Do Policy Messengers Matter? Majority Opinion Writers as Policy Cues in Public Agreement with Supreme Court Decisions

Abstract

Does the identity of a majority opinion writer affect the level of agreement a Supreme Court decision receives from the public? Using a survey experiment, we manipulate majority opinion authors to investigate whether individuals are willing to agree with Supreme Court opinions authored by ideologically similar justices even though the decisions cut against their self-identified ideological policy preferences. Our study provides insight into the extent to which policy cues—represented by a political institution’s policy messenger—affect agreement with a given policy. We find that a messenger effect indeed augments the level of agreement a given Supreme Court case receives.

Keywords

Supreme Court majority opinion writers policy cues heuristics experiments

Political and policy-making processes typically produce winners and losers. U.S. Supreme Court dispositions produce winning and losing outcomes for the parties in a given case, but such outcomes also draw the attention of political officials, the media, and members of the public. All of these actors may hold preferences as to a given policy outcome emanating from the nation’s High Court. Suffice it to say, people generally prefer to win, and obtaining a preferred outcome in a political or policy conflict can help shine a positive light on a process, institution, or interaction for an individual. Indeed, a robust literature has explored the theme of how we manage to maintain citizens’ trust, satisfaction, and feelings of legitimacy with regard to institutions and processes that necessarily yield political “losers” (e.g., Craig et al. 2006; Moehler and Lindberg 2009; Verba 2006). As Verba (2006, 518) put it, “in the struggle for control over personnel and outcomes, some will win and others lose. The democratic dilemma is gaining the voluntary acquiescence of the losers to the results they do not favor.”

The oft-cited phrase regarding public support and belief in the integrity and authority of the U.S. Supreme Court (as well as other institutions and processes) is “legitimacy is for losers” (e.g., Gibson 2012; Gibson, Lodge, and Woodson 2012). It rather efficiently encapsulates the dynamic popularly described by David Easton (1965) and others concerning how institutions manage to persist and function effectively despite their policy outputs regularly running contrary to public preferences. As Easton and subsequent scholars have explained, the Supreme Court has accumulated, and can draw upon, a “reservoir of goodwill” (i.e., diffuse support) whereby the public believes that even disagreeable decisions are legitimately made and should be considered authoritative. In similar fashion, the Court itself is considered to be a legitimate and upright institution that makes decisions in a fair manner and has the public’s best interest in mind in making policy. A substantial scholarly literature has been dedicated to defining, measuring, describing, and explaining public perceptions of the legitimacy of the Supreme Court (Caldeira 1986; Gibson and Caldeira 1992; Gibson, Caldeira, and Spence 2003a, 2003b; Scheb and Lyons 2000; Yates and Whitford 2002). Implicit in the idea of the Court’s reservoir of goodwill, or institutional diffuse support “bank account,” is the idea that the Court expends some of its stores in getting citizens who disagree with a decision to find such decisions (and the Court) to be legitimate and authoritatively palatable. Invariably, any salient decision will turn a portion of the public into “losers” because the Court’s policy output runs contrary to their preferred outcome. Could there be a way that a loss can be turned into a win? Put differently, is there a way that losses can be crafted and presented by the Court so that they are perceived as wins by members of the public who would ordinarily disagree with such verdicts?

As noted above, a river of literature has examined public perceptions of the legitimacy of the Court and its decisions. In conceptualizing and measuring Court public legitimacy, or diffuse support, researchers typically focus their queries on whether a citizen would be willing to support a verdict or the Court’s policy making in general, even if they disagree with a particular decision. In other words, researchers are primarily interested in whether respondents accept a verdict as legitimate and authoritative—and acquiesce to it or find it acceptable—regardless of whether they think it is good policy or correctly decided (e.g., Nicholson and Hansford 2014).

Getting to “Agreement”—Why Do People Agree with Supreme Court Outcomes?

In contrast, a significantly smaller literature has examined or attempted to explain citizens’ agreement or disagreement with the Court’s specific decisions (e.g., Adamany 1973; Baas and Thomas 1984; Mondak 1994; Zink, Spriggs, and Scott 2009).¹ Experimental studies on individuals’ agreement with Supreme Court decisions have found that respondents’ propensity to agree with a decision may be influenced by a number of factors, including the Court’s invocation of a Constitutional authority basis (Baas and Thomas 1984; Mondak 1994), exposure to criticism of the Court’s decision (Mondak 1994), as well as Court majority coalition size and whether precedent was followed or overruled (Zink, Spriggs, and Scott 2009). Hence, both factors external to the Court (i.e., third-party criticism) and factors that emanate from the Court (i.e., coalition size) can help change an individual’s perception of a Court decision from being a perceived loss (disagreement with the outcome) to a win (agreement with the outcome). Given that “legitimacy is for losers,” the fact that a change in citizens’ perceptions of loss is affected by such factors has important implications for the maintenance and development of the Court’s institutional legitimacy (e.g., Mondak 1994). This small but fruitful line of research inspires our central research question: whether individuals’ heuristic processing (use of source cues) is associated with their propensity to agree with the Court’s decisions. To assess this question, we investigate whether individuals are more likely to agree with a Court outcome when they identify ideologically with the justice who authored the Court majority’s opinion in the case.

In this study, we employ a survey experiment to assay the proposition that people will more likely agree with a policy outcome if they identify with the source cue associated with that policy pronouncement. Our study unfolds as follows: In the next two sections, we outline our theoretical argument that individuals’ perceptions of the desirability of Court verdicts (winning) are positively affected by their ideological identification with the justice who delivers the majority opinion. In the section that follows, we test this basic proposition by manipulating the majority opinion writer in two actual Supreme Court case vignettes. We also assess the impact of respondents’ knowledge of the Court on the strength of the heuristic influence on their agreement with Court outcomes. In the final section of our study, we address how our results fit into the broader picture of how the Court and its decisions are perceived by the public and how individuals are affected by cognitive biases when forming opinions based on elites’ policy edicts.

