Judicial Independence and Opinion Clarity on State Supreme Courts

Abstract

We contribute to the literature on judicial independence and performance in the states by analyzing opinion clarity. Written opinions are the primary means of communication for state supreme court justices, and clarity is a core component of judicial performance. Using automated text analysis on a sample of state supreme court opinions from all 50 states, we find that variation in judicial retention systems is not associated with substantively meaningful differences in opinion clarity. Furthermore, elected judges do not seem to produce clearer opinions in salient cases notwithstanding the increased public visibility of those decisions. These results suggest that judges tend to conform to prevailing professional norms despite differing institutional pressures.

Keywords

elections political behavior judicial elections judicial politics judicial behavior comparative courts public policy

The public and scholarly debate over judicial elections continues to intensify. As states evaluate reform proposals either to increase or decrease judicial accountability, policy makers, judges, advocacy groups, and scholars make the case for their favored methods of selection and retention. Although this debate has historically focused more on rhetoric than empirical evidence concerning the consequences of different institutional design choices, a new wave of research concerning the effects of elections on voter turnout and knowledge (e.g., Bonneau and Hall 2009), perceptions of institutional legitimacy (e.g., Gibson 2012), and decision-making on the merits (e.g., Langer 2002) provides a basis for objectively evaluating reform proposals. Recently, scholars have also proposed evaluating the consequences of different selection and retention mechanisms based on objective indicators of judicial performance (e.g., Cann 2007; Choi, Gulati, and Posner 2010; Goelzhauser 2012). This is an important line of research insofar as it contributes to our understanding of the consequences of institutional design choices in the states as well as judicial behavior.

We contribute to the literature on judicial independence and performance in the states by analyzing opinion clarity. Opinion clarity is a central component of judicial performance (see, for example, Clark and Carrubba 2012; Lax and Cameron 2007). Written opinions are the primary means of communication between judges and stakeholders in the judicial process. As a result, it is considered essential for judges to write in a way that can be easily understood by a diverse group of constituents, including the public, media, policy makers, and other judges (see, for example, Aldisert, Rasch, and Bartlett 2009; Garner 2002; George 2006). Notwithstanding the widespread consensus that “good legal writing is plain English” (Wydick 2005, 5), little effort has been devoted to understanding the causes and consequences of opinion-writing clarity (but see Nelson 2013; Owens and Wedeking 2011; Owens, Wedeking, and Wohlfarth 2013; Spriggs 1997; Staton and Vanberg 2008). Ours is the first project to link the discussion of opinion clarity to the literature on judicial performance and examine its determinants across a broad range of legal issues confronting state supreme courts.

Considering the conflicting theoretical and empirical evidence concerning the link between judicial independence and performance, the expected relationship between elections and opinion clarity is ambiguous. The literature on judicial performance and behavior supports three conflicting hypotheses. Some have argued that elected judges are likely to be less able than appointed judges (Posner 2005), in which case we might expect to observe elected state courts producing less clear opinions. Alternatively, agency theory offers a theoretical basis for the expectation that the electoral connection may enhance judicial performance (Choi, Gulati, and Posner 2010; Goelzhauser 2012), resulting in elected state courts producing clearer opinions in an effort to better communicate with diverse constituencies. Last, the judicial behavior literature on role orientation and audience perspectives (Baum 2006; Gibson 1978) suggests that, in some circumstances, judges may act similar despite different institutional environments due to prevailing professional norms. In addition to exploring the baseline connection between judicial independence and opinion clarity, we also examine the potentially modifying influence of case salience. Given that the public pressures associated with accountability are most pronounced in salient cases (Cann and Wilhelm 2011), we suggest that elected judges may issue clearer opinions in those disputes.

Using a sample of decisions handed down by supreme courts in all 50 states, we find no evidence of a substantively meaningful relationship between judicial retention institutions and opinion clarity either in salient or nonsalient cases. These results are consistent with the view that at least in some respects, “judges are judges.” Still, these results have important implications for our understanding of the consequences regarding the design of judicial retention systems in the states in addition to the literature on judicial performance. Furthermore, to the extent that clarity serves as a proxy for opinion quality (see, for example, Clark and Carrubba 2012; Lax and Cameron 2007), the results contradict findings presented by Choi, Gulati, and Posner (2010) suggesting that elected judges write lower-quality opinions using case citations to capture quality. Considering the widespread criticism of the use of citations to capture opinion quality (e.g., Farber 2005; Marshall 2004; Solum 2005), scholars may wish to consider clarity as an additional indicator of quality in studies of judicial performance. This project also begins to bridge the growing literatures employing automated content analysis to uncover substance in political text and communications concerning state (e.g., Coffey 2005; Weinberg 2010) and judicial (e.g., Black et al. 2011; Corley, Collins, and Calvin 2011) politics.

Judicial Retention and Performance

The policy debate over judicial elections hinges in part on understanding the empirical connection between judicial independence and performance. State experimentation with alternative institutions for selecting and retaining judges allows scholars to explore the behavioral consequences of reform. This endeavor has been aided in recent years by attempts to objectively measure components of judicial performance to gain a more systematic understanding of the relationship between independence and performance rather than relying solely on anecdotes and normative perspectives concerning the efficacy of various institutional design choices. Although the empirical results are mixed, the literature offers three theoretical perspectives for understanding the link between judicial independence and performance.

