While numerous studies exist about how the mass public perceives the U.S. Supreme Court, little is known about how elite lawyers with specialized legal expertise and an acute understanding of the U.S. Supreme Court perceive and assess the Court. Among these “legal elites,” is the Court perceived as a political and ideologically driven institution, as legalistic and capable of objectively producing legal outputs, or something in between? While Congress and the Presidency are often seen by members of the mass public as divisive, ideologically polarizing, and uncivil, conventional scholarly wisdom suggests that the Supreme Court is seen as relatively more objective, legalistic, and above the political fray. Since many Americans do not possess a thorough understanding of the Court's policymaking, they are less aware of the extent to which the Court can be political and ideological (e.g., Reference Epstein and KnightEpstein and Knight 1998; Maltzman et al. 2000; Segal and Spaeth 2002).Footnote 1 Given that legal elites—here, lawyers admitted to federal appellate bars—do not suffer from similar informational disadvantages, it is interesting in and of itself to analyze where they fall along the “law versus politics” spectrum pertaining to the Court. Legal elites are socialized to respect legal principles and their application, but through professional practice and elevated attention to what the Court is doing, legal elites experience and observe the political nature of judicial decision making.
Moreover, we confront the following empirical foundation: the legal elites studied in this article almost universally perceive the Court as legitimate. Our theoretical innovation is to develop competing models that could be the result of this characteristic. On one hand, Gibson and Caldeira's (Reference Gibson and Caldeira2009a, Reference Gibson and Caldeira2009b) positivity theory implies that strong pre-existing legitimacy orientations may induce elites to hold near-universal apolitical perceptions of the Court, which should hold regardless of ideological disagreement with the Court's policymaking. On the other hand, legitimacy and perceptions of the Court's decision making and role in government may be more separable for legal elites than for average citizens. A motivated reasoning perspective (e.g., Reference KundaKunda 1990; Reference Taber and LodgeTaber and Lodge 2006) implies that the more legal elites disagree with the ideological direction of the Court's policymaking, the more “political” they will perceive the Court—including how it makes decisions and its role in American politics. Perceptions of the Court's decision making and role, but not legitimacy orientations, are an outlet for disagreement with Court policymaking. We test these competing models by analyzing survey data from the 2005 Annenberg Supreme Court study, which is a nationally representative survey of lawyers admitted to the U.S. Supreme Court and Courts of Appeals bars.
Understanding how legal elites perceive the Court is valuable for several reasons and possesses important sociolegal implications. Since legal elites are opinion leaders and are in positions of influence as cue givers, we can gain insight into opinion dynamics by explaining the foundations of elite attitudes toward the Court. Legal elites possess more information about the Court than the mass public. They also obtain information through different channels than even the most highly sophisticated masses; they not only consume news about the Court but also participate in the legal process and are members of legal networks. As a result, the foundations of their perceptions of the Court's business will be more informed.
Furthermore, beliefs about how the Court makes decisions might also have an effect on how legal elites craft legal arguments. This group may utilize information they possess about the political nature of the Court's decision making and role in government in order to develop more persuasive arguments. These elites are also members of an important audience to the Court, which is likely more inclined to pay more attention to their views compared to the mass public (e.g., Reference Baum and DevinsBaum 2006; Baum and Devins 2010; see Reference FriedmanFriedman 2009).
Finally, our study offers a rare opportunity to test models of political judgment with an elite sample and to offer a general examination of the extent to which motivated reasoning processes may be present even at the very upper tiers of the political knowledge distribution. This is important because questions remain about the extent to which prima facie biases in political judgment are truly due to biased reasoning processes (e.g., Bolsen et al. N.d.) or whether they are simply failures of heuristic forms of judgment. The average citizen may use perceptions of ideological disagreement as a low effort cue, that is, an affect heuristic (Reference Finucane, Peters, Slovic, Schneider and ShanteauFinucane, Peters, and Slovic 2003; Reference KahnemanKahneman 2011) for forming perceptions about the political nature of the Court, but one might wonder about the relationship between these constructs under conditions of complete information. If such relationships exist even when an individual has considerable information relevant to making a judgment, then motivated forms of reasoning likely provide a better explanation than failures of heuristic judgment (e.g., Reference Lavine, Johnston and SteenbergenLavine, Johnston, and Steenbergen 2012; Reference Taber and LodgeTaber and Lodge 2006). Our study is significant, then, given the rarity with which students of political behavior have the opportunity to conduct tests with samples possessing high levels of information relevant to a given judgment task.
Legal Elites, Legitimacy, and Political Perceptions
Legal elites possess high levels of perceived legitimacy for the Supreme Court as an institution, an empirical starting point that motivates our elaboration of competing expectations about its consequences. The 2005 Lawyer Component of the Annenberg Supreme Court Survey (which we discuss in more detail below) contained 859 elite lawyer interviews and included core indicators of institutional legitimacy tapping a willingness of individuals to reject alterations to institutions' functions (e.g., Reference Caldeira and GibsonCaldeira and Gibson 1992; Reference Gibson, Caldeira and Spence2003a). It also included indicators of institutional trust.Footnote 2 First, when asked whether they agreed or disagreed that “Congress should take away the right of the Supreme Court to decide certain types of controversial issues,” 93 percent expressed disagreement (80 percent strongly disagreed, while 13 percent disagreed somewhat), while just 6 percent expressed agreement (1 percent either refused to answer or said “don't know”). Second, when asked whether they agreed or disagreed with the notion that “The Supreme Court should have the right to say what the Constitution means, even when the majority of people disagree with the Court's decision,” 95 percent expressed agreement (76.4 percent strongly agreed, and 18.4 percent agreed somewhat), while just 4 percent expressed disagreement (1 percent refused or said “don't know”). Moving to trust in the Court, 81 percent expressed agreement that the Court “can usually be trusted to make decisions that are right for the country as a whole,” and 90 percent possessed either a fair amount or a great deal of “trust [in] the Supreme Court to operate in the best interests of the American people.”