Theoretical Framework

People form opinions as the world around them frequently demands that informed choices be made. They also find themselves regularly voicing their opinions in occupational, social, and personal environments when confronted with situations—both on professional and personal issues as well as on salient social issues and government policies. Be it at work, social events, family gatherings, or even while alone watching the news, individuals are regularly apt to form and sometimes even state an opinion on a recent event (e.g., a government policy decision). In the aggregate, this makes for continuous stimuli processing and related judging. How do individuals manage this inundation of information and the attendant sense making required to form opinions and make day-to-day choices?

This task is made more manageable via a dual information processing system that helps us form preferences and make choices. Cognitive science research has advanced a theory of motivated reasoning that involves dual processing on a bicameral structure of memory. It posits a distinction between System One (S1), subconscious, processing and System Two (S2), conscious, processing of stimuli and information to make sense of the surrounding world and to react accordingly (e.g., Gibson, Lodge, and Woodson 2012; Taber and Lodge 2006). Briefly, S1 can be thought of in lay terms as our “gut response” to stimuli. It employs lightning quick impressions and associations to provide a very rough draft of reality. It is described as “fast thinking” and can be thought of as an almost instinctive, and largely unconscious, processing of information. This manner of managing information has helped us survive and persist in this sometimes-dangerous world by enabling us to make very quick decisions when immediate judgment and action are necessary. Not only does S1 work extremely quickly, but also it works rather easily, without heavily taxing our mental capacities. Thus, it is usually our first pathway to understanding our environment. Compare this with S2, which works much slower and deliberatively by considering not just the immediate, visceral information available but also less obvious concerns. S2 is based in rationality and is also more taxing on the decision maker than S1; it allows us to complete complex tasks that require analytical approaches and abstract reasoning (Kahneman 2011).

The dual processing system outlined above suggests that, given the overload of stimuli and necessary decision-making, individuals are generally inclined to rely first on S1 to help form preferences and make choices. Studies based in motivated reasoning and cultural cognition suggest that these initial impressions formed in S1 processing are shaped by preexisting beliefs and core cultural orientations (e.g., Wells et al. 2009). When S2 is then engaged, it is typically influenced by the general impressions that were developed on a given concern earlier in S1. Because S1 processing is so pervasive, we consider how it manifests itself and influences S2 processes in formulating preferences and decisions regarding the type of political output in which we are concerned—the decision to agree or disagree with a policy decision made by the Supreme Court.

In political science, scholars generally conceive of the S1 and S2 processes in slightly different terminology, but there is substantial overlap, both conceptually and in application. To provide clarification, we turn to the theoretical language of behavioral political science and endeavor to demonstrate how these dual processes inform our understanding of how citizens may come to agree with a Court decision that does not necessarily comport with their general policy preferences. S1 processing typically involves considering information heuristically rather than systematically, which is closer to S2’s deliberative and effortful processing.² Heuristics are efficiency tools that allow individuals to make quick evaluations and judgments on complex matters without having to engage in in-depth analyses (e.g., Eagly and Chaiken 1993; Kam 2005; Turner 2007). A telling example can be found in the world of advertising in the form of celebrity endorsements: a beloved and respected celebrity attaches his or her name to a product, and people evaluate and choose to buy that product, in part, based on their admiration and respect for the celebrity who has made the choice to leverage his or her reputation and image to help sell it. The individual uses a heuristic shortcut to bypass an in-depth analysis of the merits of the product by drawing a conclusion regarding its worth through the celebrity endorsement. “If this person I admire likes it, then I probably will, too”—so the thinking goes. Political behavior researchers have demonstrated that similar heuristic shortcuts—source cues—are frequently employed by individuals to gain insight and meaning regarding a wide variety of political matters that are often multifaceted and complex, such as policy outcomes and messages (e.g., Carmines and Kuklinski 1990; Lupia 1994; Zaller 1992), candidates for office (Mondak 1993), and even the neutrality and balance of news coverage (Turner 2007). Individuals can use such cues to reduce complex problems that would generally require effortful thought and deliberation to simpler, more manageable judgments that, in most cases, largely end up comporting with their own core values and beliefs (Kahan 2011; Lupia 1994; Mondak 1993). They may also feel more confident in their views by relying on such straightforward cues than if they had relied completely on the sometimes curtailed policy information that is available in many cases (Bullock 2011).

Majority Opinion Writers as Source Cues

Given that the public’s policy opinions are weakly held (Converse 1964) and are heavily influenced by elites (Zaller 1992), the source from which individuals receive their information significantly affects specific support for a given policy (e.g., Druckman 2001; Lupia 1994; Mondak 1993). Legalistic source cues from the Supreme Court as an institution (Hoekstra 1995; Mondak 1990, 1992), however, are not the only cues conveyed by the Court. A Court decision’s majority opinion author is routinely included in media reports to the public. The majority opinion writer is, in some sense, the legal community’s rough equivalent of a celebrity endorsement—which makes the authoring justice a source cue upon which individuals can rely to save time and effort when evaluating Supreme Court holdings. Of course, such processing efficiency comes with a cost: relying on a source cue alone necessarily forgoes other available—but complex—information that would otherwise give a decision maker a more complete picture upon which he or she could rely. Thus, because justices occasionally author opinions that are seemingly incongruent with their general ideology, individuals who agree with a case because they identify with a majority opinion author’s overall ideology may be unwittingly impelled to agree with policy positions that are incongruent with the policy positions they might otherwise embrace (i.e., absent the endorsement). We therefore expect to observe two relationships regarding case-level agreement among members of the public: first, an individual’s willingness to agree with a Supreme Court opinion will increase when an ideologically congruent justice authors the decision—even when its holding cuts against the individual’s policy preferences; and, second, an individual’s willingness to agree with an otherwise ideologically congruent Supreme Court opinion will decrease when an ideologically incongruent justice authors the decision.