One theoretical perspective concerning the connection between judicial independence and performance suggests that courts with less independence are likely to perform worse on average. This view derives in part from the perception that merit selection systems are more likely to produce judges who are politically independent (e.g., Geyh 2003). Furthermore, some opponents of judicial elections contend that the rigors and costs of running an election campaign may deter well-qualified people from seeking office (Goldschmidt 1994). As Judge Posner (2005, 1268) put it, “The number of people who have both political and judicial talent . . . is probably very small, and there may even be a degree of incompatibility between the two kinds of talent”; as a result, he argues, “we can expect that other things being equal, an elective judiciary will be less able than an appointive one.” The existing empirical evidence offers some support for Judge Posner’s view concerning the link between judicial independence and performance. Survey-based measures of judicial performance, for example, indicate that judges who enjoy less independence receive lower performance ratings. Cann (2007) compares judges’ ratings of their state courts across different selection methods and finds that states with partisan election systems receive lower ratings from their judges than states with merit selection or appointment. Berkowitz and Clay (2006), meanwhile, use surveys of corporate attorneys (a sample that they admit has a number of biases relative to the population of all attorneys) conducted by the Chamber of Commerce and find higher-quality ratings for states using retention elections. There is also evidence that elected courts write higher-quality opinions as measured by out-of-state citations (Choi, Gulati, and Posner 2010).

An alternative theoretical perspective concerning the connection between judicial independence and performance derives from agency theory. On this view, accountability is likely to induce better court performance. The enhanced competitiveness of judicial elections in recent decades (Bonneau 2004; 2005; Hall 2007) may necessitate a closer link between elected court systems and performance. There is abundant evidence that elected judges are more accountable to citizens in their decision-making on the merits (Bonneau and Hall 2009; Brace and Boyea 2008; Hall 1987; 1995; 2001). Elected judges are also more predictable than appointed judges in their decision-making (Brace, Yates, and Boyea 2012; Yates, Tankersley, and Brace 2010). Furthermore, there is some evidence that elected courts perform better using alternative measures of judicial performance. For example, elected courts have been shown to be more productive (Choi, Gulati, and Posner 2010) and efficient in disposing certain types of civil and criminal cases (Goelzhauser 2012).

A third theoretical perspective suggests that there may not be substantively meaningful differences across selection and retention systems with respect to some aspects of performance. Although this perspective has received comparatively little attention in the scholarly literature and policy debate over judicial elections, it is well grounded in the judicial behavior literature suggesting that, at least under some circumstances, “judges are judges.” Judges may feel strong compulsion toward filling their professional social role (Gibson 1978) or derive utility from being respected by peers and other members of the legal community (Baum 2006; Posner 1993). As a result, judges might be expected to perform similarly regardless of their institutional constraints. Some of the existing empirical literature concerning the effects of institutional design choices on state judiciaries can be interpreted in light of this perspective. For example, some research suggests that there are few differences in qualifications across selection systems (Emmert and Glick 1988; Glick and Emmert 1987). Moreover, to the extent that judicial diversity is a component of institutional quality (Alozie 1988; Torres-Spelliscy, Chase, and Greenman 2008), the existing empirical evidence suggests that there is little or no difference in diversity across selection institutions (Glick and Emmert 1987; Goelzhauser 2011; Hurwitz and Lanier 2003, 2008).

Opinion Clarity

We approach the study of judicial independence and performance through the lens of opinion clarity. One of the axioms of legal writing is that “[g]ood legal writing is clear, simple writing” (Lebovits 2004, 65). Authoritative legal (Garner 2002) and judicial (George 2000) writing handbooks emphasize the importance of clarity. Furthermore, scholars (Wanderer 2002) and judges (e.g., Aldisert, Rasch, and Bartlett 2009; Bell 1966; Wald 1995) regularly praise the virtues of opinion-writing clarity. The notion that opinion clarity is a critical aspect of judicial performance is further highlighted by the fact that numerous state judicial performance commissions include questions concerning clarity in their survey evaluations. For example, the Judicial Performance Evaluation Questionnaire in New Hampshire asks respondents to rate the “clarity of [a] judge’s decision” on a five-point scale. A recent report published by the National Center for State Courts also urges state supreme courts to focus on producing clear opinions so that they can be understood by the public and other stakeholders in the policymaking process (Vickrey, Denton, and Jefferson n.d.).

It is important to emphasize what we mean by opinion clarity. As with the legal literature advocating clarity in opinion writing discussed above, we are interested in the clarity of opinions as a whole rather than merely the clarity of any specific legal rule (or holding). There are important reasons for this emphasis. In addition to the holding, a judicial opinion typically includes a discussion of the underlying facts of the case and the reasoning used to support the court’s judgment. Clarity across each of these dimensions is important for understanding the context and scope of an opinion. Given that holdings must usually be read in light of the underlying case facts and analysis, there is value in ensuring comprehensive opinion clarity so that the diverse group of constituents for judicial opinions (e.g., the public, media, lawyers, litigants, business interests, advocacy groups, legislators, executives, and other judges) are readily able to comprehend not only the result of a particular case but also its downstream implications.

The widespread legal consensus that opinion clarity in its broad sense is an indicator of judicial performance is a second reason for focusing on this conceptualization. Although we do not know of a movement to write less readable judicial opinions in a broad sense, there is intense debate over the extent to which particular legal rules should offer clear guidance. The debate over clarity in this sense is best represented by the literature on the choice between rules and standards. Whereas some commentators advocate for the certainty of rules, others prefer the flexibility inherent in standards. From the perspective of opinion clarity in a broad sense, however, both rules and standards may be articulated in ways that make them difficult for stakeholders to decipher linguistically (see Kaplow 1992). In any event, although it is important to understand the causes and consequences of what might be called “rule clarity” (rather than “opinion clarity”), we do not understand this to be an objective indicator of judicial performance considering the extensive debate over the value of precision in legal rulemaking.¹

Our core theoretical interest is understanding the relationship between judicial independence and opinion clarity. The existing literature on judicial elections, performance, and behavior discussed previously suggests three possibilities concerning this link. First, elected courts may produce opinions that are less clear on average because they tend to attract less able judges. Second, elected courts may produce clearer opinions on average due to pressures associated with the accountability constraint and the need to communicate clearly with diverse constituencies. Not only is opinion clarity valued in its own right as discussed above but it will also often be a necessary prerequisite for communicating quality performance along other dimensions such as the validity of legal judgments. Last, clarity may differ little across institutional settings if judges are primarily influenced by professional norms. Although we remain agnostic about the expected link between retention institutions and opinion clarity considering the competing theoretical perspectives and empirical results in the judicial performance literature more generally, we add a theoretical nuance concerning case salience.