As is evident, perceived legitimacy among these legal elites is extremely high—and much higher than it is for the mass public at large. The General Public Component of the Annenberg Supreme Court Study did not contain the first two legitimacy items discussed above.Footnote 3 However, data from a 2005 survey analyzed by Gibson (2007) show that 51.4 percent of the general public provided a supportive response to the survey item that the “Court's right to hear certain issues should be reduced.”Footnote 4 Data from 2001 (Reference Gibson, Caldeira and SpenceGibson, Caldeira, and Spence 2003a, Reference Gibson, Caldeira and Spence2003b) show that 69.2 percent of the mass public agreed that the “Supreme Court should have the right to say what the Constitution means, even when the majority of people disagree with the Court's decision.” Thus, on these two foundational legitimacy items, legal elites show substantially greater support than the mass public. The same pattern follows for trust in the Court, though the differences are not quite as large. In the General Public Component of the Annenberg survey, 76 percent expressed agreement that the Court “can usually be trusted to make decisions that are right for the country as a whole,” and 75 percent possessed either a fair amount or a great deal of “trust [in] the Supreme Court to operate in the best interests of the American people.”
These patterns most likely emerge due to legal elites being more predisposed to support democratic values and institutions (Reference McCloskeyMcCloskey 1964; Reference McCloskey and BrillMcCloskey and Brill 1983; Reference Prothro and GriggProthro and Grigg 1960; Reference StoufferStouffer 1955). Also, legal elites are socialized into their profession through both law school and professional associations. Law school inculcates norms and principles of the profession and teaches the proper performance of professional roles (Reference CramtonCalmore 2003; Cramton 1977; Reference Erlanger and KlegonErlanger and Klegon 1978). Additionally, legal elites' interactions with courts—including the Supreme Court—could motivate acceptance of the Court as a legitimate institution due to repeated exposure to legitimating symbols from the Court (e.g., Reference Gibson and CaldeiraGibson and Caldeira 2009a, Reference Gibson and Caldeira2009b). Legal elites also have a vested interest in the legitimacy of the Court and the judicial system in general, since it helps define their professional identity and role as participants in the system.
Political Perceptions of Decision Making and Role in American Government
While a great deal of consensus seems to emerge regarding perceived legitimacy, to what degree do elites perceive the Court as “political” in terms of how it makes decisions and regarding its role in government? We first discuss the three types of perceptions—which will be our key dependent variables in our empirical analysis—on which we focus.
1. Perceptions of decision making
Are the Court's rulings and the justices' votes influenced by law and objective considerations (vis-à-vis a legal model), by political and ideological considerations (vis-à-vis the attitudinal model), or a combination of the two? Such perceptions tap whether elites subscribe to the so-called “myth of legality” (e.g., Baird and Gangl Reference Baird and GanglGibson, Caldeira, and Spence 2006; Reference CaseyCasey 1974; Reference GibsonGibson and Caldeira 2011), that is, a legalistic story of decision making where justices reason impartially without regard to party or their personal preferences.
2. Perceptions of judicial activism
“Judicial activism” is a loaded and often misunderstood concept (see, e.g., Reference Lindquist and CrossLindquist and Cross 2009). We seek to tap perceptions that the Court is overstepping its bounds regarding the powers of the other branches of federal government, the powers of state governments, and deciding issues beyond its reach.
3. Perceptions of external political influence
To what extent are the Court's rulings influenced by Congress, the president, and public opinion? This facet bears some relation to the first perception above, but it also taps broader perceptions of judicial independence, that is, whether the Court is susceptible to external political forces or free from such constraints?
Competing Theoretical Models
Model 1: Legitimacy as a Frame for Apolitical Perceptions of the Supreme Court
There are two competing ways of thinking about the consequences of near-universal legitimacy among legal elites for these three perceptions discussed above. First, legitimacy may serve as a powerful lens or frame through which perceptions of a political versus apolitical Court are assessed. Because legal elites possess such strong loyalty toward the Court, they may perceive the Court as generally objective and legalistic, as acting in an appropriate manner in American government writ large (i.e., not activist), and as immune from external political influence. In line with this model, such perceptions should hold regardless of whether one agrees or disagrees with the ideological direction of the Court's policymaking.
The logic underlying this model is rooted in Gibson and Caldeira's (Reference Gibson and Caldeira2009a, Reference Gibson and Caldeira2009b); Gibson, Caldeira, and Spence Reference Gibson, Caldeira and Spence2003b “positivity theory,” which argues, inter alia, that pre-existing stores of legitimacy serve as a lens through which individuals perceive aspects of the Court, including controversial decisions such as Bush v. Gore (Reference Gibson, Caldeira and SpenceGibson, Caldeira, and Spence 2003b; see also Reference Nicholson and HowardNicholson and Howard 2003) and confirmation politics (Reference Gibson and CaldeiraGibson and Caldeira 2009a, Reference Gibson and Caldeira2009b). Legal elites may be especially susceptible to positivity theory's mechanisms. Gibson and Caldeira posit that the more one is aware of and observes the Court, the more one is exposed to the legitimizing symbols and messages emanating from the Court; “to know the Court is to love it” (see also Reference Caldeira and GibsonCaldeira and Gibson 1992; Reference CaseyCasey 1974; Gibson et al. 1998). Because individuals are “susceptible to … the influence of strong legitimizing legal symbols, they tend to wind up accepting the argument that courts are different from other political institutions and that ‘politics’ plays a limited role in the judicial process. Suspicions about partisan and ideological influences on legal processes are dispelled, owing to the frame created by standing commitments to the Court” (Reference Gibson and CaldeiraGibson and Caldeira 2009a: 140). The model suggests the following hypotheses:
H1a: Near-universal levels of institutional legitimacy among legal elites imply that we should observe near-universal apolitical perceptions of the Supreme Court with respect to judicial decision making, judicial activism, and external political influence.
H1b: Apolitical perceptions of the Court will result regardless of whether legal elites agree or disagree with the ideological direction of the Court's policymaking. Ideological disagreement with the Court will have no impact on the magnitude of political perceptions.
Model 2: Ideological Foundations of Political Perceptions of the Supreme Court
Model 2 suggests that legal elites' perceptions of the Court's decision making and its role in American government are grounded in ideological considerations. The more individuals disagree with the policy direction of the Court's rulings, the more political they will perceive various facets the Court. While legal elites are unwilling to delegitimize the structure and functioning of the Court (given the near-universal perceptions of “core” institutional legitimacy), Model 2 contends that they are motivated by ideology to heavily scrutinize disagreeable decisions and thus the mechanism by which a particular Court and its members reached such decisions.
