiable electoral gains from their financial gains.
Interestingly, the effect of the interactive variables on contributions and vote percentages places a less positive gloss on the strength of women candidates in these district court races.17 The dual effects of being a woman and a sitting district judge are strikingly different than being a woman or running as an incumbent. Though not statistically significant, the inverse correlation between the interactive variable of gender/incumbency and both the collection of money and the acquisition of votes nonetheless suggests that women judges seems to gain little electoral bang from their incumbency. A less profound yet noteworthy finding emerged from the effects of gender and electoral contest upon the dependent variables. Here again, the finding was not statistically significant; yet its correlations skewed against being a woman and being involved in a different-sex race. Although being a woman or participating in a man v. woman race separately had a positive effect on acquiring campaign funds and garnering votes, their combined effects surprisingly produced negative correlations. The implication of this result is that, in comparison to men and other women candidates, these women might be less able to attract financial and electoral support for their candidacies when they compete directly against men.
Observations from the Findings
Four broad observations may be discerned from the results of the aggregate comparisons and multivariate analysis. The first is that women candidates, as a group, were especially successful in funding their campaigns for the North Carolina District Court. In comparison to the men running in this set of trial court races, the women reported impressive contribution totals, and they used their monies to mount comparatively expensive campaigns.
The second observation is that, despite their impressive contribution totals, women failed to garner commensurately impressive electoral results. There is no doubt that women candidates won seats on the district court, but their margins of victory rarely matched their expenditures. To state this observation more bluntly, men candidates received essentially the same level of electoral support as women while reporting significantly less in campaign expenditures. It appears that, in these races, women were very good at getting money from a network of traditional contributors but less effective in convincing a large percentage of the electorate to support their candidacies.
The third observation highlights the distinctiveness of men v. women races. When women competed directly against men, they, as a group, were less successful both at raising funds and at amassing commensurately impressive vote tallies. The funding gap separating men and women narrowed significantly in races where they vied for the same district court seats. Just as women lost their financial advantages, they also were less successful at the polls. Since North Carolina voters have been reluctant to support women running at the local or district level, their reluctance might explain why women district court candidates received the same level of financial support from political actors, but less from individual donors and why, in comparison to their male opponents, women were unable to parlay their financial advantages into stronger electoral gains.
The fourth observation underscores the complexities and idiosyncracies of district court races.18 District court races are essentially local races that are likely to be affected by a wide range of factors. The low salience that characterizes most judicial elections invites voters to make electoral decisions based upon a variety of influences that might be more reflective of personal predilections or local culture. Given the low salience of judicial elections and low voter knowledge, the decision to contribute to a district court campaign and the decision to vote for a particular district court candidate may depend upon the idiosyncracies and eccentricities of the political and social culture within the judicial district.
The broad purpose of this study was to assess whether a candidate's sex affected the campaigns and elections of the men and women who ran for the North Carolina District Court. Accordingly, the sex of a candidate seemed to play an identifiable, though perhaps not a determinative, role in the general elections for the North Carolina District Court. The sex of a candidate appeared to be influential in campaign finance. It is clear that women candidates, as a group, either experienced little (or no) discrimination or had an advantage in convincing traditional funding sources to support their campaign efforts. Although one could argue that being a woman might be an asset on the finance side of electoral politics, one must nonetheless concede that women might have encountered more challenges in convincing voters to back their candidacies for the state's minor trial court. Yet it is difficult to demonstrate definitively that women candidates encountered bias or discrimination at the polls in large part because women won more district court seats than they lost.
Recognizing the methodological concerns about generalizing from a relatively small set of races that were held in only one state, this study nevertheless contributes to the existing research on judicial selection and gender politics. It is hoped that its findings will serve as a springboard for additional research that will explore whether women compete equally with men in judicial elections or whether women must overcome sex-based challenges in their pursuit of elective state judgeships.