For elected judges, the pressures associated with public accountability are most pronounced when dealing with salient issues (Cann and Wilhelm 2011). Although Owens and Wedeking (2011) find that opinion clarity is not associated with case salience using a sample of U.S. Supreme Court opinions, the electoral connection facing some judges may create a net incentive to craft more readable opinions. While some judges around the world promote their judgments through public relations strategies to enhance judicial power (Staton 2010), courts in the American states are more likely to rely solely on written opinions to communicate the importance of their decisions. Given that salient opinions are the ones most likely to be scrutinized by core constituencies, it is reasonable to expect that elected judges may be more likely to produce clearer opinions in these cases.

Data and Measurement

We use three common measures of opinion clarity. Although there are many measures of writing clarity, these measures are easy to interpret and are often included as canned features in popular word processing software such as Microsoft Word. First, we utilize the Flesch Reading Ease (FRE) statistic (Flesch 1948). FRE is calculated according to the following formula:

206.835 - 1.015 (\frac{total words}{total sentences}) - 84.6 (\frac{total syllables}{total words}) .

FRE is a widely used measure of clarity in scholarly literature across a variety of fields (e.g., B. Coleman and Phung 2010; Lowrey 2006; Terris 1949). FRE scores range from 0 to 100. Writings with FRE scores of 0–30 are considered very difficult, 31–50 difficult, 51–60 fairly difficult, 61–70 standard, 71–80 fairly easy, 81–90 easy, and 91–100 very easy. Reader’s Digest rates as very easy, the New York Times as standard, and corporate annual reports as difficult on the FRE scale (Finn 1985).

Legislatures and government agencies often use FRE scores when setting policy. For example, Florida requires insurance policies to achieve a minimum FRE score of 45, and Massachusetts requires a minimum score of 50.² Similarly, California requires financial institutions to provide notice to consumers when disclosing nonpublic personal information in a form that achieves a minimum FRE score of 50.³ Government use of FRE statistics to make policy is an important benefit of using this measure to capture opinion clarity because it suggests an objective range that consumer-oriented lawmakers believe is appropriate for ensuring public comprehension. Although we do not claim that judicial opinions might be as readable as the Reader’s Digest, these consumer protection laws offer a sense of what judges might strive toward to ensure widespread readability.

We use Flesch–Kincaid Grade-Level (FKGL) scores as our second measure of opinion clarity. FKGL is calculated according to the following formula:

0.39 - (\frac{total words}{total sentences}) + 11.8 (\frac{total syllables}{total words}) - 15.59 .

Similar to FRE scores, FKGL is a widely used indicator of clarity (e.g., Law and Zaring 2010; Milne, Culnan, and Greene 2006; Reilly and Richey 2011). One of the advantages of FKGL scores is that they translate directly to the number of years of education required to understand the text. For example, an FKGL score of 12 indicates a text that can be read by someone who completed high school, and a score of 16 indicates a text that can be read by someone with the equivalent of a bachelor’s degree.

The percentage of passive sentences in a court’s opinion is our final measure of opinion clarity. Passive voice, which occurs when the subject of a sentence receives the verb’s action, is also considered an important determinant of clarity in judicial opinions (Wanderer 2002, 63). According to Bryan Garner (2002, 41), author of a seminal treatment on legal writing titled The Elements of Legal Style, “[t]he passive voice may lead merely to vagueness, but it also lends itself to obfuscation.” As with our other measures, the percentage of passive sentences is a widely used indicator of clarity (e.g., Schroeder, Aggarwal, and Gibson 1991; Schroeder and Gibson 1990; Wandersee and Clary 2007) and is employed by many popular word processing programs. Although the percentage of passive sentences is an intuitive one to interpret, one downside relative to the other measures used here is that passive voice statistics do not necessarily translate into easily understood substantive categories.

Each of these measures accurately reflects the concept of clarity as envisioned in the legal literature discussed previously. Indeed, the Federal Judicial Center’s Judicial Writing Manual encourages judges to utilize the “active voice” while writing “simple, declarative sentences” that are “understandable by the general reader”—advice that is built into the operationalization of the measures described above.⁴ To illustrate that opinion clarity can vary considerably across these dimensions even in similar cases, we analyzed four relatively recent state supreme court majority opinions addressing the same legal question: whether warrantless searches of cell phones are constitutionally permitted incident to arrest.⁵ FRE scores ranged from 26.1 (very difficult) to 36.7 (difficult), FKGL scores ranged from 14 (corresponding to someone with two years of college education) to 16.9 (corresponding to someone with a graduate degree), and the percentage of passive content ranged from 4% to 16%.