Perhaps the most important insight to emerge from recent studies of political reasoning and judgment is that citizens' pre-existing attitudes toward a political object shape their thinking about that object. In other words, the evaluation of new evidence is shaped by priors, rather than being independent of them (Reference Lodge and TaberLodge and Taber 2013; Reference Lord, Ross and LepperLord, Ross, Lepper 1979; Reference Taber and LodgeTaber and Lodge 2006). “Affect” typically comes first, and expressed reasons for that affect emerge as post hoc rationalizations constructed to be persuasive and justifiable to the self and others (e.g., Reference HaidtHaidt 2012; Reference KahnemanKahneman 2011; Reference Lodge and TaberLodge and Taber 2013; Reference Mercier and SperberMercier and Sperber 2011; Reference Taber and LodgeTaber and Lodge 2006; Reference ZajoncZajonc 1980). As Reference Lerner, Tetlock, Schneider and ShanteauLerner and Tetlock (2003: 434) argue, “As long as people are concerned about maintaining their identities as moral, competent beings, a central function of thought is making sure that one acts in ways that can be persuasively justified or excused to observers.”
Assuming that legal elites hold meaningful ideological preferences, we expect their affect toward the Court to be shaped by the Court's policymaking. A state of ideological disagreement elicits a conflict between high pre-existing stores of legitimacy toward the Court as an institution and a negative attitude toward the contemporary instantiation of the Court. In other words, there is a disconnect between perceptions of a legitimate Court and negative attitudes deriving from the current Court's policymaking. Legal elites face a dilemma: Like most human beings, they need an outlet to rationalize and justify their negative affect, but they cannot delegitimize the Court as an institution because they have a vested interest in that system. They are participants in the judicial system, which defines their professional identity and role, so delegitimizing the system in which they operate would be indirectly delegitimizing their own professional role. Thus, Model 2 contends that ideological disagreement will manifest itself by elevating political perceptions of the Court's decision making and role in American government. Elites in ideological disagreement with the Court can maintain high levels of legitimacy toward the Court as an institution and justify their negative affect toward the current Court by scrutinizing and ultimately denigrating the Court's decision making and its role in American government. Conversely, legal elites in ideological agreement with the Court's policymaking may bolster and further justify these attitudes by strongly endorsing both how the Court makes decisions and its role in government. From this logic, we hypothesize the following:
H2a: As ideological disagreement with Supreme Court policymaking increases, perceptions of the Court's decision making and role in government will become more political.
Note that Hypotheses 1b and 2a do not differentiate how the effect of ideological disagreement might vary across the three perceptions—perceptions of decision making, activism, and external influence. We further hypothesize that ideological disagreement may have its largest effect on perceptions of decision making and less of an effect on perceptions of activism and external political influence. In other words, Model 1 may better explain perceptions of activism and external political influence, while Model 2 may better explain perceptions of decision making. The reason for this goes back to the “legitimacy frame” logic underlying Model 1. Because activism and judicial independence are so closely connected to the core of institutional legitimacy (touching upon the Court's power and authority in the political system) and because of legal elites' vested interest in judicial system, high degrees of institutional legitimacy may trump ideological disagreement. Even among those in ideological disagreement with the Court, institutional loyalty toward the Court's key functions and powers may lead legal elites to defend the Court and its role in the political system.
However, we suspect that when it comes to perceptions of decision making, ideological disagreement will still register quite potent effects via the logic of motivated reasoning. For those in ideological disagreement with the Court, perceptions of decision making will serve as an outlet to scrutinize the Court, which serves to justify any negative affect toward a particular Court's disagreeable rulings. Concurrently, they may still endorse the Court's legitimacy, in a more abstract sense, in the political system.
This discussion poses a compelling puzzle: How could legal elites, even when they disagree with the Court's policymaking, maintain high stores of institutional legitimacy while simultaneously perceiving the Court to be (1) quite political in its decision making (2) not activist or susceptible to external political influences? The issue bears some resemblance to Gibson and Caldeira's (2011) finding that even though many in the general public recognize that the justices do not strictly follow a “mechanical jurisprudence” model of judging and instead possess ample decision-making discretion, legitimacy remains quite high because individuals believe that the justices use their discretion in a principled manner. For legal elites, the core of institutional legitimacy, which they possess at near unanimous levels, and perceptions of the Court's decision making may be separable constructs. Those in ideological disagreement may rationalize their negative affect toward the Court solely via perceptions of decision making connected to the current Court while simultaneously ascribing high legitimacy to the Court in general. Why are perceptions of activism and external political influence immune to ideological disagreement? Legal elites in ideological disagreement with the Court may find it more difficult to credibly use perceptions of activism and external political influence as viable outlets for their negative affect. Given their high levels of legitimacy, they likely realize that the actual scope of the Court's actions and the functions it carries out are appropriate and within its power realm. Therefore, legal elites are very unlikely to perceive the Court as activist under any conditions. Moreover, elites may have stronger perceptions (relative to the general public) that the Court is by design actually quite insulated from direct political influence external political influences. In sum, for elites in ideological disagreement, denigrating the political and ideological manner in which the Court makes decisions is a much more viable and realistic outlet for rationalizing their negative affect compared to denigrating the Court for being activist or susceptible to external political influence. From this logic, we hypothesize the following:
H3: Ideological disagreement will have a greater impact on political perceptions of decision making than it will on perceptions of activism and external political influence.
Empirical Analysis
To test these hypotheses, we analyze data from the 2005 Lawyer Component of the Annenberg Supreme Court Survey. The survey was designed to be representative of lawyers admitted to practice before the U.S. Supreme Court and the U.S. Courts of Appeals. 859 lawyers were interviewed between March 18, 2005 and May 16, 2005.Footnote 5
Perceptions of Supreme Court Decision Making
We examine five survey items that tap perceptions of decision making. Exact question wording for all survey items used are included in Supporting Information Appendix A. The first four items ask whether respondents agree or disagree (on a four-point scale) with the following: (1) Justices “know the result they want in a case and craft their legal reasoning to reach that result.” The item taps the extent to which individuals view legal justifications as post-hoc rationalizations of ideologically-driven reasoning. (2) Justices “closely follow the Constitution, the law and the precedents in deciding cases.” (3) The Court “gets too mixed up in politics.” (4) The “decisions of the Supreme Court favor some groups more than others.” The last item (5) asks whether individuals perceive the Court as “fair and objective in its rulings” (=0) or “sometimes politically motivated” (=1).
