Picking Judges: How Judicial-Selection Methods Affect Diversity in State Appellate Courts


Vol. 101 No. 1 (2017) | Citizen-centered Courts

In the beginning, judges in the 13 original states either were appointed by the governor or selected by the legislature. Over the next 80 years, however, a majority of states turned to popular elections to choose their judges. Today, voters in 22 states elect their appellate judges. Two states leave the selection of appellate judges to the legislature, and in the remaining 26 states, the governor appoints members of the appellate courts.

Some have argued that appointive systems are more likely to produce judges with better education and more expertise. But critics of appointment by “merit selection” argue it disadvantages women, minorities, and others with nontraditional legal backgrounds who may be unable to muster support from those who can influence the nominating commission or the governor.2 Others contend that elections produce judges who are more responsive, accessible, and accountable.3

Although judicial quality is virtually impossible to measure, considerable research has sought insight into the objective characteristics of the judges chosen by the various judicial-selection systems. Most of this research has focused on the gender and race or ethnicity of the judges selected by the various methods. Although the prevailing view is that the particular judicial-selection method does not significantly affect the rates at which women and minorities together are chosen for the appellate bench,4 data continue to show that merit selection advances proportionately fewer women than other judicial-selection methods.5

I undertook to study the results of the various state judicial-selection methods using a much wider range of information about judges on state appellate benches. Although their subjective differences cannot be quantified, I analyzed several objective attributes of each judge sitting on a state appellate-court bench in February 2015, including gender, race, age, and the nature of the judge’s prior legal experience, as well as arguably objective credentials such as judicial clerkships and attendance at ranked universities and law schools. The sources of the information for the study were wide and varied: Official court websites, the website “Ballotpedia” (formerly known as “Judgepedia”), gubernatorial press releases, candidate websites, law school and university publications, Westlaw, and online versions of local newspapers. The data seem to show that certain judicial-selection methods do tend to favor certain characteristics among judicial candidates, and that merit selection seems to disadvantage women in private practice.

"The Data" All Judges By State Selection Method

Click Here to download an accessible PDF of this article with tables.

Categorizing the Selection Methods

Before examining the resulting data, let us pause to consider the legal mechanics by which the various states select their appellate judges:

Partisan election. Voters in six states — Alabama, Illinois, Louisiana, New Mexico, Pennsylvania, and Texas — elect their appellate judges in overtly partisan elections. I also included judges from Ohio in this category. Although Ohio judicial candidates run in ostensibly nonpartisan general elections, they are nominated in partisan primary elections, and substantial evidence shows that partisanship predominates in the general elections. In all of these states, however, many appellate judges initially come to the bench when they are appointed to fill a midterm vacancy that occurs upon the death, resignation, retirement, or removal of an incumbent.

Nonpartisan election. In 15 states — Arkansas, Georgia, Idaho, Kentucky, Michigan, Minnesota, Mississippi, Montana, Nevada, North Carolina, North Dakota, Oregon, Washington, West Virginia, and Wisconsin — voters elect members of the appellate bench in nonpartisan elections.6 Midterm judicial vacancies in nonpartisan states are filled by gubernatorial appointment.

Legislative selection. Two states, South Carolina and Virginia, select their appellate judges by way of legislative election.

“Merit” selection. Eleven states — Alaska, Arizona, Colorado, Florida, Indiana, Iowa, Missouri, Nebraska, Oklahoma, South Dakota, and Wyoming — use the classic version of so-called “merit” selection to choose their appellate judges. Each of these states has an appointed, independent, bipartisan commission made up of lawyers and non-lawyers, which screens applications, vets and interviews applicants, and then delivers a list of nominees to the governor for appointment. In nearly all merit-selection states, the governor must appoint one of the candidates nominated by the independent commission. In each merit-selection state, an appellate judge stands for retention at an up-or-down election at some point after appointment, and at regular intervals thereafter.

This study includes Kansas in this category because each of its appellate judges sitting in 2015 was chosen by merit selection. In 2013, however, Kansas abandoned use of a constitutional nominating commission for court of appeals judges. The governor now appoints members of that bench subject to confirmation by the state senate.