Our sample of state supreme court opinions comes from the State Supreme Court Data Project, which includes cases decided by high courts in all 50 states from 1995 to 1998.⁶ The sample begins with each of the 400 salient decisions in the database (Vining and Wilhelm 2011). Salient decisions are those that received front-page coverage from a state’s most circulated newspaper the day after they were handed down. Next, we randomly sampled 5% of each state supreme court’s remaining decisions over the sample period.⁷ This sampling procedure yielded a total of 1,797 state supreme court decisions. After excluding cases for which we could not find a written opinion, we were left with a sample of 1,783 opinions. To code opinion clarity for the cases in our sample, we downloaded each decision from Lexis, highlighted the majority opinion, and used Microsoft Word to generate the aforementioned readability statistics. Figure 1 displays histograms for each of our opinion clarity measures. Sample means for the readability statistics are as follows: FRE = 34.6 (corresponding to a difficult text), FKGL = 13.8 (corresponding roughly to someone with two years of college education), and passive sentences = 13.6%. Table 1 shows the average of each clarity measure by state.

Figure 1.

Distributions of clarity measures.

Table 1.

Average Clarity of State Supreme Court Opinions.

	Reading ease	Grade level	% Passive sentences
Alabama	34.88	13.85	14.13
Alaska	36.88	13.53	14.47
Arizona	34.05	13.90	14.18
Arkansas	35.83	13.66	13.76
California	34.87	13.86	15.21
Colorado	36.09	13.69	14.42
Connecticut	35.97	13.75	13.55
Delaware	35.49	14.23	15.07
Florida	35.20	14.07	15.33
Georgia	36.24	13.41	13.00
Hawaii	39.14	12.86	17.93
Idaho	34.64	13.86	14.88
Illinois	34.42	13.66	12.97
Indiana	35.55	13.68	17.33
Iowa	36.11	13.44	13.68
Kansas	35.64	13.34	13.09
Kentucky	37.24	13.36	12.86
Louisiana	36.00	13.58	13.24
Maine	34.60	13.76	13.20
Maryland	30.76	14.82	12.80
Massachusetts	32.22	14.35	12.13
Michigan	33.21	14.08	12.74
Minnesota	32.55	14.13	10.72
Mississippi	32.66	14.17	11.72
Missouri	33.17	13.99	10.32
Montana	32.33	14.29	13.38
Nebraska	30.88	14.92	14.71
Nevada	33.17	14.24	15.77
New Hampshire	34.03	14.15	12.58
New Jersey	33.23	14.22	11.47
New Mexico	33.70	14.11	10.93
New York	33.56	14.28	10.42
North Carolina	33.49	13.72	12.46
North Dakota	35.77	13.27	12.71
Ohio	34.34	13.62	12.67
Oklahoma	33.21	13.79	12.63
Oregon	34.84	13.64	11.00
Pennsylvania	37.04	13.28	14.00
Rhode Island	34.92	13.75	13.65
South Carolina	37.23	13.37	13.85
South Dakota	36.92	13.55	13.97
Tennessee	34.20	13.79	13.43
Texas	35.14	14.04	14.13
Utah	33.17	14.43	13.44
Vermont	33.56	13.77	15.48
Virginia	35.02	13.53	12.08
Washington	35.94	13.50	13.74
West Virginia	35.82	13.52	14.79
Wisconsin	33.71	13.88	13.45
Wyoming	35.68	13.61	14.03

Data Analysis

Bivariate Analysis

Perhaps the simplest approach to examining the association between opinion clarity and retention mechanisms is to compare the means of various measures of opinion clarity across the sundry types of judicial retention systems to see whether they differ substantively and statistically from one another. States use four general retention mechanisms for their supreme courts: partisan elections, nonpartisan elections, retention elections, and reappointment (by the governor, legislature, or a nominating commission). In addition, three states provide their high court judges with secure tenure (i.e., judges are retained for life or until they reach a mandatory retirement age). We perceive the primary differences in selection mechanism to be between contestable elections (whether partisan or nonpartisan), retention elections, reappointment, and secure tenure.⁸ Table 2 arrays the mean values of the FRE, FKGL, and the percentage of passive sentences across states using contestable elections, retention elections, a reappointment process, and states with judges who face no retention because they have life terms.

Table 2.

Mean Opinion Clarity by Retention System.

Measure	Competitive	Retention	Reappointment	Life term	F-test
Reading ease	34.64	34.57	35.16	33.87	0.79
	[34.06, 35.22]	[33.95, 35.19]	[34.20, 36.11]	[32.36, 35.38]	p = .50
Grade level	13.79	13.89	13.76	14.05	1.29
	[13.67, 13.91]	[13.75, 14.03]	[13.55, 13.96]	[13.75, 14.35]	p = .28
% Passive sentences	13.46	14.00	13.25	12.86	2.09
	[13.02, 13.90]	[13.49, 14.50]	[12.48, 14.02]	[11.80, 13.93]	p = .10

Note. Cell entries are arithmetic means with 95% confidence intervals in brackets.

The results show very little substantive variation from one scale to the next. None of the differences in mean clarity scores are statistically significant across the groups. For all pairwise comparisons (within a measure), the 95% confidence intervals overlap, but an analysis of variance (ANOVA) model provides a more appropriate test of the null hypothesis that the means in each category do not differ from one another.⁹ The F-values from the ANOVA model are arrayed for each outcome measure in the right-most column of Table 2; none of them are statistically significant.¹⁰ Moreover, even if the differences were taken as population estimates (ignoring the uncertainty around our point estimates), the differences between the largest and smallest values are substantively trivial—a difference of 1.29 points in reading ease, a difference of 0.29 grade levels, and a 1.14% difference from the lowest to highest use of passive content between retention elections and secure tenure. A substantively meaningful response of judges to the electorate would, in our opinion, require at least a full grade-level difference. A post hoc power analysis shows that given our sample sizes, we would likely have sufficient statistical power to detect such an effect if it existed.¹¹ In sum, there is a clear lack of either substantive or statistically significant differences in the clarity of opinions across retention systems even when using multiple measures of clarity.¹²

Multivariate Analysis

Here, we develop a multivariate model to test the theoretical expectation that courts with judges who face competitive elections will produce clearer opinions in salient cases in an effort to better communicate with constituents. Our dependent variable is opinion clarity, and we fit separate models for each of the measures discussed above.¹³ The primary explanatory variables are an indicator for competitive retention elections, a measure of case salience based on front-page coverage of a decision in the state’s most circulated newspaper the day after being handed down (Vining and Wilhelm 2011), and the interaction of these variables.