A brief explanation is in order given that some of these measures have been used in institutional legitimacy scales (e.g., Reference Gibson, Caldeira and SpenceGibson, Caldeira, and Spence 2003a; Reference Bartels and JohnstonBartels and Johnston 2013). Though Reference Caldeira and GibsonCaldeira and Gibson's (1992) original conceptualization of institutional legitimacy centered on the public's rejection of fundamental alterations to the structure, power, or functioning of the institution, later conceptualizations and measures included trust and perceptions of politicization versus impartiality (e.g., Reference Gibson, Caldeira and SpenceGibson, Caldeira, and Spence 2003a). Two of the four legitimacy items we discussed early in the article centered on “rejection of fundamental alterations to the institution” (e.g., jurisdiction stripping), while the other two centered on trust in the Court. Recall that Reference GibsonGibson (2011) has called into question whether trust measures truly tap the legitimacy concept. To demonstrate that our measures of political perceptions of decision making are sufficiently distinct from institutional legitimacy, we created a summative scale of the two “core” legitimacy items and correlated them with all five of the measures discussed above. The bivariate correlations are small and range from −0.05 (“sometimes politically motivated”) to −0.15 (“mixed up in politics”). When we build a summative scale of perceptions of decision making (which we discuss further below), the correlation between the two scales is just −0.14. Thus, the evidence suggests that perceptions of decision making are sufficiently different from the core institutional legitimacy items.Footnote 6
Figure 1a depicts the percentage of the sample who gave a “political” response to these five survey questions.Footnote 7 Though roughly 95 percent of the sample reported high levels of perceptions of Court legitimacy, a significant share of legal elites report that they think of the Court as a political institution. The variation across these five items is also compelling. Over 65 percent of the sample reported that they “agree” or “strongly agree” that the justices craft their legal reasoning around predetermined decisions. Nearly 65 percent believe the Court is politically motivated in its decision making, and almost 50 percent of the sample feels that the Supreme Court is more favorable to some groups than others. Just 35 percent report a belief that the Court is “mixed up” in politics, and a small percentage—18 percent—believe the court disregards the law in decision making.Footnote 8
Figure 1. Elite Perceptions of Political Decision Making, Activism, and External Political Influence Regarding the U.S. Supreme Court.
These descriptive statistics present some interesting patterns and mixed verdicts in relation to Hypothesis 1a. While respondents report perceptions of legitimacy almost unanimously, which Model 1 posits will serve as a frame for elite perceptions of the Court, legal elites view the Court as quite political on many items (refuting Hypothesis 1a), yet most believe that the Court generally follows the law (in support of Hypothesis 1a). To explore this issue further, we focus on the countervailing patterns provided by the “follow law” and “politically motivated” survey items. We cross these two items and retrieve four profiles combining low (“follows law” and “fair/objective”) and high levels (“doesn't follow law” and “politically motivated”) of both items. The results are presented in Figure 1b.
The horizontal axis of Figure 1b orders, from left to right, perceptions ranging from least political (“Follows law/Fair, objective”) to most political (“Doesn't follow law/Politically motivated”). The results essentially uncover three broad categories of legal elites, with the fourth including a minute share of elites and a counterintuitive pattern. First, 30 percent of legal elites in the sample believe that the Court closely follows the law in deciding cases and is fair and objective in its rulings. These individuals seemingly subscribe to the “legal model”—justices' ideological preferences are detached from decision making and justices reason impartially. Nearly half the sample (47.5 percent) believes the Court closely follows the law yet is sometimes political motivated in its rulings. These individuals seemingly subscribe to a mixed, “legal realism” view—the law matters but sometimes justices act on the basis of their ideological preferences.
Nearly 16 percent of legal elites perceive the Court in the most political terms (the fourth bar in Figure 1b). These individuals do not believe the Court closely follows the law and perceive that the Court is politically motivated. Individuals in this group seemingly view the justices as akin to “legislators in robes.” Finally, a very minute share of legal elites (<2 percent) have a counterintuitive perception of the Court as not following the law closely yet fair and objective in its rulings.
Given that political perceptions are not consistently low across all five decision-making perceptions and that roughly 65 percent see the Court as politically motivated, these descriptive results provide evidence against Hypothesis 1a for this facet. Perceptions of legitimacy are not strong enough to induce universally apolitical perceptions of decision making. A substantial share of elites perceives the Court's decision making as “political” to some degree.
Perceptions of Supreme Court Activism
In measuring activism, we are interested in perceptions that the Court is overstepping its bounds with regard to the powers of the other branches of federal government, the powers of state governments, and generally deciding issues beyond its purview. We rely on the following four survey items, each employing a four-point agree-disagree response format: (1) “The Court is taking on too many matters best left to Congress.” (2) “The Court's decisions are too intrusive on the rights of the state.” (3) “The Court is overturning too many federal laws.” (4) “The Court makes decisions in too many cases that are not within its mandate.”Footnote 9
Figure 1c illustrates the percentage of the sample giving “activist” responses (combining the “strongly” and “somewhat” categories). The story from Figure 1c is clear—most legal elites do not perceive the Court as activist. Unlike perceptions of decision making, these results are supportive of Hypothesis 1a. For each survey item, only a small minority of legal elites responded that the Court exceeds its powers with regards to popularly elected branches and state governments. Less than 26 percent of respondents agree that the Court is intrusive of states' rights, and less than 24 percent agree that the Court takes on too many issues better left to Congress. Legal elites are also generally unsupportive of the notion that the Supreme Court overturns too many federal laws, with only 11 percent agreeing with the statement. They are also unlikely to report that the Court often makes decisions outside of its mandate, with less than 18 percent in agreement.
These descriptive results suggest that legitimacy provides a strong frame by which individuals perceive the Court's role in American government. Though many legal elites see the Court in political terms at least to some extent, most defend the institution's role in the American political system and do not see it as exceeding its powers.
Perceptions of External Political Influence
To measure perceptions of external political influence, which also taps into judicial independence, we examine three survey items, which ask “to what extent the current Supreme Court is influenced by” (1) public opinion, (2) the President, and (3) Congress “when making their decisions.” Each item uses a four-point scale.