1 Currently, twenty-nine states use partisan or nonpartisan elections to select their judges; nine of these states use partisan or nonpartisan elections to select their trial court judges only. See American Judicature Society Web site: http://www.ajs.org/js/JudicialSelectionCharts.pdf. All information about state court selection systems can be found at this Web site.
2 The United States Supreme Court decision in Republican Party of Minnesota v. White, 536 U.S. 765 (2002), may lessen or eliminate these ethical constraints. In this ruling, the Court held that, like other candidates, judicial candidates have a First Amendment right to express their views on a wide range of issues. Recently, the North Carolina Supreme Court revised the state code of judicial conduct to allow judicial candidates running in elections to discuss issues as freely as candidates seeking political office. See, "N.C. Judicial Candidates Given More Freedom to Discuss Issues," 2003.
3 Actually, North Carolina has thirty-nine judicial districts for jurisdictional purposes. One of these districts has been divided into two districts for electoral purposes. Therefore, district court judges compete in one of forty judicial districts, but operate as judges in one of thirty-nine jurisdictional judicial districts.
4 Beginning in 2002, North Carolina used nonpartisan elections to select judges for its district court.
5 In 2002 the North Carolina General Assembly instituted a system of public financing for appellate court elections. The Judicial Campaign Reform Act, which will be in effect for the 2004 general elections, created a voluntary system for candidates competing for seats on the North Carolina Court of Appeals and North Carolina Supreme Court. Participating candidates agree to fund-raising maximums, which are based upon the judgeships' salaries, to finance their primary races, and they agree to accept public funds for the nonpartisan general elections. The amount of public money that candidates will receive is based upon the judgeship salaries for the state's intermediate appellate court and the state supreme court. For example, at the time of this writing, a state supreme court justice makes about $200,000; therefore, a can-, didate who volunteers to participate in the new public finance program will receive about $200,000 in public funds. The public funds largely are collected from a voluntary checkoff on individual state income tax forms and from a voluntary $50 contribution from attorneys when paying their privilege license taxes (see Dyer, 2002:B1).
6 See http://www.rci.rutgers.edu/~cawp/facts/stbySt/NC.html, Center for American Women and Politics in the Eagleton Institute of Politics at Rutgers University.
7 Campaign finance data for the district court candidates were culled from campaign finance reports filed at the North Carolina State Board of Elections. Candidates are required to disclose not only their cumulative contributions and expenditures but also an itemization of their receipts and spending. Beginning in 2000, North Carolina election law required candidates to include the occupations of contributors in their campaign finance reports. This requirement was not in effect during the years this study covers: 1994, 1996, and 1998.
Since the contributions and expenditures used here come from campaign finance reports by district court candidates, they include primary contests and general elections. Because of North Carolina's reporting requirements and administrative procedures during this period, there is no reliably consistent and systematic method of distinguishing expenditures and contributions for primary campaigns from those of general election campaigns. Therefore, the study could not separate contributions and expenditures for primary races from those of general elections.
However, the author recognizes that overcoming opposition in a primary might affect a general election candidate's fund raising and vote getting. This study identified general election candidates who won contested primaries. A dummy variable of "primary opposition" and "no primary opposition" was added to the multivariate models. This method should allow us to test whether successfully overcoming primary opposition has a statistically significant effect on total contributions and percentage of the vote garnered by district court candidates and whether the men and women who competed in these primary contests fared differently in the general election.
Twenty-five district court candidates faced primary opposition-eleven women and fourteen men. These candidates had higher mean contributions and expenditures than candidates who had no primary opposition, but, on average, the men and women who won their primaries reported similar contributions and expenditures. Less than $2,500 and $3,500 separated them in mean contributions and mean expenditures.
8 Of the forty-three women seeking district court seats, twenty received contributions from women's groups. Only fourteen of ninety-three men reported at least one contribution from these groups. Most of the recipients were Democrats. Forty-nine district court candidates (twenty-three women and twenty-six men) also reported at least one contribution from a PAC.