We also include a variety of control variables to account for additional influences on opinion clarity. Complex cases may generate opinions that are more difficult to read on average. To capture case complexity, we include a variable scoring the number of issues in a case, an indicator for cases with more than two litigants, and an indicator denoting the presence of amicus curiae (see Collins 2008).¹⁴ Courts with productivity constraints may find it more difficult to produce clear opinions. To capture productivity constraints, we include a measure of state supreme court professionalism that accounts for the number of law clerks employed (Squire 2008), the number of cases per judge decided by the court that year, and an indicator variable for courts with discretionary dockets.

To account for the state’s political and socioeconomic environment, we include several variables. Owens, Wedeking, and Wohlfarth (2013) find that the U.S. Supreme Court produces opinions that are more difficult to read when facing a hostile Congress to insulate its policies from political attack. Because even elected courts may face political attacks from coordinate state branches, we include a variable capturing the absolute value of the ideological distance between the state supreme court and state government.¹⁵ Second, to ensure these institutional ideal points are located in the same policy space, we use Brace, Langer, and Hall’s (2000) measure of state supreme court ideology and Berry et al.’s (1998) measure of state government ideology. We also include a variable capturing the percentage of a state’s population who are college graduates. Last, we account for issue area and time effects. Our issue variables include an indicator scored 1 for constitutional cases and issue fixed effects. We also include year fixed effects.¹⁶

Table 3 displays results from a series of ordinary least squares (OLS) regressions explaining the determinants of opinion clarity across state supreme courts. Although none the coefficients on the use of competitive elections or the interaction between competitive elections and salience are statistically significant, to probe the matter more thoroughly, we present the effects of the use of competitive elections on opinion clarity varying case salience in Table 4. For salient cases, the difference between contestable and noncontestable elections is substantively negligible (one-fourth of a reading ease point, less than one-fifth of a grade level, and less than one-fourth percent of passive content) and statistically insignificant. In nonsalient cases, the differences between elected and appointed judges are again statistically insignificant, and substantively trivial, with the grade-level effect being less than one-tenth of a grade level, and the reading ease and passive content measures suggesting that, if anything, elected judges are less clear in their opinions, although these effects are very modest. To determine whether the effects of retention method on opinion clarity differ across salient and nonsalient cases, we calculate a difference-in-differences (with standard errors calculated by the delta method). Across all three measures of opinion clarity, we find no support for the hypothesis that case salience conditions the relationship between opinion clarity and method of retention as none of the difference-in-differences figures are statistically or substantively significant.¹⁷

Table 3.

The Determinants of State Supreme Court Opinion Clarity.

	Reading ease	Grade level	% Passive sentences
Competitive elections	−0.61	0.06	−0.77
	(0.54)	(0.13)	(0.46)
Salient case	−1.27	0.36	0.31
	(0.70)	(0.19)	(0.54)
Competitive × Salient	0.85	−0.23	0.53
	(0.94)	(0.24)	(0.76)
Issues	0.45	−0.12*	0.16
	(0.22)	(0.05)	(0.18)
Multiple parties	0.54	−0.10	−0.50
	(0.47)	(0.10)	(0.42)
Amicus participation	0.36	−0.17	−0.07
	(0.63)	(0.12)	(0.50)
Court professionalism	2.25	−0.28	0.76
	(1.17)	(0.31)	(1.00)
Caseload	0.02	−0.004	0.01
	(0.03)	(0.009)	(0.02)
Discretionary docket	0.15	−0.05	−0.61
	(0.60)	(0.16)	(0.42)
% College	−0.12	0.03*	−0.102
	(0.07)	(0.01)	(0.053)
Constitutional case	−0.29	−0.07	−0.97
	(1.31)	(0.32)	(0.69)
Intercept	35.99	13.51	15.22
	(2.25)	(0.49)	(2.03)
r ²	.05	.05	.02
RMSE	7.89	1.68	6.23
N	1,783	1,783	1,783

Note. Issue and year fixed effects are not displayed; RMSE = root mean square error. Robust standard errors in parentheses.

p < .05 (two-tailed).

Table 4.

Selection Method, Salience, and Opinion Clarity.

	Reading ease	Grade level	% Passive sentences
Difference between competitive elections and noncompetitive elections on salient cases	0.25 (1.02)	−0.17 (0.25)	−0.23 (0.77)
Difference between competitive elections and noncompetitive elections on nonsalient cases	−0.61 (0.54)	0.06 (0.13)	−0.77 (0.46)
Difference-in-differences	0.85 (0.94)	−0.23 (0.24)	0.53 (0.76)

Note. Standard errors (in parentheses) are computed using the Delta method. The difference between figures in the first two rows may differ slightly from the difference-in-differences row due to rounding error.

Turning to the controls, none of the explanatory variables are consistent predictors of opinion clarity. Counterintuitively, an increase in the number of issues in a case is associated with a decrease in the average grade level associated with opinion text. However, the change is substantively small: one additional issue is associated with a decrease in the grade level required to understand the text of about 0.12 [−0.21, −0.03].¹⁸ An increase in the percentage of state residents with a college degree is associated with an increase in the average grade level associated with opinion text. But here again, the change is substantively small: an increase of 1% in state residents with a college degree is associated with about a 0.03 [<0.01, 0.05] increase in the grade level required to understand the opinion text. Although the issue area fixed effects are not jointly significant in any model, the year fixed effects are jointly significant (p < .01) in the reading ease and grade-level models.