Figure 1d illustrates the percentage of legal elites who perceive that external actors influence the Court to a moderate or great extent. Generally, legal elites do not perceive a substantial level of external political influence on the Court, offering some support for Hypothesis 1a. For each external actor, roughly 30 percent perceive that the actor has at least a moderate influence on the Court. Public opinion is perceived as the least influential of the three, with less than 29 percent responding that public opinion influences the Court. The president is perceived as the most pronounced external influence on the Court's decision making, perhaps because of the president's nominating power. Still, less than 37 percent observe the president as exerting at least a moderate influence. The results reinforce our findings for perceptions of activism, though to a lesser extent. Just as nearly all legal elites perceive the Court as legitimate—and a supermajority of elites do not see the Court as activist—most elites see the court as free from external political influence.Footnote 10
Ideological Disagreement and Political Perceptions
We next pit Models 1 and 2—and associated Hypotheses 1b and 2, respectively—against each other by testing whether ideological disagreement with the Court's policymaking influences the three types of political perceptions of the Supreme Court. We will also test Hypothesis 3 regarding how the impact of ideological disagreement may vary depending on the type of perception (decision making versus activism and external political influence).
Perceptions of Decision Making
Given the nuanced nature of this perception, we generate two dependent variables. The first is a summative scale using all five survey items, which scale well (α = 0.70). The scale was recoded to range from 0 to 1, with higher values representing more political perceptions. The second dependent variable relies on our analysis from Figure 1b, where we crossed the “follows law” and “fair/objective versus politically motivated” items and found that three groups emerged: (1) “legalists” or subscribers to the legal model (“follows law” and “fair/objective”); (2) “legal realists/individuals with mixed perceptions” (“follows law” but “sometimes politically motivated”); and (3) those who think the Court is highly political (“doesn't follow law” and “sometimes politically motivated”).Footnote 11
For perceptions of activism and perceptions of external political influence, we build summative scales using the four activism items (α = 0.80) and the three external influence items (α = 0.72), respectively. Each scale was recoded to range from 0 to 1, with higher values representing greater perceptions of activism and external political influence. For the three summative scales, we estimate OLS regression models. For the three-category perceptions of decision making variable, we treat the categories as nominal and estimate a multinomial logit model.
Independent Variable of Interest: Ideological Disagreement with the Court
To measure ideological disagreement with the Court's policymaking, we rely on Reference Bartels and JohnstonBartels and Johnston's (2013: 190–92) measurement strategy for the American public. The measure seeks to capture whether individuals perceive they are in ideological disagreement with the Court's policymaking. Bartels and Johnston find that subjective ideological disagreement has deleterious consequences for perceptions of Supreme Court legitimacy. The measure entails combining one's self-reported ideology (liberal, moderate, or conservative)Footnote 12 with one's perception of the ideological nature of the Court's policymaking. To measure perceived Court ideology, the survey asked, “Judging by its recent decisions, do you think the Supreme Court is generally liberal (6.5 percent), generally conservative (49.9 percent), or is it making decisions more on a case-by-case basis (41.9 percent)?” The last response option presents some obstacles to the overall measure, which Reference Bartels and JohnstonBartels and Johnston (2013) discuss at length. The authors argue that the “case-by-case” option roughly captures a perception of “moderate” policymaking on the part of the Court—that the Court is not necessarily predisposed to decide a given case in a liberal or conservative direction.Footnote 13 We believe that legal elites, who are more aware of the Court's policymaking than the mass public, understand that the “case-by-case basis” option, when given in the context of liberal and conservative options, is reflective of a “moderate” response on a general left-right spectrum.
Bartels and Johnston measure this concept by crossing an individual's self-reported ideology (liberal, moderate, conservative) with his or her perceived ideology of the Court (liberal, moderate, conservative). There are four categories to the measure. (1) When the two variables match up perfectly (e.g., conservative respondent who perceives the Court as conservative), individuals are in strong agreement. (2) Reference Bartels and JohnstonBartels and Johnston (2013) demonstrate that liberal and conservative individuals who perceive the Court as moderate (or case-by-case basis) possess tacit agreement with the Court's policymaking that more closely resembles strong agreement than moderate disagreement. (3) Moderate individuals who perceive the Court as liberal or conservative are coded as having moderate disagreement. (4) Liberals and conservatives who perceive that the Court's outputs are ideologically incongruent (e.g., liberal individual perceiving a conservative Court) are coded as being in strong disagreement. In our regression models, we operationalize subjective ideological disagreement as nominal and dummy out the categories, specifying “strong agreement” as the baseline (excluded) category.
An additional and complementary test of the impact of ideological disagreement—which Reference Bartels and JohnstonBartels and Johnston (2013) employ in addition to using the measure above—is to interact individual ideology with perceptions of Supreme Court ideology to estimate the impact of ideology on Court perceptions conditional on how one perceives the Court's ideology. We report results from this analysis after discussing our primary results. We note the limitations of this additional analysis due to only 56 elites in our sample (6.5 percent) perceiving the Court as liberal. This issue did not affect Reference Bartels and JohnstonBartels and Johnston's (2013) analysis of the mass public sample, which had a sufficient sample size for each perception of Court ideology.
This last point raises an issue about the ideological context of the Supreme Court in 2005 during which this survey was conducted. As Reference Bartels and JohnstonBartels and Johnston (2013) discuss, while many often assume that the Court has been solidly conservative in the contemporary era, it has actually been more of a center-right Court when accounting for all of its decisions. And when looking at the high-profile, salient cases, the Court actually decided more cases in the liberal direction than the conservative direction from 1994–2005. For the mass public, depending on how closely one followed the rulings of the Court, there were arguably rational bases to perceive the Court as conservative, moderate, and even liberal during this period (Reference Bartels and JohnstonBartels and Johnston 2013). On the other hand, legal elites are more likely to be able to discount the disproportionate impact of salient outcomes and make inferences based on the totality of the Court's rulings. Therefore, it makes sense that we would see most legal elites perceiving the Court as conservative or moderate and very few perceiving it as liberal.Footnote 14
Unfortunately, the issues described above imply that our ideological disagreement variable does not contain as much information in the legal elites sample as it does in the mass public sample analyzed by Reference Bartels and JohnstonBartels and Johnston (2013). Ideological disagreement is defined almost exclusively against the backdrop of either moderate or conservative perceptions of the Court's ideology and not liberal perceptions of the Court. Importantly, there is still sufficient and meaningful variation ranging from low to high values of ideological disagreement (e.g., liberals who see the Court as conservative to conservatives to see the Court as conservative) with a good number of individuals in the middle as well (based on the large share of individuals who see the Court as moderate).