9 Candidates may accrue two types of benefits from a political party. One advantage stems from the coordination of a candidate's campaign with the party organization. These joint activities allow candidates to draw upon the party's organization, fund-raising contacts, and electoral base to assist their campaigns. Many candidates reported these coordinated efforts as in-kind contributions or expenditures. In addition to coordinated activities, political parties also contribute money directly to the candidate. The figures reported here include direct contributions and only those coordinated activities reported as in-kind contributions; coordinated activities listed as expenditures on campaign reports were excluded from tallies of total receipts.
10 The percentages of the vote received by the district court candidates who competed in contested races in the 1994 to 1998 general elections were acquired from the North Carolina State Board of Elections.
11 Interestingly, the impact of incumbency was blunted for women judges and men judges in different-sex contests. As a group, incumbents outspent their challengers, and they largely retained their judgeships. However, incumbents of both sexes outspent their different-sex challengers by approximately $6,000, while their colleagues spent more than $13,000 (average) against their same-sex challengers. The same phenomenon characterized the voting results in these races. Winning incumbents who competed in different sex races won by an average vote margin of 2 percent; their colleagues, fighting back same-sex challengers, retained their seats by an impressive 7 percent vote margin, on average.
12 The variables gender, incumbency, and political party affiliation were coded as dichotomous variables: gender was coded as 1 for a woman candidate and O for a man candidate; candidates were coded 1 for being an incumbent or O for a nonincumbent; and a candidate's political party affiliation was coded as 1 for Republican and O for Democrat. Expenditures for candidates and their opponents include all expenses that were reported on all campaign finance reports filed by candidates competing in contested races in the 1994, 1996, and 1998 general elections. Expenditures were measured in thousands of dollars.
13 Statewide election results were used to calculate party strength because district court districts do not all correspond to county lines nor do they dovetail with congressional or state legislative district lines. Like many of its neighbors, North Carolina is a competitive two-party state. In 1998, mean percentages of the vote for Democrats and Republicans in statewide elections, for United States Senate, two North Carolina Supreme Court seats, and five North Carolina Court of Appeals seats, were 49.89 and 50.11. In 1996 Democratic candidates garnered 51.40 percent of the vote and Republican candidates received 48.15 percent for statewide elections for United States president, United States Senate, North Carolina governor and lieutenant governor, two North Carolina Supreme Court seats, and one North Carolina Court of Appeals seat. The only statewide elections in 1994 were two North Carolina Supreme Court races and two court of appeals races; the mean vote percentages for Democrats and Republicans were calculated by combining the 1994 judicial races with the 1992 statewide political elections for United States president, United States Senate, and North Carolina governor. Therefore, party strength for 1994 district court candidates was 47.34 percent for Democrats and 52.64 percent for Republicans.
Unlike congressional elections research, this study did not use "judicial district" as an independent variable for methodological and practical reasons. First, as stated above, North Carolina judicial districts do not correspond to any other electoral districts. second, access to voting data is limited because electoral/voting data from precincts are classified, and precincts do not neatly fit into judicial districts. Third, demographic information cannot be easily attained because judicial districts can cross county and city boundaries. Fourth, consistent patterns of electoral behavior in judicial elections are hard to discern because, unlike political seats, which are generally contested on a regular basis, challenges to specific district court seats within a single judicial district vary widely. In some districts, a number of seats are challenged at every election; in contrast, a significant proportion of the seats are rarely challenged. This inconsistency can cloud the analysis.
14 The electoral contest variable was a dummy variable; candidates who ran in races pitting women against men were coded as 1 and candidates who competed in other races were coded as O.
15 For our regression model, primary challenge was coded as a dummy variable with 1, indicating candidates who survived a contested primary before running in the general election for the district court; the other candidates were coded as O.