Conclusion

Ultimately, the debate over how to structure judicial selection and retention systems is best informed by empirical evidence concerning the consequences of institutional design choices. Recent work employing objective indicators of judicial performance to examine the effects of these institutional choices represents an important advancement in the study of judicial politics in the states. The theoretical and empirical results in this literature support three alternative views regarding the link between judicial independence and performance in the states: elected courts perform better in light of the accountability constraint and need to communicate clearly with diverse constituents, elected courts perform worse because they attract less able judges on average, and performance is similar due to the influence of professional norms. We contribute to this debate by examining the relationship between judicial retention and opinion clarity, which is a core component of judicial performance. The results suggest that there is no substantively meaningful difference in opinion clarity across retention systems. Moreover, elected courts do not appear to produce clearer opinions in salient cases notwithstanding enhanced public scrutiny. Overall, these results are consistent with the view that there are certain dimensions of performance that judges manage in similar ways despite facing different institutional environments.

Understanding the link between judicial independence and opinion clarity also has implications for the developing literature on opinion quality. The prevailing view has long been that “the difficulties inherent in attempting to measure opinion quality are possibly insurmountable” (Stewart and Heck 1987, 373n3). Recently, however, scholars have employed innovative objective measures to capture opinion quality. The most prominent study at the intersection of state politics and judicial behavior is Choi, Gulati, and Posner’s (2010) use of external citations to examine opinion quality in state supreme courts, finding that elected courts produce lower-quality opinions on average. Although the use of citations to capture opinion quality is popular in the study of judicial politics, it has been widely criticized (e.g., Farber 2005; Marshall 2004; Solum 2005), with the most common concern being that citations are a better indicator of influence. To the extent that clarity is a component of opinion quality (see, for example, Clark and Carrubba 2012; Lax and Cameron 2007), our results contradict those presented by Choi, Gulati, and Posner (2010). Of course, there are limitations to using clarity as a measure of quality. Although clarity may be a necessary condition for a quality opinion, it is not sufficient. For example, a clear opinion might be poorly organized, unpersuasive, or lack fidelity to the law. Notwithstanding these limitations, we encourage scholars interested in opinion quality to consider clarity as an additional measure.

Although our results suggest that there is no substantively meaningful difference in opinion clarity across judicial retention systems in the states, there are several questions deserving of attention in future research. Our sampling procedure yielded cases from 1995 to 1998 across many issue areas, but it is possible that incentives to produce clear opinions have changed in light of recent increases in the competitiveness of state judicial elections or that there is more variation within particular issue areas. In addition, while we focus specifically on the clarity of the opinion as a whole, a similar approach can be invoked to examine the causes and consequences of the clarity of legal rules. While there may be some overlap between opinion and rule clarity, these are distinct concepts. For example, judges might articulate complex legal rules in opinions that are highly readable. Legal rules might also be difficult to implement due to vagueness despite being quite clear in the sense that we use the term here. In Brown v. Board of Education II (1955), for example, the Supreme Court ordered the lower courts to implement desegregation plans with “all deliberate speed.” As a result, it is important for scholars to be clear about the underlying concept of interest. Last, judges sometimes have incentives to produce vague opinions (Staton and Vanberg 2008). For example, judges might be less likely to produce clear opinions when facing threats of noncompliance from key political actors. This expectation can be incorporated into future research by modeling the ideological direction of decisions and the preferences of actors charged with implementation.

Footnotes

Acknowledgements

We thank Mark Miller and participants at the 2013 annual meeting of the Midwest Political Science Association for helpful comments, and Amanda Leifson for excellent research assistance.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Greg Goelzhauser is an assistant professor of political science at Utah State University.

Damon M. Cann is an associate professor of political science at Utah State University.

References

Aldisert

Ruggero J.

Rasch

Meehan

Bartlett

Matthew P.

2009. “Opinion Writing and Opinion Readers.” Cardozo Law Review 31:1–44.

Alozie

Nicholas O.

1988. “Black Representation on State Judiciaries.” Social Science Quarterly 69 (4): 979–86.

Baum

Lawrence

. 2006. Judges and Their Audiences: A Perspective on Judicial Behavior. Princeton: Princeton University Press.

Bell

Griffin B.

1966. “Style in Judicial Writing.” Journal of Public Law 15:214–19.

Berkowitz

Daniel M.

Clay

Karen

. 2006. “The Effect of Judicial Independence on Courts: Evidence from the American States.” Journal of Legal Studies 35 (2): 399–400.

Berry

William D.

Ringquist

Evan J.

Fording

Richard C.

Hanson

Russell L.

1998. “Measuring Citizen and Government Ideology in the American States, 1960-93.” American Journal of Political Science 42:327–48.

Black

Ryan C.

Treul

Sarah A.

Johnson

Timothy R.

Goldman

Jerry

. 2011. “Emotions, Oral Arguments, and Supreme Court Decision Making.” Journal of Politics 73:572–81.

Bonneau

Chris

. 2004. “Patterns of Campaign Spending and Electoral Competition in State Supreme Court Elections.” Justice System Journal 25:21–38.

Bonneau

Chris

. 2005. “Electoral Verdicts: Incumbent Defeats in State Supreme Court Elections.” American Politics Research 33:818–41.

10.

Bonneau

Chris

Hall

Melinda G.

2009. In Defense of Judicial Elections. New York: Routledge.

11.

Brace

Paul

Boyea

Brent D.