Control Variables
We control for a number of variables, including first, differential media exposure. Reference Johnston and BartelsJohnston and Bartels (2010) find that exposure to “sensationalist” (talk radio and cable news) relative to “sober” (newspapers and network news) media has deleterious consequences for judicial legitimacy. We may expect the same to hold true for the Court perceptions we examine here. We use Johnston and Bartels's measure:
Differential media exposure = (talk radio + cable news) – (newspaper + network news)
We recoded the measure to range from 0 to 1 (1 = high sensationalist). We also control for the number of years of practice by the lawyer, as we might expect political perceptions to change across time. The variable ranges from 1 to 64. We divide the variable by 10, such that a one-unit increase is a ten-year increment. Because there are conflicting expectations about how experience may influence such perceptions, we employed multiple functional forms—linear, quadratic, and a log transformation. For all models, a quadratic operationalization fit the data the best.Footnote 15 We also control for whether each lawyer was admitted to practice in the Supreme Court, of which 49 percent were. Finally, we account for general political trust. We rely on a survey item asking, “How much do you trust the federal government as a whole to operate in the best interests of the American people—a great deal, a fair amount, not too much, or not at all?” We recoded the variable to range from 0 to 1, where higher values represent greater trust.
Results for Perceptions of Decision Making
Regression model results for the four models are presented in Table 1, and post-estimation graphical depictions of our primary results are shown in Figure 2. The first two models (A and B) in Table 1 report the impact of ideological disagreement on perceptions of decision making using the summative scale of all five items (Model A) and the three categories derived from crossing the “follows law” and fair/objective versus politically motivated” items (Model B).
Figure 2. Post-Estimation Results Showing the Impact of Subjective Ideological Disagreement on Perceptions of Political Decision Making, Activism, and External Political Influence.
Table 1. Regression Models Testing the Impact of Subjective Ideological Disagreement on Perceptions of Political Decision Making, Activism, and External Political Influence
** p < 0.01; *p < 0.05; †p < 0.10.
What is abundantly clear from Models A and B in Table 1 is that subjective ideological disagreement exhibits a statistically and substantively significant impact on perceptions of decision making. The results provide strong support for Hypothesis 2 and Model 2's motivated reasoning logic; they provide evidence against Hypothesis 1a, and Model 1's legitimacy frame story, for this particular perception. The more individuals disagree with the direction of the Court's policymaking, the more political they perceive the Court's decision making. For the four-category ideological disagreement variable, recall that “strong agreement” is the baseline (excluded) category. In Table 1, model A, those in strong disagreement with the Court's policymaking possess significantly stronger political perceptions of decision making than those in strong agreement. And those in moderate disagreement also hold significantly stronger political perceptions than those in strong agreement. Elites who “tacitly agree” with the Court's policymaking (i.e., liberals and conservatives who perceive the Court as proceeding on a moderate, “case-by-case” basis) have very similar degrees of political perceptions of decision making as those in strong agreement do; the difference between these two groups is not statistically significant.
To visualize the strength of the relationship, Figure 2a plots post-estimation predicted values (
) of the perceptions of decision making scale as a function of subjective ideological disagreement, while controlling for the other independent variables.Footnote 16 As seen in Figure 2a, the more elites disagree with the ideological direction of Supreme Court policymaking, the more political their perceptions. This impact is fairly substantial. The move from strong agreement to strong disagreement leads to a change in the perceptions of decision making scale from 0.41 to 0.61—a difference representing about 1/5 of the entire scale. Following the motivated reasoning logic, the finding also suggests that political perceptions of how the Court makes decision will change among individuals as they transfer across ideological agreement/disagreement categories as a result of different types of decisions and longer-term membership change.Footnote 17
Moving to Model B in Table 1, which is a multinomial logit model, the first column of results within Model B represents the effects of the independent variables on the likelihood of being in the “mixed/legal realists” relative to the “legalists/legal model” category; the latter is the baseline category for the dependent variable in the multinomial logit model. The second column of results within Model B represents effects on the likelihood of being in the “political” relative to the “legalists/legal model” category. The results in Table 1 reveal that the likelihood of a legal elite being in both the “legal realist” and “political” categories, relative to the “legal model” category, increases significantly as ideological disagreement with the Court's policymaking increases. Figure 2b presents a much clearer substantive depiction of this effect. The graph reports the predicted probability of being in each dependent variable category as a function of subjective ideological disagreement, while averaging over the other independent variables.Footnote 18
The strongest pattern to emerge is that the predicted probability of holding a legalistic perception significantly and dramatically decreases as ideological disagreement increases—from 0.52 among the strong agreement group, down to 0.10 for the strong disagreement group. The differences in the probability of holding this perception between all six pairwise combinations of ideological disagreement groups are statistically significant.Footnote 19 Similarly, the probability of holding “realist/mixed” perceptions of the Court's decision making significantly increases as a function of ideological disagreement; all differences in the probability of this perception among ideological disagreement groups are statistically significant except for moderate disagreement versus tacit agreement. Parallel results emerge for holding strongly political perceptions of Court decision making—such perceptions increase as a function of ideological disagreement. The differences in the probability of holding such perceptions are statistically significant for each pairwise combination of ideological disagreement groups except for two—(1) strong agreement versus tacit agreement and (2) strong disagreement versus moderate disagreement.
Once again, the results from Model B in Table 1 and Figure 2b complement the results from Model A and Figure 2a and provide strong support for the motivated reasoning logic contained in Model 2. The results also suggest a strong basis for the argument that legalistic versus political perceptions of Court decision making among legal elites contain a strong situational component. So long as legal elites agree with the Court's rulings in general, they are inclined to perceive that the Court is legalistic and objective, but if the Court were to make a series of rulings antithetical to their ideological predispositions, our findings suggests that such perceptions would change course and become more political.
Results for Perceptions of Activism
The third model (C) in Table 1 reports the impact of ideological disagreement on perceptions of activism using the summative scale of all five survey items. Figure 2c plots predicted activism as a function of ideological disagreement. Unlike the prior finding with regard to perceptions of decision making, these results make it clear that ideological disagreement does not exhibit an impact on perceptions of activism among elites, thereby lending strong evidence to the “legitimacy frame” perspective in Model 1 and Hypothesis 1b. For activism, elites do not appear to engage in the motivated reasoning process posited in Model 2. The results also provide support for Hypothesis 3 regarding the differential impact of ideological disagreement. In the regression model, the joint impact of the ideological disagreement dummies is statistically insignificant (F = 1.05, p = 0.37). The coefficient for the strong disagreement (relative to strong agreement) dummy is marginally significant (at the alpha = 0.10 level), but the effect size is minute. None of the other pairwise differences between disagreement categories come close to statistical significance. Figure 2c makes it abundantly clear that there is no relationship to speak of between ideological disagreement and perceptions of activism among legal elites. Across levels of ideological disagreement, activism remains constant and at a fairly low level, once again, in support of Hypotheses 1a and 1b.