As is the Court’s custom, the Chief Justice assigns the opinion writer to a case if the Chief is a member of the majority coalition; if not, then the senior justice within the majority is tasked with the assignment. In either scenario, the assignor can select him or herself to write the majority’s opinion. Most studies that investigate strategic opinion assignment assume that the assignor’s ideal policy position is his or her foremost priority when selecting an opinion writer (e.g., Hammond, Bonneau, and Sheehan 2005; Lax and Cameron 2007; Maltzman, Spriggs, and Wahlbeck 2000; Maltzman and Wahlbeck 1996; Rohde 1972). For the most part, these investigations consider the bargaining that takes place among the justices based on their individual policy preferences and not how the public at large will react to the resulting opinion. However, members of the Court are indeed cognizant of how the public will receive the Court’s opinions. Chief Justice Stone, for example, reassigned Smith v. Allwright’s (1944) majority opinion writer after Justice Jackson raised concerns about the fact that Stone’s original assignee, Justice Frankfurter, was a less-than-ideal policy messenger given that the case’s disposition struck down the Texas Democratic Party’s “white primary” system (Abraham 2008). In Jackson’s words, “the Court’s decision . . . will be much less apt to stir ugly reactions if the news . . . is broken to [the public] . . . by a Southerner [and] a Democrat” (Mason 1956, 614–16).

Public agreement with Court decisions facilitates Court loyalty and diffuse support (e.g., Mondak 1994), whereas repeated unpopular decisions can slowly whittle away this loyalty base (Caldeira 1986; Gibson 2007). Given the relatively fungible nature of opinion author assignment and the Court’s interest in public support and legitimacy, opinion assignment may present itself as an important tool available to the Court, which enables manipulation of the public’s agreement with the Court’s policy output—simply by changing the policy messenger.

Finally, individuals rely on source cues in varying degrees primarily because some are generally more likely to use heuristic shortcuts in a given situation or context. The conventional wisdom on political heuristics suggests that people who are unfamiliar with a complex subject matter have an incentive to rely on easily accessible heuristics rather than systematically interpreting the issue’s substance—which can be quite taxing and may still not lead to forming opinions on complex policy pronouncements, such as Court decisions, that are congruent with their personal world vision. In contrast, individuals who possess high knowledge of an institution or process may have far less need to employ heuristic shortcuts. People who are politically aware—here, those who are more well versed in the intricacies of the Court, how it works, and the judicial players—are not simply citizens who know more about the Court, they are also generally more effortful processors of judicial policy matters (e.g., Kam 2005). Accordingly, this view posits that high-knowledge individuals should be less likely to rely upon source cues (i.e., opinion author) in assaying legal outcomes. It could also be the case that one’s reliance on source cues may be conditioned on the amount of information given to that individual (see Bullock 2011). When only short descriptions of policy are given in conjunction with party cues, for example, Bullock (2011, 512) finds that cues meaningfully influence policy views compared with instances when cues are supplemented by considerable policy information.

However, a competing line of research expects the messenger effect to be heightened among high-knowledge individuals. These studies reason that knowledgeable citizens know enough to realize that they lack the requisite background needed to make informed decisions on complex policy matters, so they rely on source cues to a greater degree than less knowledgeable citizens (Clarke et al. 2013; Gigerenzer 2008; Gigerenzer, Hertwig, and Pachur 2011). Related studies show that when a heuristic “behaves badly”—in other words, a government official makes a decision that is inconsistent with that official’s typical or expected policy-making type—knowledgeable, sophisticated citizens are actually more susceptible to being misled by such elite cues, leading to suboptimal decisions that are not congruent with their preexisting policy preferences (Clarke et al. 2013; Dancey and Sheagley 2013; Kahneman 2003; Kahneman and Frederick 2002). Finally, with regard to citizens’ views of Supreme Court policy making, a rich literature suggests that to know a Supreme Court is to love it (Gibson, Caldeira, and Baird 1998; Gibson and Caldeira 2009). Accordingly, individuals with a high level of Supreme Court knowledge are likely to hold the Supreme Court, and presumably its members, in high esteem. This literature shows that individuals who are highly familiar with the Court are repeatedly exposed to symbols of legitimacy and tend to think positively toward the Court and its justices—and therefore, we believe, may be susceptible to the lead, or cue, of the elite judicial actor providing the majority opinion for the High Court in a given case.³

Experimental Design

To test whether the identification of majority opinion writers affects individuals’ agreement with Supreme Court decisions, we constructed a survey containing demographic, political, and Court knowledge questions, as well as two short vignettes of recent Supreme Court cases. We chose two recent criminal procedure cases that were decided by atypical voting blocs⁴ but were authored by ideologically consistent justices: Bullcoming v. New Mexico (2011) and Maryland v. King (2013). Justice Ginsburg authored Bullcoming, and it is generally considered to be a liberal holding.⁵ In this case, the Court ruled that forensic evidence used during trial was Constitutionally inadmissible because the lab analyst who conducted the tests did not take the witness stand and hence, violated the Confrontation Clause of the Sixth Amendment. Justice Kennedy authored King and it is considered to be a conservative holding.⁶ In this case, the Court ruled that enforcement agents’ taking and analyzing a cheek swab of an arrestee’s DNA is a legitimate police booking procedure that is Constitutionally reasonable under the Fourth Amend-ment’s Search and Seizure Clause.