I also count California as merit-selection state because it uses a variant of the traditional merit system: Applicants for the bench in California are reviewed by an independent commission created by statute and made up of lawyers and nonlawyers appointed by the California State Bar Association. Although the commission’s ratings do not formally constrain the governor’s appointment prerogative, over the years, the governor has appointed only a couple of appellate judges whom the commission found not qualified. After the governor has made his or her selection, the appointee must be confirmed by the Commission on Judicial Appointments, a body made up of the Chief Justice, the Attorney General, and the senior presiding justice of the court of appeal. Given its makeup, the Commission on Judicial Appointments seems unlikely to fail to confirm any appointment by the governor.

Finally, members of New York’s intermediate appellate court (but not the members of its highest court) are selected through a variation of merit selection. New York’s general jurisdiction trial court is called the Supreme Court, and its intermediate appellate court is called the Appellate Division of the Supreme Court. Under the state constitution, the governor appoints members of the Appellate Division from among the current justices on the Supreme Court. For many years, a Departmental Judicial Screening Committee, created by gubernatorial executive order, has evaluated and recommended candidates for all judicial positions in the Appellate Division. Under the executive orders, the governor may appoint to the Appellate Division only justices that the judicial screening committee has found to be “highly” or “well” qualified.

“Merit-confirmation.” In the remaining 12 states — Connecticut, Delaware, Hawaii, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, Rhode Island, Tennessee, Utah, and Vermont (and New York’s Court of Appeals) — the governor appoints appellate judges subject to the approval of another elected body.7

Earlier studies tended to treat in one category all states that use strict merit-selection protocols, and in another category, all other nonelection states in which the governor fills judicial vacancies by appointment.8 In characterizing the latter group as “appointment” states without further explanation, however, the prior literature did not acknowledge two significant constraints on the governor’s appointment power in those states.

The first constraint is that, in each of the 12 states, a judicial nominating commission must screen applicants before the governor may make the appointment.9 In roughly half the states, that commission is created by the state constitution, just as in traditional merit-selection states. In the remaining states, judicial nominating commissions have been created by a series of long-standing executive orders.

The second constraint is that in each of the 12 states (and with respect to appointments to New York’s Court of Appeals), the governor’s appointee may not take the bench unless and until he or she is confirmed by a separately elected body. Although the prior literature has not assessed how a confirmation requirement may constrain the gubernatorial appointment prerogative, at the federal level, the constitutional requirement that the Senate must give its “advice and consent” to any judicial nomination has significantly restrained the President’s power to appoint members of the bench: Through 2011, the Senate failed to confirm 36 of 160 presidential nominations to the United States Supreme Court. It stands to reason that in the states, the mere existence of a confirmation process will constrain to some degree a governor’s choice of whom to appoint. And that constraint likely will be more significant when the confirming body is of a different political party than the governor or does not in general reflect the governor’s judicial ideology.

Overview of the Results

The data allow a fresh look at the characteristics other researchers have studied and also allow analysis of a host of other objective characteristics, including more detailed information about the judges’ prior legal careers and academic credentials. Altogether, they show that merit-confirmation tends to produce the most distinctive appellate bench: Judges selected by merit-confirmation tend to have more years of legal experience and more judicial experience, and include more former prosecutors, more judges with other government-law experience, and more former judicial clerks than judges in other states. Judges in merit-selection states, by contrast, tend to bear objective characteristics roughly comparable to those in election states.

Generally, the characteristics of judges in election states do not differ significantly from those of judges in other states. To a large extent this is because so many of them first come to the bench by appointment. Those judges tend, by large margins, to bear the same characteristics of judges appointed through merit selection and merit-confirmation. As reported previously, merit selection disadvantages women. The data suggest that may be a consequence of the nature of women’s respective career paths.

As a caveat, although a study like this may reveal relationships, proof of cause and effect can be elusive. For one thing, the various states’ selection methods are not employed randomly across all regions of the United States. With important exceptions, judicial-election protocols predominate in the South and in the Great Lakes, while merit selection tends to be found in and around the West and the Midwest, and merit-confirmation protocols tend to cluster in the Northeast. For that reason, when a particular characteristic is more commonly found among judges chosen by a particular selection method, we cannot know whether that is because of the states’ respective selection methods, the preferences of their citizens, or the attributes of the judicial candidates who live in those states.