This finding obviously presents an interesting and stark contrast from the perceptions of decision making results, which also boosts support for Hypothesis 3 and the expectation that perceptions of activism do not seem to be motivated by ideological predispositions for legal elites. Without regard to ideological disagreement, elites' strong legitimacy frame compels them to defend the institution against charges that it both overreaches on the issues it takes up and encroaches on the territory of the other branches of government. Perceptions of a lack of activism are linked to this legitimacy frame more so than perceptions of decision making.
Results for Perceptions of External Political Influence
Model D in Table 1 reports the relationship between ideological disagreement and perceptions of external political influence. Figure 2d presents the post-estimation plot for this relationship. On the whole, the results show that ideological disagreement exhibits a moderate impact on perceptions of external influence—one that is stronger than the effect for activism but weaker than the effect for perceptions of decision making. The results provide some support for Hypothesis 2 and Model 2 and also Hypothesis 3 regarding the comparison of the effect of ideological disagreement across the three facets. In the regression model, the joint effect of the ideological disagreement dummies is statistically significant (F = 16.8, p < 0.001). Regarding pairwise comparisons between groups, elites in both strong and moderate disagreement with the Court have significantly stronger perceptions of external influence than those in strong agreement. Those in tacit agreement possess similar perceptions to those in strong agreement. And those in moderate disagreement possess significantly elevated perceptions of external influence compared to those in tacit agreement.Footnote 20
As seen in Figure 2d, the effect of ideological disagreement on perceptions of external influence is moderately strong, though certainly not overwhelming. There is an elevated level of perceptions that the Court is influenced by external political forces among those who express some degree of disagreement (strong or moderate) compared to those who express some degree of agreement (strong or tacit). This lends support to Hypothesis 2 and the notion that such perceptions are very difficult to separate from ideological predispositions. But ideological disagreement does not lead to a spike in such perceptions to the degree that it does for perceptions of decision making, which suggests support for Hypothesis 3. Like activism, perceptions of external influence touch a nerve because they are related to the key value of judicial independence. Our results suggest that ideological disagreement leads elites to perceive that the Court is increasingly influenced by external forces and judicial independence is compromised, but this ideological effect is not as overwhelming perhaps because of the strong value placed on judicial independence, which is fairly closely linked to the legitimacy frame.
Effects of Control Variables
While our focus has been on core theoretical propositions, our control variables also exhibit some interesting effects. First, differential media exposure has its most potent impact on perceptions of activism; the effect is fairly strong and statistically significant. The greater the balance of “sensationalist” over “sober” media exposure, the more activist are perceptions of the Court.Footnote 21 This result is sensible, since sensationalist media tend to frame judicial issues around debates over “judicial activism.” This variable exhibits a marginally significant impact on perceptions of decision making (Models A and B) and a statistically insignificant effect for external influence. On the whole, the number of years of practice has a minor to null impact across the four models. Recall that we modeled this as a quadratic effect, which we found was a statistically superior operationalization to alternative functional forms. While the effect is essentially nonexistent for perceptions of decision making, the squared term in the activism and external influence models is positive and statistically significant, meaning a u-shaped effect occurs. For activism, while the effect is not particularly large, post-estimation analysis shows that early on in one's career activism perceptions are moderate and lessen toward the middle of one's career; then, from the middle to the end of one's career, activism perceptions increase. The same dynamic holds for perceptions of external political influence. The effects could also be cohort effects, though further examination of this topic is beyond the scope of the current article. Whether or not one is admitted to practice before the Supreme Court has no effect on any of the dependent variables. General political trust significantly decreases political perceptions with respect to Court decision making and activism; political trust exhibits a statistically insignificant impact on perceptions of external influence.
Complementary Analysis
We now turn to results from our additional, complementary analysis showing how the impact of individual ideology on political perceptions of the Court is conditional on perceptions of Court ideology. Instead of using our subjective ideological disagreement measure as our independent variable of interest, we include: (1) the five-point measure for respondent's ideology, recoded from 0 to 1, where higher values are more conservative; (2) two dummy variables for whether elites perceive the Court as conservative or moderate (liberal is the baseline category); and (3) two interaction terms, between individual ideology and each of the two Court ideology dummies. We include the same control variables. For each facet of political perceptions of the Court, the model allows us to communicate how the marginal effect of individual ideology is conditional on how one perceives the Court's ideology. Figure 3 reports these conditional effects from OLS models for the three dependent variables using summative scales; the dot in each graph represents the conditional marginal effect of ideology, while the line through the dot represents the 95 percent confidence interval.Footnote 22 Full model results are included in Supporting Information Appendix B. Because of space considerations, we do not report results for the three-category political decision making dependent variable; the results are substantively comparable to those in using the political decision making scale.
Figure 3. Complementary Analysis Showing Marginal Effects of Individual Ideology on Political Perceptions of the Supreme Court, Conditional on Perceptions of Court Ideology.
For perceptions of decision making, the results comport quite well with our prior analysis. Among elites who perceive the Court as conservative, ideology has a negative and statistically significant effect—conservative legal elites are significantly less likely to perceive the Court as political in its decision making relative to moderates and liberals.Footnote 23 This effect is the exact opposite—positive and statistically significant—among those perceiving the Court as liberal, with conservative elites perceiving the Court as more political than liberals. Though the result is telling, we again raise its limitations due to the small number of legal elites who perceive the Court as liberal. For those perceiving the Court as moderate, ideology has no effect, suggesting that it behooves the Court to be perceived by legal elites as moderate.
The results for activism again comport with our main results but they tell an interesting story about ideological differences aside from raw ideological disagreement. Note that regardless of how the Court is perceived ideologically, the impact of ideology on activism perceptions is positive and statistically significant. The results show that conservatives see the Court as significantly more activist than do moderates and liberals. Thus, while there is an ideological divide in perceptions of activism, these results, in conjunction with our prior results, show that it is not ideological disagreement per se that is driving perceptions of activism. Instead, conservatives simply appear to be more predisposed to perceive the Court as activist than liberals or moderates. The size of the ideological effect diminishes as perceptions of Court ideology move from liberal to conservative, meaning that as the Court is perceived to be more conservative, conservatives begin to see the Court as less activist and the extent to which they see it as more activist than liberals diminishes.