2008. “State Public Opinion, the Death Penalty, and the Practice of Electing Judges.” American Journal of Political Science 52:360–72.

12.

Brace

Paul

Langer

Laura

Hall

Melinda G.

2000. “Measuring the Preferences of State Supreme Court Judges.” Journal of Politics 62:387–413.

13.

Brace

Paul

Yates

Jeff

Boyea

Brent D.

2012. “Judges, Litigants, and the Design of Courts.” Law & Society Review 46:497–522.

14.

Brown v. Board of Education II, 349 U.S. 294. 1955.

15.

Cann

Damon M.

2007. “Beyond Accountability and Independence: Judicial Selection and State Court Performance.” Judicature 90:226–32.

16.

Cann

Damon M.

Wilhelm

Teena

. 2011. “Case Visibility and the Electoral Connection in State Supreme Courts.” American Politics Research 39:557–81.

17.

Choi

Stephen J.

Gulati

G. Mitu

Posner

Eric A.

2010. “Professionals or Politicians: The Uncertain Empirical Case for an Elected Rather than Appointed Judiciary.” Journal of Law, Economics, & Organization 26:290–336.

18.

Clark

Tom S.

Carrubba

Clifford J.

2012. “A Theory of Opinion Writing in a Political Hierarchy.” Journal of Politics 74:584–603.

19.

Coffey

Daniel

. 2005. “Measuring Gubernatorial Ideology: A Content Analysis of State of the State Speeches.” State Politics & Policy Quarterly 5:88–103.

20.

Cohen

Jacob

. 1988. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale: Lawrence Erlbaum.

21.

Coleman

Brady

Phung

Quy

. 2010. “The Language of Supreme Court Briefs: A Large-scale Quantitative Investigation.” Journal of Appellate Practice & Process 11:75–103.

22.

Collins

Paul M.

Jr.

2008. Friends of the Court: Interest Groups and Judicial Decision Making. New York: Oxford University Press.

23.

Corley

Pamela C.

Collins

Paul M.

Jr. Calvin

Bryan

. 2011. “Lower Court Influence on U.S. Supreme Court Opinion Content.” Journal of Politics 71:31–44.

24.

Dunn

Olive J.

1961. “Multiple Comparisons among Means.” Journal of the American Statistical Association 56:52–64.

25.

Emmert

Craig F.

Glick

Henry R.

1988. “The Selection of State Supreme Court Justices.” American Politics Quarterly 16:445–65.

26.

Farber

Daniel A.

2005. “Supreme Court Selection and Measures of Past Judicial Performance.” Florida State University Law Review 32:1175–96.

27.

Faul

Franz

Erdfelder

Edgar

Buchner

Axel

Lang

Albert-Georg

. 2009. “Statistical Power Analyses Using G*Power 3.1: Tests for Correlation and Regression Analyses.” Behavior Research Methods 41:1149–60.

28.

Finn

Seth

. 1985. “Unpredictability as Correlate of Reader Enjoyment of News Articles.” Journalism Mass Communication Quarterly 62:334–45.

29.

Flesch

Rudolph

. 1948. “A New Readability Yardstick.” Journal of Applied Psychology 32:221–33.

30.

Garner

Bryan A.

2002. The Elements of Legal Style. 2nd ed. New York: Oxford University Press.

31.

George

Joyce J.

2006. The Judicial Opinion Writing Handbook. 5th ed. New York: William S. Hein.

32.

Geyh

Charles

. 2003. “Why Judicial Elections Stink.” Ohio State Law Journal 64:43–79.

33.

Gibson

James L.

1978. “Judges’ Role Orientations, Attitudes, and Decisions: An Interactive Model.” American Political Science Review 72:911–24.

34.

Gibson

James L.

2012. Electing Judges: The Surprising Effects of Campaigning on Judicial Legitimacy. Chicago: Chicago University Press.

35.

Glick

Henry

Emmert

Craig

. 1987. “Selection Systems and Judicial Characteristics: The Recruitment of State Supreme Court Justices.” Judicature 70:228–35.

36.

Goelzhauser

Greg

. 2011. “Diversifying State Supreme Courts.” Law Society Review 45:761–81.

37.

Goelzhauser

Greg

. 2012. “Accountability and Judicial Performance: Evidence from Case Dispositions.” Justice System Journal 33:249–61.

38.

Goldschmidt

Jona

. 1994. “Merit Selection: Current Status, Procedures, and Issues.” University of Miami Law Review 49:1–126.

39.

Hall

Melinda G.

1987. “Constituent Influence in State Supreme Courts: Conceptual Notes and a Case Study.” Journal of Politics 49:1117–24.

40.

Hall

Melinda G.

1995. “Justices as Representatives: Elections and Judicial Politics in the American States.” American Politics Quarterly 23:485–503.

41.

Hall

Melinda G.

2001. “State Supreme Courts in American Democracy: Probing the Myths of Judicial Reform.” American Political Science Review 95:315–30.

42.

Hall

Melinda G.

2007. “Competition as Accountability in State Supreme Court Elections.” In Running for Judge, ed. Streb

Matthew

New York: New York University Press. Pp. 165-85.

43.

Hurwitz

Mark S.

Lanier

Drew N.

2003. “Explaining Judicial Diversity: The Differential Ability of Women and Minorities to Attain Seats on State Supreme and Appellate Courts.” State Politics & Policy Quarterly 3:329–52.

44.

Hurwitz

Mark S.

Lanier

Drew N.

2008. “Diversity in State and Federal Appellate Courts: Change and Continuity across 20 Years.” Justice System Journal 29:47–70.

45.

Kaplow

Louis

. 1992. “Rules versus Standards: An Economic Analysis.” Duke Law Journal 42:557–629.

46.