That being said, here is a broad snapshot of the demographics of the 1,285 appellate judges:

  • Average age at selection is 51
  • 35 percent are women
  • 15 percent are nonwhite

With respect to their career paths:

  • 83 percent have private-practice experience
  • 64 percent have prior judicial experience (for example, service on a state trial court)
  • 38 percent have prosecutorial experience
  • 25 percent have served as a government lawyer in a capacity other than as a prosecutor
  • 12 percent served in legal aid or as a public defender

As for their education and other scholarly credentials:

  • 73 percent went to a nonranked, in-state law school
  • 63 percent went to a nonranked, in-state undergraduate institution
  • 27 percent served a judicial clerkship
  • 26 percent have served on a college or law school faculty, either part time or full time, and either before coming on the bench or after
  • 15 percent went to a ranked undergraduate school
  • 13 percent published law review articles before taking the bench
  • 12 percent went to a ranked law school
  • 7 percent are elected members of the American Law Institute

There are many similarities in the attributes of the judges across the various benches, but some differences stand out:

Age: The data reveal statistically significant differences in the ages at which judges are chosen across the various selection methods. By a slight but statistically significant margin, judges in election states are younger than those in merit-selection, merit-confirmation, and legislative-selection states. Correspondingly, appointed judges have roughly two more years of legal experience than their elected counterparts.

Gender: A little more than 35 percent of the judges, across all selection methods, are women. Proportionately, more women are on the bench in partisan-election states than in states that use any other selection method. As noted, significantly fewer judges chosen by merit selection (29 percent) are women.

Race/Ethnicity: Overall, nearly 15 percent of the appellate judges in the dataset are nonwhite. No statistically significant differences appear across the various selection methods.

Prior Judicial Experience: Across all selection methods, 64 percent of the appellate judges have some prior judicial experience (on a trial court or, in the case of a judge on a state’s court of last resort, on an intermediate appellate court). By a statistically significant margin, however, appellate judges in merit-confirmation states are most likely to have prior judicial experience, at nearly 75 percent, and judges in nonpartisan election states are least likely to have prior judicial experience, at 58 percent.

Private Practice Experience: Judges in election states and those chosen through merit selection are statistically significantly more likely to have practiced in the private sector than judges in merit-confirmation states.

Former Prosecutors: Nearly 38 percent of the judges in the current dataset are former prosecutors. To a statistically significant degree, merit-confirmation states choose more former prosecutors for their appellate benches than states using any other judicial selection method. Partisan-election states follow not far behind.

Other Former Government Lawyers: Perhaps not surprisingly, given that so many lawyers practice in state government, judges whose appointments required confirmation by a government body or who were selected by the legislature are significantly more likely to have government-law experience.

Former Legal Aid/Criminal Defense Lawyers: On the average, 12 percent of all the judges have some experience in legal aid, poverty law, or criminal defense agencies. Only slight differences appear across the various selection methods.

Judicial Clerkships: Merit-confirmation puts the highest percentage of former judicial law clerks on the bench. Nearly 32 percent of all judges in the dataset who were selected through merit-confirmation served judicial clerkships; slightly more than 26 percent of judges appointed through merit selection clerked.

Education: Judges in election states have strong in-state education connections. Of judges in partisan-election states, 84 percent attended in-state law schools, as did 72 percent of judges in nonpartisan election states. Judges selected by merit-confirmation are the least likely to have done so (61 percent). Judges in partisan-election states also are the most likely to have attended an in-state undergraduate institution (75 percent).

Judges also were sorted according to whether they graduated from ranked law schools or undergraduate institutions.10 The proportion of judges who graduated from ranked law schools is not large (11.5 percent), but there are wide differences among selection methods. Larger proportions of the judges in legislative-selection (21 percent) and merit-confirmation (17 percent) states attended ranked law schools. By contrast, only 3 percent of the judges in partisan-election states did so. Judges sitting in partisan-election states also are significantly less likely to have attended a ranked undergraduate institution. But, by contrast to the results with law schools, judges selected by the legislature are least likely (not most likely) to have attended a ranked undergraduate institution. The highest proportion of judges who attended ranked undergraduate schools is seen in nonpartisan election states (19 percent), followed closely by judges selected by merit-confirmation and by merit selection.