Finally, the results for external political influence tell a story distinct from the prior two models, though once again, the results largely comport with our prior analyses. From Figure 3, it is easy to see how the effect of subjective ideological disagreement in our main results is somewhat weak. Among legal elites who perceive the Court as conservative, ideology has a predictable and statistically significant negative effect—conservative elites assess the Court as subject to a lower degree of external influence than moderates and liberals. But unlike the political decision making model, where ideology exhibited the opposite effect for liberal perceptions of the Court, ideological divisions underlying external political influence are statistically insignificant among those perceiving the Court as moderate or liberal. Once again, we qualify these results for the perception of the Court as liberal, given the small sample size.
Discussion and Conclusion
Is the U.S. Supreme Court perceived as a “political” institution? We have argued that new insights regarding this important question can be attained by analyzing the perceptions of legal elites, who are more informed about the Court than the mass public and who represent a highly relevant audience to the Court. Using as a launching pad the fact that legal elites almost universally perceive the Court as legitimate, we derived competing hypotheses grounded in two theoretical frameworks—positivity theory (Model 1) and motivated reasoning (Model 2). On the whole, we find pieces of evidence supportive of each model, depending on the facet of political perceptions. Refuting Model 1, many elites see the Court as at least somewhat political and ideological in the manner in which it renders decisions. In support of Model 1, a large majority of elites do not perceive the Court as activist or as overly influenced by external political actors.
Pitting our competing models directly against each other, the more legal elites perceive that they are in ideological disagreement with the Court, the more political they perceive the manner in which it makes decisions. These results provide strong support for the motivated reasoning logic underlying Model 2. Legal elites' perceptions of the Court are situational; they could become more political as one increasingly perceives that s/he is in ideological disagreement with the Court. In contrast, and in support of Model 1, ideological disagreement exhibits no impact on perceptions of activism, suggesting that a strong legitimacy frame leads legal elites to perceive that the Court acts within appropriate bounds. Finally, providing strong support for neither Model 1 nor Model 2, we find a moderate effect of ideological disagreement on perceptions of external political influence, suggesting at least some evidence that ideological disagreement leads people to question whether the Court maintains its judicial independence. Results are largely in line with our expectations from Hypothesis 3 about the differential impact of ideological disagreement on the three types of perceptions.
Our results uncover important insights regarding how this important group of legal elites judges the Court. To justify their negative affect derived from ideological disagreement with the Court, legal elites appear to scrutinize and highlight the political nature of the Court's decision making rather than engage in “global delegitimization” of the institution itself. This dynamic—ideological motivated reasoning constrained by the mechanisms underlying positivity theory and the realities of a legitimate Court—fits well with more general psychological perspectives on human judgment that posit a negotiation between what we wish to believe and what reality can afford: “when one wants to draw a particular conclusion, one feels obligated to construct a justification for that conclusion that would be plausible to a dispassionate observer” (Reference KundaKunda 1990: 493). We cannot believe whatever we wish to believe, but we can often bend reality in the direction of our desires.
Returning to the puzzle surrounding Hypothesis 3: How can legal elites, even when they disagree with the Court's policymaking, maintain high stores of institutional legitimacy while perceiving the Court to be political in its decision making but not activist or highly susceptible to external political influences? We think our evidence provides a plausible answer to this question. We contend that “core” perceptions of institutional legitimacy (support of the key powers and functions of the Court), which legal elites possess at near unanimous levels, and perceptions of the Court's decision making are separable constructs. Because legal elites have a vested interest in the judicial system, they cannot delegitimize the Court as an institution because doing so would indirectly delegitimize their own professional role and purpose. But motivated reasoning implies that individuals require a viable outlet for rationalizing their negative affect. Legal elites in ideological disagreement with the Court appear to rationalize this negative affect by scrutinizing and highlighting the political nature of the current Court's decision making. But they can simultaneously ascribe high legitimacy to the Court as an institution.
Furthermore, legal elites in ideological disagreement with the Court may find it difficult to credibly use perceptions of activism and external political influence as viable outlets for their negative affect. With respect to activism, perhaps they simply do not see the Court as generally operating outside of its scope or jurisdiction. They know and embrace the reality of the Court's scope and power. With respect to external political influence, elites are perhaps much more likely than the general public to know that by design, the Court is more immune from overt political influence than other institutions of American government. On the other hand, legal elites realize that the justices, perhaps because of the Court's insulated nature, have the capacity to make decisions in political and ideological ways. In sum, among legal elites, activism and external political influence are not compelling rationalizations of disliked policy outcomes, and thus, lawyers are simply more likely to use the more compelling argument regarding political decision making: “I didn't get what I want because, as everyone knows, the justices are political and ideological.” In line with motivated reasoning, elites attempt to construct an argument that supports their views and would be compelling and persuasive to a disinterested observer.
Our findings also have important implications for the political judgment literature more generally. Potent effects of political affiliation are often viewed as evidence of biased reasoning processes (e.g., Reference BartelsBartels 2002; Reference Lavine, Johnston and SteenbergenLavine, Johnston, and Steenbergen 2012; Reference Lodge and TaberLodge and Taber 2013), but others have argued that these associations are better understood as a form of low-information rationality and a resulting reliance on heuristics, or low-effort cues (e.g., Reference Lupia and McCubbinsLupia and McCubbins 1998; Reference Sniderman and StiglitzSniderman and Stiglitz 2012). Our findings offer a rare glimpse into the judgment processes of political actors with very high levels of sophistication and thus, presumably, relatively little need for heuristic substitutes. Yet we still observe forms of judgment that are consistent with motivated reasoning, suggesting the importance of bias in political perceptions.
We think that these results motivate a larger conversation about the political perceptions elites possess of federal institutions, including the Supreme Court, and how these perceptions can vary based both on contemporaneous ideological disagreement as well as on the underlying frames that influence particular perceptions of those institutions. While some political perceptions of the Court are subject to change, others are more rigid and stem from strong inclinations toward defense of the Court as an institution. Unpacking these trends and others will be valuable for studies of both elite and mass perceptions of the Court and other federal institutions.
Supporting Information
Additional Supporting Information may be found in the online version of this article at the publisher's web-site:
Appendix A. Question Wording for Survey Items: 2005 Annenberg Supreme Court Survey of Legal Elites.
Appendix B. Regression Model Results for Complementary Analysis Associated with Figure 3.
