Case Salience and Media Coverage of Supreme Court Decisions

Abstract

Judicial behavior is contingent on case salience. Unfortunately, existing measures of case salience have met with some skepticism. After discussing the characteristics of an ideal measure of salience, the authors construct a new measure of case salience. This new measure expands on prior studies by examining coverage in four diverse newspapers and includes coverage anywhere in the paper, instead of concentrating on front-page coverage only. By developing this new measure, the authors uncover patterns about national media coverage of the Court and provide a potentially more useful measure of case salience.

Keywords

judicial decision-making salience media Supreme Court

On June 10, 1996, the United States Supreme Court upheld the convictions of Michael Whren and James Brown. While the defendants were initially pulled over for minor traffic infractions, they were convicted of serious drug offenses.¹ The young defendants were driving a new car in a known drug area and were African American. The defense in Whren v. United States contended that the investigating officers needed probable cause of drug activity for the stop. To rule otherwise, the defense argued, could justify racially discriminatory behavior (Sklansky 1997, 289). The Supreme Court’s decision, stating that the judiciary should not look at the officer’s subjective state of mind, was quickly heralded by law enforcement (Biskupic 1996), but criticized by many who suggested the opinion made all drivers “become prey to police officers’ arbitrary whims, hunches, suspicions, and prejudices” (Levit 1996, 187).

On the same day as the Whren decision, New York City’s mayor Rudi Giuliani agreed to a new budget; the city of Newark, New Jersey, implemented new policies to improve some of its failing schools; and presidential candidate Bob Dole announced changes to the Republican Party’s platform concerning abortion. We know these events occurred because they were reported on the front page of the New York Times on June 11, 1996. Absent, however, from the Times front-page coverage is any mention of the Whren case. Although contemporary critics called this case a “clear step . . . toward authoritarianism, toward racist policing, and toward a view of minorities as criminals” (Harris 1997, 547), the Times placed a story about Whren on page 22. On the same day, however, the editors of Washington Post, Chicago Tribune, and Los Angeles Times all devoted space on the front page of their papers to the Whren holding.

There could have been many reasons why the editors of the New York Times did not place the Whren decision on page 1. Perhaps they felt the case was not that important. Perhaps they felt that local issues, such as the Newark schools situation and the city’s budget, were more interesting to their subscribers. Perhaps the conservative, pro–law enforcement decision did not coincide with the ideology of some of the editors, and these editors wanted to downplay the decision’s relevance. Whatever their motivations, the result is that a Times reader may have a very different impression of the importance of the Whren decision, covered on page 22, than a reader of the many other papers across the nation that gave the decision front-page attention.

This example is not an indictment on the coverage of the Times, as reporters and editors must weigh a number of different characteristics of newsworthiness when deciding what to cover and where to place certain stories (Gans 1979). The problem arises when scholars use news coverage in the New York Times as a measure of a case’s importance, relevance, or “salience.” Using front-page coverage of the Times as the only indicator of salience could be problematic as many cases, just like Whren, may not appear on the Times’ front page but may be considered salient based on other media outlets.

In this article, we further explore the idea of case salience, examine multiple newspapers’ coverage of Supreme Court decisions, and propose a new measure of case salience. This article speaks to multiple audiences and makes a significant contribution to the discipline. For judicial scholars, this article proposes an alternative and potentially better salience measure. For researchers examining political behavior outside of the judicial context, this article examines whether the practice of coding coverage from one media source may be an appropriate way to measure the salience of specific topics. Lastly, for scholars of media politics, this case provides an excellent opportunity to inform debates over standards of newsworthiness.

Salience and the Courts

Most scholars agree that a salient issue or event is one that actors care more about (Niemi and Bartels 1985, 1212) or one that they feel is more “important” (Baird 2004, 760). Others define salience as awareness of information about events or issues (Hutchings 2001). This definition is often related to the amount of media coverage devoted to an issue (Canes-Wrone and de Marchi 2002). These two aspects are invariably intertwined as “people generally have only a few issues that are particularly important to them and to which they pay attention” (Edwards, Mitchell, and Welch 1995, 110-11).

Whichever definition we choose, issues or events deemed highly salient are more likely to influence decisions of individual citizens and elite actors. For example, the presence of highly salient issues on the ballot can increase voter turnout (Smith 2001), and citizens often choose candidates based on salient issues (Abramowitz 1995). Issue salience also has an important influence on legislative behavior (Hutchings 1998), and Ang and Peksen (2007) find that economic sanctions are much more likely to be effective if the nation enforcing the sanction sees the issue as important.