Other Scholarly Qualities: The study also identified law review articles the judges wrote before they went on the bench. There are no significant differences in those numbers among the judges, by selection method. By a significant margin, however, judges in merit-confirmation states are more likely to have been elected to the American Law Institute; judges in partisan-election states are next most likely to be elected ALI members.

Merit Selection and Merit-Confirmation

The data show that the objective characteristics of the judges chosen by merit selection are not dissimilar in most respects from those chosen by all selection methods, taken together. Merit-selected judges are slightly less likely than the average to be former prosecutors and slightly less likely to have practiced other forms of government law. Although they are slightly more likely than the average to have attended a ranked law school, other selection methods boast a higher percentage of ranked law-school graduates. And proportionately far fewer merit-selected judges than the average are elected members of ALI.

But all selection methods do not produce judges with about the same mix of characteristics. Merit-confirmation, not merit selection, produces benches that are the most distinctive among all the selection methods. Judges selected by merit-confirmation have significantly more years of legal experience and are significantly older (by two years or more) than other judges. Significantly higher percentages of merit-confirmation judges served judicial clerkships, are elected members of ALI, and have taught in college or graduate school. A significantly greater percentage of merit-confirmation judges arrived at the appellate court having already served on another bench. On the other hand, a significantly smaller percentage of merit-confirmation judges come from private practice. Instead, they are the most likely of all the judges to have served as prosecutors and in other government-law positions.

Do the constraints imposed by the confirmation process in merit-confirmation states cause those benches to have a different mix of characteristics than those in merit-selection states? By and large, just as judges chosen by merit-confirmation are significantly different in many respects from other judges overall, benches in merit-confirmation states are significantly different from those in merit-selection states. By significant margins, more merit-confirmation judges than merit-selected judges are former prosecutors or other government lawyers, and more came to their current positions with prior judicial experience. On the other hand, merit-selected judges are significantly more likely than merit-confirmation judges to come from private practice. With respect to education credentials, judges selected by merit-confirmation are significantly less likely than merit-selected judges to have attended in-state law schools and in-state undergraduate institutions. Correspondingly, merit-confirmation judges are more likely to have attended ranked law schools or undergraduate schools, although the difference is not statistically significant. Judges selected by merit-confirmation are significantly more likely than merit-selected judges to have taught at the college or law school level, to have been elected to the American Law Institute, and to have published in a law review before coming to the bench.

The fact that judges selected by merit- confirmation are significantly different in many objective respects from those chosen by merit selection indicates that the additional constraint imposed by a confirmation process may have demonstrable effects. As noted above, in the federal judicial appointment process, the Senate’s power of advice and consent significantly constrains presidential appointment power. In similar fashion, it would seem likely that, when governors in merit-confirmation states decide whom to nominate to the appellate bench, they choose applicants they expect will appeal to the members of the elected body that must confirm them. That may help explain the data that show that more appellate judges in merit-confirmation states share objective credentials such as prior judicial experience, longer legal careers, and judicial clerkships.11

Judicial Election

Taken together, some significant differences between the characteristics of appellate judges in election states and those in merit-selection and merit-confirmation states. Judges in election states are significantly more likely to be female than judges in merit-selection and merit-confirmation states.12 Judges in election states are significantly younger and have correspondingly less legal experience than those selected by merit or merit-confirmation. By significant margins, judges in election states are less likely to have prior judicial experience but more likely to have private practice experience. Not surprisingly, a significantly greater proportion of judges in election states attended in-state law schools and undergraduate schools. By contrast, a significantly greater percentage of merit-selected and merit-confirmation judges attended ranked law schools. Of course, the data do not tell us whether merit selection and merit-confirmation prefer paper credentials more than voters do, or whether lawyers who have attended ranked law schools are more inclined to seek the bench in merit states than in election states.