The cases’ liberal and conservative policy distinctions are based on Harold Spaeth’s original Supreme Court Database coding scheme in that they are based on the winning/losing party’s status. Regarding the criminal procedure issue area, the relevant party is the criminal defendant—if he or she loses, then it is considered a conservative outcome, and if he or she wins, then it is considered a liberal outcome. Spaeth’s basic logic is largely consistent with lay interpretations of the law as well. While there might exist some doubt as to this coding convention in the case of white-collar or political crime cases, that is the not the situation for either of our vignettes.⁷

Both Bullcoming and King have drawn considerable media attention since being decided in 2011 and 2013, respectively.⁸ In fact, the New York Times ran a front-page editorial on the repercussions stemming from King’s disposition the day after the Court released its written opinion (Liptak 2013). What is more, during oral argument, Justice Alito underscored King’s importance, stating that King “is perhaps the most important criminal procedure case that this Court has heard in decades.”⁹ Our research design therefore entails both a highly salient case (King) and a less-salient case (Bullcoming) as classified by traditional measures (see Epstein and Segal 2000).¹⁰

Respondents were randomly assigned to one of two groups. Individuals in Group 1 were told the cases’ actual authors, and individuals in Group 2 were told different justices authored the case opinions. Specifically, Group 2 respondents were told Justice Thomas authored Bullcoming and that Justice Breyer authored King. Immediately following the authors’ identification, respondents were also informed whether the author was a liberal or conservative justice (see the appendix). The author identities given to Group 2 respondents, then, are inconsistent with the cases’ actual dispositions (i.e., a conservative justice, Clarence Thomas, authored the liberal holding in Bullcoming, and a liberal justice, Stephen Breyer, authored the conservative holding in King). It is important to note that Justice Thomas and Justice Breyer were selected to be the authors of the Group 2 vignettes because they were, in fact, members of the majority bloc in Bullcoming and King, respectively. Differences in agreement between Groups 1 and 2, because of proper randomization, represent the support a given case would have received had the opinion simply been assigned to another member of the majority.¹¹

While we could have constructed the vignettes around only the authoring justice’s name, the added ideological label was sensible because of the well-documented inability of many citizens to place Supreme Court justices along an ideological spectrum. What is more, the manner by which the media reports Supreme Court decisions is similar to the approach used in our vignettes,¹² so we are satisfied that the justice’s name and ideological label are reasonably representative depictions of the information available to citizens upon which they formulate their level of agreement.¹³

The surveys given to Group 1 and Group 2 were identical except for the authoring justices’ names and ideological labels and respondents were randomly assigned to these groups, so endogeneity concerns are eliminated. Furthermore, the order of the presentation in both of the vignettes was randomized to help allay any issues regarding the ordering of the case vignettes or related concerns. Given that the case disposition and author ideology are incongruent in Group 2, we are able to isolate the effect that majority opinion writers have on the level of agreement a Supreme Court decision receives compared with the effect of pure policy outcome agreement.

Results

Our survey yielded 443 adult U.S. citizen respondents from 44 states. Group 1 contained 218 individuals and Group 2 contained 225. We recruited respondents through Amazon’s Mechanical Turk platform (AMT). Recent assessment studies indicate that the AMT subject pool is a reasonable substitute for other convenience samples that have traditionally been employed in experimental settings (Huber, Hill, and Lenz 2012). As with other AMT survey experiments, our sample’s demographics are satisfactorily representative (e.g., Berinski, Huber, and Lenz 2012). Berinski, Huber, and Lenz (2012) found AMT samples to be more diverse than in-person convenience samples typically used in experiments and similar in representativeness to national Internet samples such as Knowledge Networks. They were also able to replicate well-known experiments using AMT samples (Berinski, Huber, and Lenz 2012, 362–66).¹⁴ Our survey respondents agreed with the Court’s decision in both cases more often than they disagreed.¹⁵

The hypothesized relationships between ideological congruence with a Supreme Court’s majority opinion writer and individual-level agreement are evident in the entire sample. As illustrated in Figure 1, self-identified liberals were less likely to agree with Bullcoming, a liberal case disposition, when delivered by Justice Thomas compared with Justice Ginsburg; the inverse relationship is observed among self-identified conservatives. As Figure 2 shows, self-identified conservatives were less likely to agree with King, a conservative case disposition, when delivered by Justice Breyer compared with Justice Kennedy; again, the inverse relationship is observed among self-identified liberals.¹⁶

Figure 1.

Proportion of agreement with Bullcoming v. New Mexico among the entire sample.

Figure 2.

Proportion of agreement with Maryland v. King among the entire sample.

To determine whether messenger effects similarly affected high-knowledge respondents, we divided the sample into high and low Court knowledge groups. Our Court knowledge distinction is based on five closed-ended Supreme Court knowledge questions¹⁷ asked to each respondent (see the appendix). Low-knowledge respondents are those individuals who answered less than three questions correctly.

Even among high-knowledge respondents, however, individuals were less likely to agree with an otherwise congruent Supreme Court decision when an ideologically dissimilar justice delivered the opinion compared with the same decision being delivered by an ideologically proximate justice. In Figure 3, self-identified liberals were more likely to agree with Bullcoming, a liberal decision, when authored by Justice Ginsburg compared with Justice Thomas. Likewise, in Figure 4, self-identified conservatives with a high degree of Court knowledge were more likely to agree with King, a conservative decision, when authored by Justice Kennedy compared with Justice Breyer. Both of these patterns are statistically significant and, when taken together, indicate that perceived ideological congruence plays a significant role in individual-level agreement with Supreme Court decisions.¹⁸

Figure 3.

Proportion of agreement with Bullcoming v. New Mexico among high-knowledge respondents.

Figure 4.

Proportion of agreement with Maryland v. King among high-knowledge respondents.

Experimental Implications

The results of our survey experiment support the majority opinion writer heuristic mechanism. To further investigate the degree to which author identification may affect an everyday citizen’s level of agreement with Supreme Court decisions, we extend our analysis with an ordered logit regression model, using the data collected from our survey. Our dependent variable is the response to the agreement questions that immediately followed the Bullcoming and King vignettes, which gives us six distinct categories ranging from “strongly disagree” to “strongly agree” (see the appendix). Each respondent was given both vignettes, and we have pooled the responses to the Bullcoming and King case vignettes. Postestimation tests suggest that the proportional odds assumption associated with ordered models is reasonable given our data.