In the judicial context, regardless of whether we call them “landmark” decisions (Segal and Spaeth 1996, 976), “controversial” cases (Pritchett 1941, 890), or simply “big” cases (Barrow and Zuk 1990, 466), prior research suggests that justices act differently when deciding salient cases. Judicial scholars have identified two types of salience: political salience and legal salience. A case that is politically salient is one that is important to the public. A case that is legally salient, however, is important because it influences “the development of the law” regardless of whether it is known to the public (Bailey, Kamoie, and Maltzman 2005, 79). Like most recent studies, we focus on political salience as opposed to legal salience. Although legal salience can be important, the theoretical framework underlying the two most popular paradigms for judicial behavior, the attitudinal and the strategic models of judicial decision making, both point to the importance of political salience, rather than legal salience.

The attitudinal model posits that justices have policy preferences and suggests that their behaviors on the bench are influenced by those preferences (Segal and Speath 2002). Thus, a liberal justice will support liberal outcomes in most cases and a conservative justice will generally support conservative outcomes. While this model of judicial behavior has received substantial support in the literature, the influence of the attitudinal model may be conditioned by the salience of the case (Spaeth and Segal 1999, 309). A justice may be more willing to follow the majority than dissent if the issue at hand is relatively unimportant or is perceived to lack wide-ranging ramifications (Spriggs, Maltzman, and Wahlbeck 1999, 491).

In their analysis of civil rights cases from the Supreme Court, Unah and Hancock (2006) find that justices are more likely to follow their policy preferences in highly salient cases than in those cases deemed less significant. Like other political actors, judges may devote their attention to issues they feel are the most important and that may have larger policy impact. Similarly, in their examination of the influence of attorney experience and Supreme Court decisions, McAtee and McGuire (2007) find that attorney experience influences a justice’s behavior only in nonsalience cases, where perhaps the justice is not as knowledgeable of or interested in the issues at hand and is more easily swayed by attorney arguments.

The strategic model of judicial decision making suggests an alternative way that salience may influence judicial behavior. This model recognizes that justices do not make decisions in a vacuum; they are constrained by the preferences of other actors (Epstein and Knight 1998, 10). Aside from conditioning the effect of personal preferences, case salience can factor into individual justice’s calculations of how he or she expects other political actors or the public to react. For example, justices may negotiate more in salient decision than in more ordinary cases (Epstein and Knight 1998, 98). Spriggs, Maltzman, and Wahlbeck (1999) find that justices make more suggestions and threats during the opinion-writing process and author more separate opinions in those cases that are deemed salient. Others suggest that the Chief Justice is more likely to write the opinion in highly salient cases, presumably to have more control over the court-created policy in major decisions (Brenner 1993; Epstein and Segal 2000).

Operationalizing Salience in the Courts

While case salience may play a role in attitudinal and strategic models, there is considerable disagreement over the operationalization of the concept. Supreme Court justices are generally reticent to discuss the salience of current or pending cases (Epstein and Segal 2000, 67); therefore, researchers have substituted proxy measures to capture judicial salience. In their study of case salience, Epstein and Segal (2000, 69-70) cite seven prior measures of case salience used in previous research: (1) cases that have been reprinted in constitutional law textbooks, (2) cases that are included in Congressional Quarterly’s list, (3) cases on The Supreme Court Compendium’s list, (4) cases that generate “substantial” Supreme Court citations within five years of their decision date, (5) cases that generate eight or more law review articles within two years of their decision date, (6) cases headlined on the advance sheets of the Lawyer’s Edition, and (7) cases generating substantial amicus curiae participation. Epstein and Segal (2000) convincingly cite the flaws of each approach and, consequently, derive their own measure—whether the case is discussed on the front page of the New York Times.