On close inspection, however, the comparison between the benches in election states and in appointed states is not as simple as it might appear. As noted above, when an interim vacancy occurs in a state in which judges are elected, the governor usually is empowered to fill the vacancy by appointment. That appointed judge then runs as an incumbent in the next election. Of the appellate judges sitting in election states at the time of this study, 41 percent first came to the bench by appointment.

It has been commonly thought that the precise means by which judges come to the bench in election states (i.e., by election or by interim appointment) does not matter very much.13 In several respects, however, appointed judges in election states bear significantly different characteristics than those of their elected colleagues. Interim appointments usually have the effect of bringing the appellate benches in election states more in line with the norm produced by other selection methods. For example, among the statistically significant differences between elected and appointed judges in election states is that a significantly greater proportion of the appointed judges are nonwhite. This is particularly true in nonpartisan election states, where more than 18 percent of the appointed judges but only 5 percent of the elected judges are nonwhite. (By contrast, there is virtually no distinction between the proportion of women first appointed and elected to the bench in election states).

Judges in election states who first come to the bench by appointment, particularly in partisan-election states, are considerably more likely to have some commercial law or business law experience than those who are elected. The opposite is true with prosecutorial experience, particularly in partisan-election states. Judges appointed to fill interim vacancies in partisan-election states are far less likely than those first elected to have practiced in legal aid or criminal defense, but no such distinction is seen in nonpartisan election states. As for prior government-law experience, the proportion of judges actually elected in nonpartisan election states who have practiced in government is smaller than any other group of judges. When appointing authorities fill interim vacancies in nonpartisan election states, however, nearly 24 percent of their appointments have prior government-law experience.

Turning to education, appointed judges in election states are significantly less likely to have attended in-state law schools than those who are elected. Although proportionately more appointees in partisan-election states attended ranked law schools than those first elected, the opposite is true (by a wide margin) in nonpartisan election states. With respect to other scholarly credentials, significantly more elected members of the American Law Institute come to the bench via appointment than by election in partisan-election states, but there is virtually no distinction among the numbers of ALI members appointed and elected in nonpartisan-election states. Finally, by a significant margin, judges who are first appointed in election states are more likely to have published a law review article before coming to the bench than those who are first elected. The difference is particularly striking among judges in partisan-election states.

Merit Selection and Women

The table below shows women’s representation among appellate judges in the dataset, according to when they took the bench.

Although fewer than half of all state appellate judges on the bench in 2015 are women, their proportional representation does not compare unfavorably with U.S. employment data. That is, according to the U.S. Statistical Abstract, 22.2 percent of the lawyers employed in 1989 were women. By 2014, when more than 35 percent of all judges selected for state appellate benches were female, women represented 32.9 percent of all employed lawyers in the country. According to the American Bar Association, 35 percent of the licensed lawyers in the United States in 2015 were women.14

The progress of women is not uniform across all selection methods, however. Women are chosen for state appellate courts at a statistically significantly lower rate in merit-selection states than in other states. Although more than 35 percent of all the appellate judges in the dataset are women, fewer than 29 percent of the judges in merit-selection states are women. A separate multivariate analysis reveals that, holding all other factors constant and by a statistically significant margin, a state appellate judge selected by merit is 31 percent less likely to be a woman than a judge selected by any other method.15


Click Here to download an accessible PDF of this article with tables.

Why do women seem to struggle for appointment in merit-selection states? By itself, a “merit” system does not disproportionately disadvantage women: More than 36 percent of the appellate judges on the bench in 2015 who were selected through merit-confirmation are women. Whether governors in merit-selection states tend to disproportionately select male nominees over female nominees because of political influences that favor men more than women cannot be determined. But “politics” in general is not a sufficient explanation. After all, for women, the single most favorable appellate judicial-selection method is partisan election (with nonpartisan election not far behind). At the end of the day in merit selection, the governor’s appointment is subject to an unquantifiable mix of factors and influences, including, of course, the governor’s ideological predilections. Perhaps some explanation, however, lies in the characteristics of the judges chosen by the various selection methods.