Our key independent variables test whether a respondent’s ideological congruence with a case’s outcome or ideological congruence with a case’s author affects agreement with Supreme Court output. To do so, we collapse respondent, case, and author ideologies to dichotomous measures (liberal/conservative). We then use three specific indicator variables to determine whether a case outcome or author identification influence agreement. First, we consider instances when a case ideology outcome and author ideology both match a respondent’s ideology (Author/Outcome Match); second, we include a variable that indicates whether a case ideology outcome matches a respondent’s ideology, but the opinion writer’s ideology does not (Outcome ID Match Only); and third, we use a variable to designate whether the author’s ideology matches the respondent’s, but the case’s outcome does not (Author ID Match Only).¹⁹ Finally, we include a dichotomous case identifier variable (Maryland v. King vignette = 1; Bullcoming v. New Mexico vignette = 0) to control for any case-specific influences associated with the vignettes.

The results in Table 1 further demonstrate the messenger effect attendant to public agreement with Supreme Court decisions. When author ideology and case outcome both match an individual’s ideology, the probability of one’s level of agreement with a Court decision increasing is statistically significant. Notably, however, when a case outcome is incongruent with an individual’s ideology but the opinion author’s ideology matches the individual’s, the probability of one’s level of agreement increasing is also significant—albeit the coefficient is smaller compared with instances when the case outcome is also ideologically congruent. When a case outcome is congruent with an individual’s ideology, but the author’s ideology is incongruent, the coefficient is significant but, again, much smaller in magnitude compared with instances when an ideologically congruent justice delivers the majority opinion. Auxiliary analysis confirms that the messenger effect remains robust when we include race, age, gender, and education controls. The probability change values listed in Table 2 provide for more intuitive interpretations of the substantive impact of the variables on the likelihood of a respondent’s choice for a given level of agreement.²⁰ The impact of Author ID Match Only is substantively interesting, especially when considered in conjunction with instances in which both the author and case outcome match the respondent’s ideology (relative to instances in which only the case outcome matches).

Table 1.

Ordered Logit Estimates for Level of Agreement with Supreme Court Policy Output.

Variable	All	Low knowledge	High knowledge
Author ID match and outcome ID match	0.5171** (0.0076)	0.4689** (0.1019)	0.5274** (0.1722)
Author ID match only	0.3220** (0.00061)	0.4387** (0.0190)	0.2164** (0.0376)
Outcome ID match only	0.1661** (0.0012)	0.4390** (0.0918)	−0.2968 (0.2249)
Specific case control	0.3887** (0.0024)	0.6228** (0.1993)	0.0300 (0.4235)
τ1	−2.3215 (0.0208)	−2.2279 (0.0383)	−2.5118 (0.1636)
τ2	−1.0335 (0.0387)	−0.9961 (0.0139)	−1.1451 (0.2404)
τ3	−0.2402 (0.0116)	−0.0998 (0.1631)	−0.4589 (0.3454)
τ4	0.9298 (0.0215)	1.2360 (0.2044)	0.5122 (0.3492)
τ5	2.5570 (0.0345)	2.9074 (0.0832)	2.1162 (0.1993)
LnL	−1,472.96	−819.04	−643.33
n	886	504	382

Robust standard errors are in parentheses and are clustered on the treatment groups.

p < .01, one-tailed.

Table 2.

Change in Probabilities for Level of Agreement with Supreme Court Policy Output by Knowledge Level of Respondents (All, Low, and High).

Variable	Strongly disagree	Disagree	Slightly disagree	Slightly agree	Agree	Strongly agree
Outcome and author match
All	−.024	−.044	−.038	−.020	.068	.058
Low	−.020	−.035	−.039	−.019	.065	.047
High	−.029	−.053	−.033	−.015	.066	.063
Author match only
All	−.016	−.029	−.023	−.008	.044	.034
Low	−.019	−.033	−.036	−.017	.061	.044
High	−.013	−.023	−.013	−.003	.029	.023
Outcome match only
All	−.009	−.016	−.012	−.003	.023	.016
Low	−.019	−.033	−.036	−.017	.061	.044
High	—	—	—	—	—	—

Table 3 shows the logit regression analysis model results when we collapse the six categories of agreement into simple “Agree” (strongly agree, agree, and slightly agree) and “Disagree” (strongly disagree, disagree, and slightly disagree) categories. It demonstrates the robustness of our model to this alternative modeling strategy. Figure 5 displays the marginal effects of our independent variables on respondents’ decision to agree or disagree with a Supreme Court policy decision. It is evident that the majority opinion writer—as the Court’s policy messenger—has a meaningful influence on how citizens view the policy outcomes of the High Court.

Table 3.

Logit Estimates for Agreement with Supreme Court Policy Output.

Variable	All	Low knowledge	High knowledge
Author ID match and outcome ID match	0.5929** (0.0265)	0.5011** (0.0743)	0.6840** (0.1556)
Author ID match only	0.3252** (0.0001)	0.3363** (0.0125)	0.3162** (0.0468)
Outcome ID match only	0.2694** (0.0264)	0.4647** (0.0654)	−0.0680 (0.1990)
Specific case control	0.3481** (0.0641)	0.4910** (0.1593)	0.1205 (0.4476)
Constant	0.2076** (0.0641)	0.1610 (0.1120)	0.1205 (0.2977)
LnL	−561.67	−314.37	−244.72
n	886	504	382

Robust standard errors are in parentheses and are clustered on the treatment groups.

p < .01, one-tailed.

Figure 5.

Average marginal effects on individuals’ agreement with Supreme Court policy output.