Epstein and Segal (2000) make compelling arguments for their measure and it has been widely accepted and used in subsequent studies (e.g., Bailey, Kamoie, and Maltzman 2005; Baird 2004; Giles, Blackstone, and Vining 2008; McAtee and McGuire 2007; Unah and Hancock 2006). Despite its common usage, however, there are at least three potential systematic flaws in using the New York Times (or any other individual media outlet) as a proxy for salience (Brenner and Arrington 2002). First, some papers may lean to the ideological left or right. In the New York Times example, there is some evidence that the Times leans farther left than many other papers (Groseclose and Milyo 2005). If the Times exhibits a liberal bias, it may be more likely to cover liberal than conservative decisions, and thus, relying on the Times measure may overrepresent the salience of liberal cases (Maltzman and Wahlbeck 2003). Unah and Hancock (2006) find merit in this critique, finding that 28 percent of liberal Supreme Court decisions are covered on the front page of the Times, compared to 19 percent of conservative outcomes (χ²= 9.008, p < .01).

Second, Brenner and Arrington (2002) argue that cases must compete with other news stories (and other cases) for front-page coverage. For example, if a natural disaster occurs on the same day the Supreme Court announces an important decision, the disastrous event may push an otherwise salient case off the front page and to other parts of the paper. The case is still important, yet it would not appear salient under the measure. Similarly, two important cases may compete for front-page coverage, but due to space limitations, only one may make it to page 1. Obviously, this problem is exacerbated with any media measure that relies solely on the front page, rather than considering coverage anywhere in the paper.

A third criticism levied at coverage from a single source is that newspapers may be more likely to report stories from their home area, as their readership and advertisers are likely to live and work close by (Brenner and Arrington 2002; Maltzman and Wahlbeck 2003). For the New York Times, Unah and Hancock (2006) verify this critique, concluding that 42 percent of cases originating from New York State are published on the front page of the Times, compared to 23 percent of cases originating elsewhere (X²= 5.65, p < .05).

The Whren case and associated coverage provides one concrete example of a case illustrating many of these critiques. Whren was a conservative decision that strongly supported (if not greatly expanded) law enforcement’s ability to use evidence obtained during minor traffic infractions. The same time as the Court decided Whren, two local stories appeared to be very prominent in the Times, as New York City’s mayor and city council were gripped in difficult budget negotiations and New Jersey schools struggled with multiple policy and monetary issues. These two local issues and the fact that it was a conservative decision may have been responsible for bumping coverage of Whren to page 22.

Despite the concerns expressed in the judicial literature, many have continued to use the front page of the Times as a proxy for salience, concluding that “its advantages over other measures lie in its immediacy, which makes it facially the most valid measure, and in its content neutrality, which makes it transportable to virtually any policy area of interest” (Unah and Hancock 2006, 304). We agree that the New York Times measure has many advantages over currently extant measures, but we believe that building off of the logic of this measure and expanding its reach may yield a better measure. To improve upon this existing measure of salience, we next discuss the characteristics of an ideal measure of case salience.

Defining an Ideal Measure of Salience

We believe that there are six characteristics of an ideal measure of salience. First, if one is examining the influence of salience on behavior, it follows that the measure of salience should be contemporaneous with those behaviors (Baird 2004, 760). After all, justices are not influenced by history’s assessment of a case’s significance but the case’s apparent importance at the time it is being decided (Cook 1993, 1128; Epstein and Segal 2000, 67-8). Media measures score well on this criterion, but the need to have contemporary measures of salience excludes many previous measures that are retrospective in nature, such as The Supreme Court Compendium, Congressional Quarterly lists, and cases printed in law textbooks (Epstein and Segal 2000, 68).

Second, a good measure of salience should be replicable. The reliability of any measure may be questionable if future studies and/or other researchers cannot replicate that measure (King 2007). Again—media measures score quite well on this criterion. Most researchers have access to major newspapers, whereas other options, like scoring cases from the lawyer’s edition, are more difficult for many scholars.

Third, a good measure should be transportable, meaning that it can be used across multiple types of studies regardless of the subfield (Epstein and Segal 2000, 67). Having a measure of salience that can be used to study all three branches of government would allow more insight into the role of salience in political behavior than a measure that is specific to the courts. Using the media as a measure of salience again appears appropriate in this context, as nonjudicial studies have used the media as a proxy for salience (Canes-Wrone and de Marchi 2002). Those measures specific to judicial studies, such as the number of amicus curiae briefs, would not be transportable to other fields.²

Fourth, a good measure should reflect the potential range of salience. It is unrealistic to claim that a case or issue either is or is not salient. Just as partisanship varies from strong partisan to weak partisan to undecided, so too do assessments of salience and ambivalence vary across a spectrum of beliefs. In a similar example, scholars in the judicial field have moved away from using dichotomous measures of judicial ideology (e.g., partisanship) to more refined and continuous measures, such as judicial common space scores (Epstein et al. 2007) or measures derived from NOMINATE scores (Giles, Hettinger, and Peppers 2001). Just as with judicial ideology, a more nuanced measure of salience that includes more potential variation will open the door for new examinations of salience and its importance.