Looking first at the attributes of female judges and the attributes of male judges across all selection methods, by statistically significant margins, the women are less likely to have private-practice experience than the men, and the female judges who practiced in the private sector did so for fewer years. Holding all other factors constant, a judge with private-practice experience is 38 percent less likely to be a woman than a judge with another career path. The disparity in private-practice experience is not surprising. Historically, more male lawyers than female lawyers have joined private law firms and remained in private practice for significant periods. Beyond these differences, however, there are no other statistically significant distinctions between the career paths of female judges in the overall dataset and their male counterparts. Given that, generally speaking, women are represented overall on state appellate benches at about the same rate as they are present in the practicing bar, among most selection methods it appears that career paths other than private practice provide women with satisfactory routes to the bench. Conversely, it appears that only in merit-selection states may a lack of private-practice experience disadvantage female judicial aspirants.

Holding all other factors constant and by statistically significant margins, a judge who has been elected to the American Law Institute is much more likely to be a woman, as is a judge who has served a judicial clerkship. While nearly 30 percent of the female judges are former judicial clerks, only 25 percent of the male judges chose to clerk. For a woman who for reasons of temperament or work/life balance has decided not to pursue partnership in a large law firm or a high-profile position in another legal field, a judicial clerkship might constitute an alternate way of demonstrating the intelligence, integrity and legal ability required for the appellate bench. Academic credentials might afford the same woman another means of demonstrating she is worthy of judicial appointment. But by a significant margin, fewer of the female judges than the male judges boast a degree from a ranked law school; more of the women than the men attended law school in the state in which they sit.

We turn next to considering differences in the characteristics of male and female judges produced by the respective selection methods. This analysis may show whether merit selection treats certain characteristics in women differently than other selection methods treat those characteristics, and ultimately, whether merit selection treats those characteristics differently in women than in men.

There are few statistically significant differences in the relative characteristics of men and women chosen by the respective selection methods. That is, when male or female judges tend to be more or less likely to have had a particular career path, that same tendency usually is seen across all selection methods. There are three exceptions, however, arising on benches in merit-selection states. First, the significant overall private-practice differential between the female members of the bench and the male members is greatest by far in merit-selection states. Second, the difference between the men and women who have business-law experience is much greater in merit-selection states than elsewhere. Third, a significantly greater percentage of women chosen by merit selection have served judicial clerkships than their male counterparts. Although, as noted, female judges overall are more likely than male judges to have served clerkships, in no other selection method is the margin statistically significant.

Multivariate analyses show that, holding all other factors constant, judges chosen by merit selection who have private-practice experience are significantly less likely to be female than those without private-practice experience. On the other hand, holding everything else constant, a merit-selected judge who has practiced in government is 40 percent more likely to be a woman than a merit-selected judge without government-law experience. At the same time, a judge with prior government experience who is selected by any other method is no more likely to be a woman than one without prior government experience. A judge selected by merit who served a judicial clerkship is more than twice as likely to be a woman as a judge who did not clerk; no similar disparity is seen among judges selected by other methods.

In sum, female judges in merit-selection states are considerably less likely to have private-practice or business-law experience than male judges in merit-selection states or than female judges in other states.16 By itself, the relatively low rate of women in private practice does not explain why proportionately fewer women are chosen by merit selection than by other selection methods. Other selection methods in which women succeed at about the same rate as they are represented in the bar choose even higher percentages of judges with private-practice experience than does merit selection.

One possible explanation is that merit selection does not value women’s private-practice and business-law experience to the same degree as it values men’s private-practice and business-law experience, and does not value women’s private-practice and business-law experience to the same degree that other selection methods do. Why might this be so? The explanation may be in the nature of a female judicial applicant’s private practice.

Recall that to a statistically significant degree, the female judges’ private-practice careers are shorter than those of male judges. Data concerning the judges’ practices do not allow close study of the duration of their experience in business law or complex litigation, but there is no reason to conclude that the same gender differences do not occur in that subset of private practice.