As demonstrated in Tables 1 and 3, when high- and low-knowledge individuals are considered separately, the messenger effect remains significant for both groups. Among low-knowledge individuals, the Author/Outcome Match, Author ID Match Only, and Outcome ID Match Only variables are statistically significant and in the anticipated direction. The same holds true among high-knowledge respondents except that when an ideologically incongruent justice delivers an ideologically congruent case outcome, the variable (Outcome ID Match Only) estimate is not statistically different from zero. Thus, while both groups are affected by source cues from majority opinion writers, high-knowledge respondents are not more likely to agree with an ideologically congruent opinion delivered by an ideologically incongruent justice. This finding supports the notion that high-knowledge individuals are actually more susceptible to the lead of elite jurisprudential actors, arguably because they are “smart enough to know that they are not smart enough” to make a fully informed decision—hence, they sensibly rely on expert heuristic cues (Clarke et al. 2013, 230).²¹

Overall, the messenger effect is confirmed by our results—respondents are more likely to agree with a decision associated with an ideologically compatible judicial actor and less likely to agree with a decision associated with a justice who is ideologically distant—irrespective of the ideological outcome of the decision. As expected, the messenger effect is especially prominent among high-knowledge respondents. Again, respondents generally agreed with both of the Court’s decisions more often than they disagreed—which is generally consistent with extant research highlighting the Court’s “reservoir of goodwill” among the public (e.g., Caldeira and Gibson 1992; Gibson 2007; Gibson, Caldeira, and Baird 1998; Gibson, Caldeira, and Spence 2003a). However, the messenger effect wields a substantively meaningful influence on agreement at the case-specific level and, in the case of high-knowledge citizens, has an even stronger impact on an individual’s agreement with a Supreme Court policy output.

Conclusion

Our experimental design is remarkably uncomplicated, which allows for the majority opinion writer mechanism to be readily isolated. Our results present a novel perspective of the Supreme Court’s agreement-cueing capabilities—the choice of a particular justice within the majority coalition to write an opinion can modify the level of agreement that a decision would otherwise receive from a given individual. Our findings suggest that the Court’s jurisprudential outputs are not assayed by members of the public in a neutral fashion but rather are interpreted through a lens that is affected by the ideological identity of the justice delivering the judicial policy pronouncement. Given the general length and complexity of Supreme Court written opinions, it is understandable to observe individuals tempering their agreement based on the majority opinion writer when they are given only a short synopsis of the policy disposition. Practically speaking, however, short synopses are precisely the form that Supreme Court policy pronouncements take when disseminated to the public by the media. Thus, our results raise normative democratic concerns about the public’s practical ability to consume Supreme Court jurisprudence in an effortful manner (see, for example, Estlund 2007).

The extant literature on Supreme Court legitimacy suggests that when individuals agree with a legal outcome, they are unlikely to need reassurance that the decision (or the Court generally) is legitimate—after all, legitimacy is generally just for losers. However, Court gains made with one individual through messenger effects may have the opposite impact on another person who is not ideologically compatible with the majority opinion writer. Thus, the Court’s potential use of a strategy based on this phenomenon could be complex to say the least. Furthermore, as our previously mentioned example of Chief Justice Stone’s strategic choice of a more palatable policy messenger in a politically charged Texas case suggests, the Court’s use of such a stratagem might turn on a number of specific contextual concerns regarding how a legal outcome will be perceived by various populations. Still, our findings demonstrate that the Court’s choice of opinion writer on a basic ideological dimension has consequential implications for how a Court decision is received and interpreted by members of the public generally.

Our next step is to extend this research to various legal issues beyond criminal procedure. If the patterns we observed hold across issue dimensions—that is, if liberal respondents disagree with a Supreme Court decision simply because it was authored by a conservative justice, and vice versa—then opinion assignments carry with them even more weight than justices simply being concerned with ideal policy positions. Furthermore, our study has implications for policy assessment by the public more generally. Our results suggest that government officials can enhance policy agreement by the public (and presumably improved policy legitimation) by being cognizant of the phenomenon of messenger effects and strategically adjusting the manner and form of policy communication and presentation. Such adjustments may pay dividends in public perceptions of the legitimacy of government policies and the government more generally in an era of partisan polarization and public skepticism of political actors and institutions.

Footnotes

Appendix

Supreme Court knowledge questions:

1) Who is the current Chief Justice of the U.S. Supreme Court?

Anthony Kennedy; Antonin Scalia; Ruth Bader Ginsburg; John Roberts; Samuel Alito

2) Can U.S. Supreme Court justices be impeached?

Yes; No; Do not know

3) Must the U.S. Supreme Court decide every case that is appealed to it?

Yes; No; Do not know

4) Do all nine justices need to be present for the U.S. Supreme Court to hear a case?

Yes; No; Do not know

5) Of the choices given below, which justice is currently serving on the U.S. Supreme Court?

Sandra Day O’Connor; William Rehnquist; Stephen Breyer; David Souter; John Paul Stevens

Vignettes: Each respondent was given a vignette from Bullcoming and King. The order in which the vignettes were given to respondents was randomly assigned. Emphasis added to highlight treatments.

Bullcoming v. New Mexico vignette:

Maryland v. King vignette:

After each vignette, respondents were asked the following question: After reading the summary above, what is your opinion of the Supreme Court’s decision?

Strongly agree; agree; slightly agree; slightly disagree; disagree; strongly disagree

Acknowledgements

The authors would like to thank Brandon Bartels, Jeffrey Rachlinski, Charles Anthony Smith, David Law, Damon Cann, Jamie Carson, and the anonymous reviewers for their advice and constructive feedback.

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Abraham

Henry J.

2008. Justices, Presidents, and Senators. 5th ed. Lanham: Rowman & Littlefield.

Adamany

David

. 1973. “Legitimacy, Realigning Elections, and the Supreme Court.” Wisconsin Law Review 3:790–846.

Baas

Larry R.

Thomas

Dan

. 1984. “The Supreme Court and Policy Legitimation: Experimental Tests.” American Politics Research 12 (3): 335–60.

Barnes

Robert

. 2013. “With the Fourth Amendment, Things Aren’t So Predictable.” The Washington Post, April 1, A11.

Berinski

Adam J.

Huber

Gregory A.

Lenz

Gabriel S.

2012. “Evaluating Online Labor Markets for Experimental Research: Amazon.com’s Mechanical Turk.” Political Analysis 20 (3): 351–68.

Bishin

Benjamin

Hayes

Thomas

Incantalupo

Matthew

Smith

Charles Anthony

. 2013. “Testing Backlash: The Influence of Political Institutions on Public Attitudes on Gay Rights.” Paper presented at the annual meeting of the American Political Science Association, Chicago.