16 All three interactive variables are dummy variables. Gender/incumbency creates an independent in which being a woman and an incumbent is coded as 1 and other combinations are coded as O. Gender/electoral contest creates a dummy variable in which being a woman running in a man v. woman race is coded as 1 and other combinations are coded as O. Gender/primary opposition creates an independent variable in which being a woman who faced primary opposition is coded as 1 and other combinations are coded as 0.
17 Different results emerged when only women and their opponents were included in multiple regression analysis. None of the independent variables were statistically significant with contribution totals. Since gender no longer significantly correlates with contribution totals, women lost their financial advantages when they compete against men in these races. In contrast, several variables have a statistically significant impact on the dependent variable vote percentage. Perhaps the most interesting finding is that the importance of incumbency, though still significant, was nevertheless diminished in these races. Here, too, being a woman and an incumbent correlated negatively. The findings presented in Table 3 suggest that women who competed directly against men for seats on the North Carolina District Court encountered a different electoral climate than candidates running in other races (see Appendix).
18 Despite the findings presented here, other variables not included in our model might further explain the district court candidates' contribution totals and vote percentages: work experience, name recognition, and length of residency in the district. Arguably, work experience as police officers, district attorneys, or family law attorneys might be an important credential for voters selecting candidates to a court that hears criminal cases as well as family cases. Furthermore, there is no reason to assume that name recognition would be less significant in district court races than in political races. Name recognition for district court candidates may be measured not only by traditional indicators, such as media buys, but also by nontraditional measures, such as membership in civic, professional, religious, recreational, and other community groups. Candidates who have been active in a wide range of organizations throughout the district are likely to be known, or at least have their name recognized, by a sizable number of potential contributors and voters. In addition to work experience and name recognition, the length of time candidates have lived in the district may also have a positive effect on their ability to marshal money for their campaigns and votes for their election. Indeed, according to political and judicial elections research, there is a positive "friends and neighbors effect" for local candidates running in statewide or multidistrict elections (Aspin and Hall, 1987; Rice and Macht, 1987).
Aloize, N. O. (1996). "Selection Methods and the Recruitment of Women to the State Courts of Last Resort," 77 Social Science Quarterly 110.
American Judicature Society. "Judicial Selection Chart." Online at http://www.ajs.org/js/JudicialSelection.
Aspin, L. T., and W. K. Hall (1987). "The Friends and Neighbors Effect in Judicial Retention Elections," 40 Western Political Quarterly 703.
Baum, L. (1987). "Explaining the Vote in Judicial Elections: The 1984 Ohio Supreme Court Elections," 40 Western Political Quarterly 361.
Bernstein, A. (1998). "Campaign 98: Following the Money Trail in Judicial Race," Houston Chronicle, September 13, p. 37A.
Borreson, S. (1998). "End of the Stone Age? Kathy Stone, the Only Incumbent Democratic State District Judge in Harris County Seeking Re-election, Hopes Her Experience Outweighs Party Affiliation," Texas Lawyer, September 14, p.1.
Burrell, B. (1998). "Campaign Finance: Women's Experience in the Modem Era," in S. Thomas and C. Wilcox (eds.), Women and Elective Office. New York: Oxford University Press.
_____ (1994). A Woman's Place h in the House. Ann Arbor: University of Michigan Press.
_____ (1985). "Women's and Men's Campaigns for the U.S. House of Representatives, 1972-82," 13 American Politics Quarterly 251.
Carp, R. A., R. Stidham, and K. C. Manning (2004). Judicial Process in America. Washington, DC: Congressional Press, 2004.
Carroll, S. J. (1994). Women as Candidates in American Politics, 2d ed. Bloomington: Indiana University Press.
Center for American Women and Politics. "Women in Elective Office." Online at http://www.rci.rutgers.edu/~cawp.
Champagne, A., and K. Cheek (1996). "PACs and Judicial Politics in Texas," 80 Judicature 26.
Champagne, A., and J. Haydel (1993). Judicial Reform in the States. Lanham, MD: University Press of America.