Langer

Laura

. 2002. Judicial Review in State Supreme Courts: A Comparative Study. Albany: State University of New York Press.

47.

Law

David S.

Zaring

David

. 2010. “Law versus Ideology: The Supreme Court and the Use of Legislative History.” William and Mary Law Review 51:1653–747.

48.

Lax

Jeffrey R.

Cameron

Charles M.

2007. “Bargaining and Opinion Assignment on the US Supreme Court.” Journal of Law, Economics, & Organization 23:276–302.

49.

Lebovits

Gerald

. 2004. “Free at Last from Obscurity: Clarity—Part 2.” New York Bar Association Journal 76: 64.

50.

Lowrey

Tina M.

2006. “The Relation between Script Complexity and Commercial Memorability.” Journal of Advertising 35:7–15.

51.

Marshall

William P.

2004. “Be Careful What You Wish For: The Problems with Using Empirical Rankings to Select Supreme Court Justices.” Southern California Law Review 78:119–36.

52.

Milne

George R.

Culnan

Mary J.

Greene

Henry

. 2006. “A Longitudinal Assessment of Online Privacy Notice Readability.” Journal of Public Policy & Marketing 25:238–49.

53.

Nelson

Michael J.

2013. “Elections and Explanations: Judicial Elections and the Readability of Judicial Opinions.” Working Paper available online at http://mjnelson.wustl.edu/research.html.

54.

Owens

Ryan J.

Wedeking

Justin P.

2011. “Justices and Legal Clarity: Analyzing the Complexity of U.S. Supreme Court Opinions.” Law & Society Review 45:1027–61.

55.

Owens

Ryan J.

Wedeking

Justin

Wohlfarth

Patrick C.

2013. “How the Supreme Court Alters Opinion Language to Evade Congressional Review.” Journal of Law and Courts 1:35–59.

56.

Posner

Richard A.

1993. “What Do Judges Maximize? (The Same Thing Everybody Else Does).” Supreme Court Economic Review 3:1–41.

57.

Posner

Richard A.

2005. “Judicial Behavior and Performance: An Economic Approach.” Florida State University Law Review 32:1259–79.

58.

Reilly

Shauna

Richey

Sean

. 2011. “Ballot Question Readability and Roll-off: The Impact of Language Complexity.” Political Research Quarterly 64:59–67.

59.

Schroeder

Nicholas

Aggarwal

Raj

Gibson

Charles

. 1991. “Financial Reporting by Japanese Firms on the NYSE: An Analysis of Linguistic Content.” Management Information Review 31:233–51.

60.

Schroeder

Nicholas

Gibson

Charles

. 1990. “Readability of Management’s Discussion and Analysis.” Accounting Horizons 4:78–87.

61.

Solum

Lawrence B.

2005. “A Tournament of Virtue.” Florida State University Law Review 32:1365–1400.

62.

Spriggs

James F.

II . 1997. “Explaining Federal Bureaucratic Compliance with Supreme Court Opinions.” Political Research Quarterly 50:567–93.

63.

Squire

Peverill

. 2008. “Measuring the Professionalization of U.S. State Courts of Last Resort.” State Politics & Policy Quarterly 223–38.

64.

Staton

Jeffrey K.

2010. Judicial Power and Strategic Communication in Mexico. New York: Cambridge University Press.

65.

Staton

Jeffrey K.

Romero

Alexia

n.d. “Clarity and Compliance in the Inter-American Human Rights System.” Working Paper available online at http://paperroom.ipsa.org/papers/paper_26179.pdf.

66.

Staton

Jeffrey K.

Vanberg

Georg

. 2008. “The Value of Vagueness: Delegation, Defiance, and Judicial Opinions.” American Journal of Political Science 52:514–9.

67.

Stewart

Joseph

Jr. Heck

Edward V.

1987. “Caseloads and Controversies: A Different Perspective on the Overburdened U.S. Supreme Court.” Justice System Journal 12:370–83.

68.

Terris

Fay

. 1949. “Are Poll Questions Too Difficult?” Public Opinion Quarterly 13:314–19.

69.

Torres-Spelliscy

Ciara

Chase

Monique

Greenman

Emma

. 2008. Improving Judicial Diversity. New York: Brennan Center for Justice at New York University Law School.

70.

Vickrey

William C.

Denton

Douglas G.

Jefferson

Wallace B.

n.d. Opinions as the Voice of the Court: How State Supreme Courts Can Communicate Effectively and Promote Procedural Fairness. Williamsburg, VA. National Center for State Courts.

71.

Vining

Richard L.

Jr. Wilhelm

Teena

. 2011. “Measuring Case Salience in State Courts of Last Resort.” Political Research Quarterly 64:559–72.

72.

Wald

Patricia M.

1995. “Rhetoric of Results and the Results of Rhetoric.” University of Chicago Law Review 62:1371–420.

73.

Wanderer

Nancy A.

2002. “Writing Better Opinions: Communicating with Candor, Clarity, and Style.” Maine Law Review 54:47–70.

74.

Wandersee

James H.

Clary

Renee M.

2007. “Learning the Trail: A Content Analysis of a University Arboretum’s Exemplary Interpretive, Science Signage System.” American Biology Teacher 69:16–23.

75.

Weinberg

Micah

. 2010. “Measuring Governors’ Political Orientations Using Words as Data.” State Politics & Policy Quarterly 10:96–109.

76.

Wydick

Richard C.

2005. Plain English for Lawyers. 5th ed. Durham: Carolina Academic Press.

77.

Yates

Jeff

Tankersley

Holley

Brace

Paul

. 2010. “Assessing the Impact of State Judicial Structures on Citizen Litigiousness.” Political Research Quarterly 63:797–810.