Blumner

Robyn

. 2013. “The Price of Punishment.” Tampa Bay Times, October 27, 5P.

Buhrmester

Michael

Kang

Tracy

Gosling

Samuel

. 2011. “Amazon’s Mechanical Turk: A New Source of Inexpensive, yet High Quality, Data?” Perspectives on Psychological Science 6 (1): 3–5.

Bullcoming v. New Mexico, 131 S. Ct. 2705 (2011).

10.

Bullock

John G.

2011. “Elite Influence on Public Opinion in an Informed Electorate.” American Political Science Review 105 (3): 496–515.

11.

Caldeira

Gregory A.

1986. “Neither the Purse Nor the Sword: Dynamics of Public Confidence in the Supreme Court.” American Political Science Review 80 (4): 1209–26.

12.

Caldeira

Gregory A.

Gibson

James L.

1992. “The Etiology of Public Support for the Supreme Court.” American Journal of Political Science 36 (3): 635-64.

13.

Carmines

Edward G.

Kuklinski

James H.

1990. “Incentives, Opportunities, and the Logic of Public Opinion in American Political Representation.” In Information and Democratic Processes, edited by Ferejohn

John A.

Kuklinski

James H.

, 240-68. Urbana: University of Illinois Press.

14.

Clarke

Harold D.

Sanders

David

Stewart

Marianne C.

Whiteley

Paul

. 2013. “Leader Heuristics, Political Knowledge and Voting in Britain’s AV Referendum.” Electoral Studies 32 (2): 224–35.

15.

Converse

Philip E.

1964. “The Nature of Belief Systems in Mass Publics.” In Ideology and Discontent, edited by Apter

David E.

, 206-61. New York: Free Press.

16.

Craig

Stephen

Martinez

Michael

Gainous

Jason

Kane

James

. 2006. “Winners, Losers, and Election Context: Voter Responses to the 2000 Presidential Election.” Political Research Quarterly 59 (4): 579–92.

17.

Dancey

Logan

Sheagley

Geoffrey

. 2013. “Heuristics Behaving Badly: Party Cues and Voter Knowledge.” American Journal of Political Science 57 (2): 313–25.

18.

Druckman

James N.

2001. “On the Limits of Framing Effects: Who Can Frame?” Journal of Politics 63 (4): 1041–66.

19.

Eagly

Alice H.

Chaiken

Shelly

. 1993. The Psychology of Attitudes. Fort Worth: Harcourt Brace Janovich College Publishers.

20.

Easton

David

. 1965. A Systems Analysis of Political Life. New York: John Wiley.

21.

Epstein

Lee

Segal

Jeffrey A.

2000. “Measuring Issue Salience.” American Journal of Political Science 44 (1): 66–83.

22.

Estlund

David M.

2007. Democratic Authority: A Philosophical Framework. Princeton: Princeton University Press.

23.

Gibson

James L.

2007. “The Legitimacy of the U.S. Supreme Court in a Polarized Polity.” Journal of Empirical Legal Studies 4 (3): 507–38.

24.

Gibson

James L.

2012. “Can the Supreme Court Survive a Health-Care Decision?” Pacific Standard, March 26. http://www.psmag.com/legal-affairs/can-the-supreme-court-survive-a-health-care-decision-40598/.

25.

Gibson

James L.

Caldeira

Gregory A.

1992. “Blacks and the United States Supreme Court: Models of Diffuse Support.” Journal of Politics 54 (4): 1120–45.

26.

Gibson

James L.

Caldeira

Gregory A.

2009. “Knowing the Supreme Court? A Reconsideration of Public Ignorance of the High Court.” Journal of Politics 71 (2): 429–41.

27.

Gibson

James L.

Caldeira

Gregory A.

Baird

Vanessa A.

1998. “On the Legitimacy of National High Courts.” American Political Science Review 92 (2): 343–58.

28.

Gibson

James L.

Caldeira

Gregory A.

Spence

Lester Kenyatta

. 2003a. “Measuring Attitudes toward the United States Supreme Court.” American Journal of Political Science 47 (2): 354–67.

29.

Gibson

James L.

Caldeira

Gregory A.

Spence

Lester Kenyatta

. 2003b. “The Supreme Court and the U.S. Presidential Election of 2000: Wounds, Self-Inflicted or Otherwise?” British Journal of Political Science 33 (4): 535–56.

30.

Gibson

James L.

Lodge

Milton

Woodson

Benjamin

. 2012. “The Symbols of Judicial Legitimacy: Thinking Fast and Slow about the U.S. Supreme Court.” Paper prepared for presentation at the annual meeting of the American Political Science Association, New Orleans.

31.

Gigerenzer

Gerd

. 2008. Rationality for Mortals: How People Cope with Uncertainty. New York: Oxford University Press.

32.

Gigerenzer

Gerd

Hertwig

Ralph

Pachur

Thorsten

, eds. 2011. Heuristics: The Foundations of Adaptive Behavior. New York: Oxford University Press.

33.

Grimmer

Justin

Messing

Solomon

Westwood

Sean

. 2012. “How Words and Money Cultivate a Personal Vote: The Effect of Legislator Credit Claiming on Constituent Credit Allocation.” American Political Science Review 106 (4): 706–19.

34.

Hammond

Thomas H.

Bonneau

Chris W.

Sheehan

Reginald S.

2005. Strategic Behavior and Policy Choice on the U.S. Supreme Court. Palo Alto: Stanford University Press.

35.

Hoekstra

Valerie J.

1995. “The Supreme Court and Opinion Change: An Experimental Study of the Court’s Ability to Change Opinion.” American Politics Research 23 (1): 109–29.

36.

Hoekstra

Valerie J.

Segal

Jeffrey A.

1996. “The Shepherding of Local Public Opinion: The Supreme Court and Lamb’s Chapel.” Journal of Politics 58 (4): 1079–102.

37.

Horton

John

Rand

David

Zeckhauser

Richard

. 2011. “The Online Laboratory: Conducting Experiments in a Real Labor Market.” Experimental Economics 14 (3): 399–425.