Conover, P. (1981). "Political Cues and Perceptions of Candidates," 9 American Politics Quarterly 427.
Cook, E. A., S. Thomas, and C. Wilcox, eds. (1994). The Year of the Woman: Myths and Realities. Boulder, CO: Westview Press.
Darcy, R., S. Welch, and J. Clark, eds. (1994). Women, Elections and Representation. Lincoln: University of Nebraska Press.
Dubois, P. L. (1986a). "Financing Trial Court Elections: Who Contributes to California Judicial Campaigns," 70 Judicature 8.
_____ (1986b). "Penny for Your Thoughts? Campaign Spending in California Trial Court Elections, 1967-82," 39 Western Political Quarterly 264.
_____ (1980). From Ballot to Bench: judicial Elections and the Quest for Accountability. Austin: University of Texas Press.
Dyer, E. (2002). "Lawmakers Approve Bill Altering Judicial Elections; Nonpartisan Publicly Financed Campaigns for Appellate Courts Could Start in 2004," News and Record, October 2, p. B1
Duersti-Lahti, G. (1998). "The Bottleneck: Women Becoming Candidates," in S. Thomas and C. Wilcox (eds.), Women and Elective Office. New York: Oxford University Press.
Eismeier, T. J., and P. H. Pollock III (1984). "Political Action Committees: Varieties of Organization and Strategy," in M. J. Malbin (ed.), Money and Politics in the united States. Chatham, NJ: Chatham House Publishers, Inc.
Fowlkes, D. L., J. Perkins, and S. T. Rinehart (1979). "Gender Roles and Party Roles," 73 American Political Science Review 772.
Fox, R. L. (1997). Gender Dynamics in Congressional Elections. Thousand Oaks, CA: Sage Publications.
Friedman, T. (1998). "The Politicization of the Judiciary," 82 Judicature 6.
Gertzog, I. N. (1995). Congressional Women: Their Recruitment, Integration and Behavior, 2d ed. Westport, CT: Praeger
Hall, M. G. (2001). "State Supreme Courts in American Democracy: Probing the Myths of Judicial Reform," 95 American Political Science Review 315.
_____ (1992). "Electoral Politics and Strategic Voting in State Supreme Courts," Journal of Politics. 54: 427-446.
Herrick, R. (1995). "A Reappraisal of the Quality of Women Candidates," 15 Women and Politics 25.
_____ (1996). "Is There a Gender Gap in the Value of Campaign Resources?" 24 American Politics Quarterly 68.
Herrnson, P., J. C. Lay, and A. K. Stokes (2001). "Women Candidates: Shattering the Glass Ceiling?" 22 Campaigns anci Elections 8.
Hojnacki, M., and D. C. Kimball (2001). "PAC Contributions and Lobbying Contacts in Congressional Committees," 54 Political Research Quarterly 161.
Huckfeldt, R., and J. Sprague (1992). "Political Parties and Electoral Mobilization: Political Structure, Social Structure, and the Party Canvas," 86 American Political Science Review 70.
Jones, R. S. (1984). "Financing State Elections," in M. J. Malbin (ed.), Money and Politics in the United States. Chatham, NJ: Chatham House Publications, Inc.
Leeper, M. (1991). "The Impact of Prejudice on Female Candidates: An Experimental Look at Voter Inference," 19 American Politics Quarterly 248.
Lovrich, N. D., Jr., C. H. Sheldon, and E. Wassman (1988). "The Racial Factor in Nonpartisan Judicial Elections: A Research Note," 41 Western Political Quarterly 807.
Mahtesian, C. (1998). "Bench Press," Governing, August, p. 17.
Maisel, L. S. (1990). "The Incumbency Advantage," in M. L. Nuguent and J. R. Johannes (eds.), Money, Elections and Democracy. Boulder, CO: Westview Press.
Martin, E. (1999). "Women Judges: The New Generation," in L. D. Whitaker (ed.), Women in Politics: Insiders or Outsiders? 3d ed. Upper Saddle River, NJ: Prentice Hall.