Fifth, current theories of judicial decision making suggest that salience is important to the Court because it reflects the national mood. Justices must “rationally anticipate whether [their] preferred outcomes will be respected and faithfully followed by relevant publics” (McGuire and Stimson 2004, 1019). As they lack the institutional ability to enforce their own rulings, strategic models predict that justices should display different behaviors because they anticipate that the public feels strongly about an issue. If the High Court issues a decision contrary to public opinion, the Court may lose legitimacy in the public’s eye. In response, elected officials, fearing negative electoral consequences, may also reject the Court’s ruling. Some prior studies support the view that justices’ behaviors change with public opinion (e.g., McGuire and Stimson 2004; but see Norpoth and Segal 1994). However, the influence of public opinion on judicial behavior may be conditioned on the salience of the case. For example, Giles, Blackstone, and Vining (2008) find that public opinion influences judicial decisions in nonsalient cases but not in salient cases. Given this important line of research on judicial responsiveness to public mood (either directly or indirectly through elected officials’ attentiveness to public opinion), a good measure of salience should reflect those cases thought important by the public.³

Finally, an appropriate variable for any study should be free from systematic bias. For example, if a measure of salience consistently favors liberal outcomes, then it introduces measurement error into studies employing that measure. This type of error, although unintentional, might cause us to draw faulty conclusions about the effects of salience.

Sample Selection and Data Collection

To evaluate media sources as a measure of salience, we examine coverage of cases argued during William Rehnquist’s tenure as Chief Justice (1986-2005). Once we selected this era, we then had to determine which media outlets to search. We immediately concluded that we must search the New York Times. The importance of the Times is obvious—it is the preeminent indicator of salience (Epstein and Segal 2000) as well as perhaps the most decorated and influential paper in the country. We then considered searching the Vanderbilt Television Archives for indications of coverage on television, but we quickly discovered that far too few cases are covered on the major newscasts to yield accurate results. We also considered searching popular Internet sites focusing on the courts, but then determined that these sites would not be relevant for older cases. After considering these options, we concluded that Epstein and Segal (2000) are correct that newspaper coverage is the ideal media indicator. Because of the problems with using any single media outlet, however, we wanted to combine the Times with three other papers that were respected as excellent news outlets while simultaneously (at least in the aggregate) maintaining ideological and regional balance. We eventually settled on the Washington Post, Los Angeles Times, and Chicago Tribune, in addition to the New York Times. These papers have been previously used in the judicial literature because they are both geographically and ideologically diverse. The New York Times and the Washington Post tend to take more liberal stances while the L.A. Times and the Chicago Tribune are generally considered more conservative (Segal and Cover 1989, 559).

After selecting the papers, we then searched all cases orally argued during this period to determine whether they were covered in the New York Times, Washington Post, Los Angeles Times, and/or Chicago Tribune, yielding a total of 7,356 searches.⁴ If the case was covered, we coded the page and section number on which it appeared. We then merged the coverage data with the cases from the U.S. Supreme Court Database (Spaeth 2008) data set to better understand the characteristics of the cases that are covered.

Comparing Court Coverage in Major Newspapers

Scholars have noted the variations among media coverage of political events and political actors. For example, Canes-Wrone and de Marchi (2002) found that coverage of roll call votes differed between even “national” papers like the New York Times, the L.A. Times, and the Washington Post. To help answer whether coverage differs by paper, Figure 1 presents the percentage of Supreme Court cases covered on the front page of four major newspapers, the New York Times, Los Angeles Times, Washington Post, and Chicago Tribune. Figure 2 shows a similar graph, but is adjusted to reflect coverage anywhere in the paper (not just on the front page).

Figure 1.

Percentage of Supreme Court cases covered on the front page of four major newspapers, 1986-2004

Figure 2.