38.

Huber

Gregory

Hill

Seth

Lenz

Gabriel

. 2012. “Sources of Bias in Retrospective Decision Making: Experimental Evidence on Voters’ Limitations in Controlling Incumbents.” American Political Science Review 106 (4): 720–41.

39.

Kahan

Dan M.

2011. “Foreword: Neutral Principles, Motivated Cognition, and Some Problems for Constitutional Law.” Harvard Law Review 125 (1): 1–77.

40.

Kahneman

Daniel

. 2003. “A Perspective on Judgment and Choice: Mapping Bounded Rationality.” American Psychologist 58 (9): 697–720.

41.

Kahneman

Daniel

. 2011. Thinking Fast and Slow. New York: Farrar, Straus and Giroux.

42.

Kahneman

Daniel

Frederick

Shane

. 2002. “Representativeness Revisited: Attribute Substitution in Intuitive Judgment.” In Heuristics and Biases: The Psychology of Intuitive Judgment, edited by Gilovich

Thomas

Griffin

Dale

Kahneman

Daniel

, 49-81. New York: Cambridge University Press.

43.

Kam

Cindy D.

2005. “Who Toes the Party Line? Cues, Values, and Individual Differences.” Political Behavior 27 (2): 163–82.

44.

Lax

Jeffrey R.

Cameron

Charles M.

2007. “Bargaining and Opinion Assignment on the US Supreme Court.” Journal of Law, Economics, & Organization 23 (2): 276–302.

45.

Liptak

Adam

. 2011. “Supreme Court Ruling Accepts No Substitutes in Crime Lab Testimony.” The New York Times, June 24, A16.

46.

Liptak

Adam

. 2013. “Justices Allow DNA Collection after an Arrest.” The New York Times, June 4, A1.

47.

Lupia

Arthur

. 1994. “Shortcuts versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections.” American Political Science Review 88 (1): 63–76.

48.

Maltzman

Forrest

Spriggs

James F.

Wahlbeck

Paul J.

2000. Crafting Law on the Supreme Court: The Collegial Game. New York: Cambridge University Press.

49.

Maltzman

Forrest

Wahlbeck

Paul J.

1996. “Strategic Policy Considerations and Voting Fluidity on the Burger Court.” American Political Science Review 90 (3): 581–92.

50.

Maryland v. King, 133 S. Ct. 1958 (2013).

51.

Mason

Alpheus T.

1956. Harlan Fiske Stone: Pillar of the Law. New York: Viking Press.

52.

Moehler

Devra

Lindberg

Steffan

. 2009. “Narrowing the Legitimacy Gap: Turnovers as a Cause of Democratic Consolidation.” Journal of Politics 71 (4): 1448–66.

53.

Mondak

Jeffrey J.

1990. “Perceived Legitimacy of Supreme Court Decisions: Three Functions of Source Credibility.” Political Behavior 12 (4): 363–84.

54.

Mondak

Jeffrey J.

1992. “Institutional Legitimacy, Policy Legitimacy, and the Supreme Court.” American Politics Quarterly 20 (4): 457–77.

55.

Mondak

Jeffrey J.

1993. “Public Opinion and Heuristic Processing of Source Cues.” Political Behavior 15 (2): 167–192.

56.

Mondak

Jeffrey J.

1994. “Policy Legitimacy and the Supreme Court: The Sources and Contexts of Legitimation.” Political Research Quarterly 47 (3): 675–92.

57.

Nicholoson

Stephen P.

2011. “Dominating Cures and the Limits of Elite Influence.” Journal of Politics 73 (4): 1165–77.

58.

Nicholson

Stephen P.

Hansford

Thomas G.

2014. “Partisans in Robes: Party Cues and Public Acceptance of Supreme Court Decisions.” American Journal of Political Science 58 (3): 620–36.

59.

Petty

Richard E.

Cacioppo

John T.

1986. Attitudes and Persuasion: Classic and Contemporary Approaches. Boulder: Westview Press.

60.

Rohde

David W.

1972. “Policy Goals, Strategic Choices, and Majority Opinion Assignment, 1946-1962.” Western Political Quarterly 16: 652–82.

61.

Ryan

Timothy

. 2012. “What Makes Us Click? Demonstrating Incentives for Angry Discourse with Digital-Age Field Experiments.” Journal of Politics 74 (4): 1138–52.

62.

Scheb

John M.

II Lyons

William

. 2000. “The Myth of Legality and Public Evaluation of the Supreme Court.” Social Science Quarterly 81 (4): 928–40.

63.

Smith v. Allwright, 321 U.S. 657 (1944).

64.

Taber

Charles S.

Lodge

Milton

. 2006. “Motivated Skepticism in the Evaluation of Political Beliefs.” American Journal of Political Science 50 (3): 755–69.

65.

Turner

Joel

. 2007. “The Messenger Overwhelming the Message: Ideological Cues and Perceptions of Bias in Television News.” Political Behavior 29 (4): 441–64.

66.

Verba

Sidney

. 2006. “Fairness, Equality and Diversity: Three Big Words.” Social Research 73 (2): 499–540.

67.

Wells

Chris

Reedy

Justin

Gastil

John

Lee

Carolyn

. 2009. “Information Distortion and Voting Choices: The Origins and Effects of Factual Beliefs in Initiative Elections.” Political Psychology 30 (6): 953–69.

68.

Yates

Jeff

Whitford

Andrew B.

2002. “The Presidency and the Supreme Court after Bush v. Gore: Implications for Legitimacy and Effectiveness.” Stanford Law & Policy Review 13 (1): 101–18.

69.

Zaller

John

. 1992. The Nature and Origins of Mass Opinion. New York: Cambridge University Press.

70.

Zink

James R.

Spriggs

James F.

Scott

John T.

2009. “Courting the Public: The Influence of Decision Attributes on Individuals’ Views of Court Opinions.” Journal of Politics 71 (3): 909–25.