Viewing these facts through the lens of merit selection, women who have spent less time as partners at large law firms or in other business-law practices may be less likely than their male counterparts to have formed significant relationships with clients that can influence appointment in merit-selection states. As noted, critics of merit selection contend it unduly favors applicants supported by the traditionally powerful, meaning bar leaders and business interests. It seems possible that to the extent those interests carry weight in merit selection, they tend to favor male candidates over female candidates. Implicit bias may be at work, but more fundamentally, the nature of women’s experiences in private practice may hinder them from winning support among business interests and bar leadership in merit selection. Put simply, to the extent that business interests wield influence in merit selection, they may be more likely to exercise that influence in favor of the lawyers they have hired to represent them in significant matters, and, at least to date, those lawyers tend to be male, not female.

Merit selection does not seem to advantage men over women when they share career paths other than private practice. In fact, merit is a more favorable path to the bench for women than men with prior nonprosecutorial government service. Proportionately more female than male judges are former government lawyers, although the difference is not statistically significant across all selection methods. The difference is greatest among merit-selected judges, however, where 28 percent of the women but only 19 percent of the men have prior governmental service. Consistent with these results, holding all other factors constant, a merit-selected judge who comes from government service is 40 percent more likely to be a woman than a judge with another career path.

Other objective credentials can advantage a woman seeking appointment through merit selection. A judicial clerkship is much more valuable to a woman in a merit-selection state than in a state with another selection method. Recall that, holding all other factors constant, a merit-selected judge who has served a clerkship is more than twice as likely to be a woman than one without a clerkship. (A judicial clerkship also advantages a female judicial aspirant in other selection systems, but not to the same degree.)

By the same token, about the same proportion — 13 percent — of the women and the men on merit-selected benches graduated from ranked law schools. That is about the average among all the male judges, but only 8 percent of all the female judges graduated from ranked law schools. These data indicate that, by contrast to private-practice experience and business-law experience, a degree from a ranked law school is an objective credential to which merit selection gives equal (and considerable) weight, regardless of gender.

In summary, it seems quite possible that the underrepresentation of women on appellate benches in merit-selection states may be related to how merit treats women’s private-practice experience. By all analyses, female judicial candidates from private practice and those with business-law experience do not succeed at the same rate as men with those career paths in merit selection and as comparable women in other selection systems. These results may lend support to the proposition that merit selection unduly advantages judicial applicants who have developed strong relationships with business and other establishment interests. To the extent that, for whatever reason, women in private practice have not developed those relationships, they may be disadvantaged by merit selection. The same gender differences are not seen among benches in merit-confirmation states, where the governor’s choice of whom to appoint is constrained by a confirmation requirement. Why this so deserves further study.

At the same time, data show that merit selection fairly recognizes and values other prior legal experience and credentials in women. To a greater extent than other selection methods, merit selection seems to value a woman’s prior career in government law. And merit selection seems to give equal weight to a woman’s prior prosecutorial experience as it gives to a man’s. Moreover, merit selection seems to afford significantly greater weight to a judicial clerkship served by a woman than by a man; this may afford success to a woman who has served a clerkship but lacks private-practice or business-law experience. A degree from a ranked law school is another credential signifying intellectual competence that may help a female applicant succeed in merit selection even if she is unable to call on business interests for support with the nominating commission or the governor. Nevertheless, the persistent gender disparity on merit-selected appellate benches argues in favor of continued efforts among judicial nominating commissions and governors in those states to recruit diverse judicial applicants and to make conscious efforts to give fair weight to female applicants’ records and legal experience.