Moog, R. S. (1992). "Campaign Financing for North Carolina Appellate Courts," 76 Judicature 68.
"N.C. Judicial Candidates Given More Freedom to Discuss Issues" (2003). Associated Press, April 11.
Newfield, J., and C. Miner (2003). "New Indictments Handed Up in Brooklyn," New York Sun, November 13, p.1.
Nicholson, M. A., and N. Nicholson (1994). "Funding Judicial Campaigns in Illinois," 77 Judicature 294.
Reid, T. V. (2000). "The Role of Gender in Judicial Campaigns: North Carolina Trial Court Races," 28 Southeastern Political Review 551.
_____ (1999). "The Politicization of Retention Elections: Lessons from the Defeat of Justices Laphier and White," 83 judicature 68.
_____ (1996). "PAC Participation in North Carolina Supreme Court Elections," 80 Judicature 21.
Republican Party of Minnesota v. White (2002). 536 U.S. 765.
Rice, T. W., and A. A. Macht (1987). "Friends and Neighbors Voting in Statewide General Elections," 31 American ]oumal of Political Science 448.
Richey, W. (2000). "Justice for Sale? Cash Pours Into Campaigns," Christian Science Monitor, October 25, p. 2.
Sabato, L. (1990). "PACs and Parties," in M. L. Nuguent and J. R. Johannes (eds.), Mone;y, Elections and Democracy. Boulder, CO: Westview Press.
_____ (1984). PAC Power: inside the World of Political Action Committees. New York: W.W. Norton and Company.
Sapiro, V. (1981-82). "If Senator Baker Were a Woman: An Experimental Study of Candidates' Images," 3 Political Psychology 61.
Schotland, R. (1985). "Elective Judges' Campaign Financing: Are State Judges' Robes the Emperor's Clothes of American Democracy?" 57 Journal of Law and Policy 57.
Schreiner, M. (2003). "McCain Touts State's Judicial Election Reform," Star News, October 21, p. IB.
Selzer, R. A., J. Newman, and M. V. Leighton (1997). Sex as a Political Variable. Boulder, CO: Lynne Rienner Publishers, Inc.
Smith, B. (2002). "Symposium on Judicial Elections: Selecting Judges in the 21th Century," 30 Capital University Law Review 437.
Smith, C. E. (1997). Courts, Politics, and the Judicial Process, 2d ed. Chicago: Nelson-Hall Publishers.
Smith, E. R. A. N., and R. C. Fox (2001). "The Electoral Fortunes of Women Candidates for Congress," 54 Political Research Quarterly 205.
Solimine, M. (2002). "The False Promise of Judicial Elections in Ohio," 30 Capital University Law Review 559.
Sorauf, E (1992). inside Campaign Finance: Myths and Realities. New Haven, CT: Yale University Press.
_____ (1988). Money in American Elections. Glenville, IL: Scott, Foresman.
Thielemann, G. S. (1993). "Local Advantage in Campaign Financing: Friends, Neighbors, and Their Money in Texas Supreme Court Elections," 55 Journal of Politics 472.
Uelman, G. F. (1997). "Crocodiles in the Bathtub: Maintaining the Independence of State Supreme Courts in an Era of Judicial Politicization," 72 Notre Dame Law Review 1133.
Uhlaner, C. J., and K. L. Scholzman (1986). "Candidate Gender and Congressional Campaign Receipts," 48 Journal of Politics 30.
Werner, B. L. (1997). "Financing the Campaigns of Women Candidates and Their Opponents: Evidence from Three States, 1982-1990," 18 Women and Politics 81.
Westlye, M. C. (1991). Senate Elections and Campaign Intensity. Baltimore: Johns Hopkins University Press.
Wohl, A. (2000). "Justice for Rent: The Scandal of Judicial Campaign Financing," American Prospect, May 22, p. 34.…