1 The study discussed here was conducted in partial fulfillment of the requirements for a Master of Laws in Judicial Studies from Duke University School of Law. See Diane M. Johnsen, Building a Bench: A Close Look at State Appellate Courts Constructed by the Respective Methods of Judicial Selection, 53 San Diego L. Rev. (2017) (forthcoming).
2 See, e.g., Mark S. Hurwitz & Drew Noble Lanier, Diversity in State and Federal Appellate Courts: Change and Continuity Across 20 Years, 29 Justice Sys. J. 47, 49 (2008); Nicholas O. Alozie, Selection Methods and the Recruitment of Women to State Courts of Last Resort, 77 Soc. Sci. Q. 111, 112 (1996); Barbara Luck Graham, Do Judicial Selection Systems Matter? A Study of Black Representation on State Courts, 18 Am. Pol. Q. 316, 32 (1990).
3 See, e.g., Victor Eugene Flango & Craig R. Ducat, What Difference Does Method of Judicial Selection Make? 5 Justice Sys. J. 25, 25 (1979). See also Stephen J. Choi, G. Mitu Gulati & Eric A. Posner, Professionals or Politicians: The Uncertain Empirical Case for an Elected Rather than Appointed Judiciary, 26 J. L. Econ. & Organ. 290, 327 (2008) and Chris W. Bonneau & Melinda Gann Hall, In Defense of Judicial Elections 7 (2009).
4 Alozie, supra note 2, at 112; Hurwitz & Lanier., supra note 2, at 52; Malia Reddick, Michael J. Nelson & Rachel Paine Caufield, Examining Diversity of State Courts: How Does the Judicial Selection Environment Advance — And Inhibit — Judicial Diversity?, Am. Judicature Soc’y 7 (2008), available at http://www.judicialselection.us/uploads/documents/Examining_Diversity_on_State_Courts_2CA4D9DF458DD.pdf.
5 Hurwitz & Lanier, supra note 2, at 53, Table 2; Reddick, et al., supra note 4, at 4, 6. Reddick, et al., concluded that the method of selection did not affect the gender composition of state courts of last resort, but that, taking into account the relevant political environments, “merit selection placed significantly fewer women on intermediate appellate courts than did partisan or nonpartisan elections.”
6 West Virginia switched from partisan elections to nonpartisan judicial elections in 2016, after the effective date of this study.
7 Tennessee added a confirmation requirement to its merit-selection system effective in January 2016. For purposes of this study, its judges are treated as having been chosen through merit selection.
8 See, e.g., Hurwitz & Lanier, supra note 2, and Reddick, et al., supra note 4. See also Mark S. Hurwitz & Drew Noble Lanier, Explaining Judicial Diversity: The Differential Ability of Women and Minorities to Attain Seats on State Supreme and Appellate Courts, 3 State Pol. & Pol’y Q. 329, 335 (2003).
9 In some of the states, the governor’s appointment must be approved by one or both houses of the legislature; in a handful of other states, a separately elected body must confirm the appointment.
10 For this purpose, I used U.S. News and World Report rankings of the top ten law schools and top 20 colleges and universities. A host of law schools and undergraduate institutions that are unranked by these measures provide excellent educations to their students, and many high-achieving students who are admitted to ranked institutions cannot afford to enroll. National rankings nevertheless provide some measure of their students’ intellect and, to a greater or lesser extent, the quality of their graduates’ legal training.
11 Although plenty of bright young law graduates choose not to seek judicial clerkships, I credit a judicial clerkship as an objective badge of merit because judges tend to hire their clerks from students at the top of their law school classes.
12 Women are selected for the appellate bench in election states at a higher rate than they are elected to other state offices. Of all the appellate judges in election states on the bench in 2015, nearly 39 percent are women. By contrast, in 2016, women represented about 25 percent of all statewide elective officials and state legislators. Susan J. Carroll, Women in State Government: Still Too Few, in The Book of the States 450, 452 (National Conference of State Governments, 2016).
13 See, e.g., Henry R. Glick & Craig F. Emmert, Selection Systems and Judicial Characteristics: The Recruitment of State Supreme Court Judges, 70 Judicature 228, 231 (1987).
14 A.B.A., Lawyer Demographics (2015), https://www.americanbar.org/content/dam/aba/administrative/market_research/lawyer-demographics-tables-2015.authcheckdam.pdf.
15 Looking separately at intermediate appellate and high-court benches, fewer women are selected for state courts of last resort (30 percent) and intermediate appellate courts (28 percent) by merit selection than by any other selection method. The margin across all selection methods is statistically significant among the latter but not among the former.
16 Of the female judges selected by merit selection, 74 percent have practiced in the private sector; by contrast, 80 percent of female judges selected by other means have private practice experience. For purposes of comparison, proportionately more men (nearly 89 percent) chosen by merit selection have practiced in the private sector than men chosen by other methods (85 percent).