Percentage of Supreme Court cases covered anywhere in four major newspapers, 1986-2004

Taken together, these figures tell a number of important stories. First, coverage varies across time. Both figures suggest that coverage is much more prevalent some years than others. For example, Figure 1 reveals that in 2000 almost 30 percent of cases were covered on the front page of the New York Times and the L.A. Times, while page-1 coverage in the New York Times just a few years before did not top 10 percent. This suggests that studies of the media and the Supreme Court should closely consider context and not assume that trends found in one year continue in other years. This trend of temporal variation is even more obvious in Figure 2, where overall coverage is declining over time, regardless of the paper in question. Second, these figures demonstrate that the coverage differs radically across papers. In most years, the New York Times covers more Supreme Court cases—both on the front page and anywhere in the paper—than the Washington Post, Chicago Tribune, or L.A. Times. The Washington Post and L.A. Times have similar coverage most years and the Chicago Tribune consistently devotes the smallest amount of ink to Supreme Court cases. While this may be an indication of more thorough coverage by the New York Times, it may also provide evidence that using the Times measure overrepresents the number of salient cases, as others have suggested (Brenner and Arrington 2002, 104).

Toward a New Salience Measure

We contend that a quality salience measure must be contemporaneous, replicable, transportable, nondichotomous, a measure of mass national mood, and free from systematic biases. We also believe that existing proxies of salience all fail on at least one of these attributes. Even the front page of the New York Times, while probably the best measure currently available, is suboptimal on many of these criteria (Brenner and Arrington 2002; Unah and Hancock 2006).

To alleviate these problems, we developed a Case Salience Index (CSI) that includes components of all of the media measures previously discussed. We begin with the presumption that combining four papers from four different regions will reduce (although not necessarily eliminate) regional biases. Furthermore, we believe it is important not only to recognize the significance of the front page, but also to include cases mentioned anywhere in the paper. We think this is particularly noteworthy, as less than one-fourth of cases argued before the Court are covered on the front page in any given year, but many receive some coverage off page 1. We contend that coverage anywhere indicates some level of salience, although clearly front-page stories are more important.

Given these criteria, we develop an index where each occurrence of coverage on the front page of any paper receives a score of 2, coverage anywhere else in each paper (but not on page 1) receives a score of 1, and no coverage at all receives a 0. We then sum these scores across all papers, meaning that each case receives a CSI score varying between 0 (meaning that it was not covered anywhere in any paper) and 8 (indicating that it was covered on the front page of all four papers).⁵ Our index includes all the advantages of using the front page of the Times while minimizing the potential problems of using a single media outlet.

Given what we know theoretically about salience, a good sign of face validity would be if the index places the plurality of cases in the 0 or “not salient” category and the percentage of cases would decrease with each step on the scale. Using this measure of salience, we would expect that few cases would receive an 8—the highest potential point on the case salience index.

Figure 3 presents a histogram showing the distribution of the Case Salience Index. Fortunately, it largely conforms to our expectations previously outlined. Approximately one-fourth of cases receive a score of 0, and three-fourths of the cases score a 4 or lower. Further investigation suggests that the Case Salience Index identifies many potentially important cases that are counted as not salient if we examine just one particular newspaper source. For example, we find that 1,494 out of 1,876 cases (80 percent) were not covered on the front page of the New York Times. Out of these cases that would be deemed “not salient” by the Times measure, over 1,000 were covered somewhere in at least one paper. Over 18 percent (270) of the cases deemed not salient by the New York Times measure actually appeared on the front page of at least one other paper, with 71 cases appearing on the front page of at least two papers, but not the Times.

Figure 3.

Distribution of the Case Salience Index

Examples of cases appearing on multiple front pages but evading page-1 coverage of the New York Times are numerous. In addition to Whren, cases involving election laws,⁶ cable television regulation,⁷ governmental immunity,⁸ contempt orders in child custody hearings,⁹ and racial discrimination¹⁰ all appeared on the front page of at least two other major papers but did not appear on page 1 of the Times. In addition, many cases from the New York area appeared on the front page of the Times but were not significantly covered in any other paper, front page or otherwise. For example, one case dealing with New York cigarette taxing policies¹¹ appeared on page 1 of the Times but was not covered anywhere in the three other papers. In all, 122 cases received front-page coverage in the New York Times yet did not appear on the front page of any other paper. Thus, this measure has the potential of overrepresenting some cases’ salience while underrepresenting the importance of other cases.

To further illustrate the potential differences, Figure 4 displays histograms of the CSI when cases are reported on the New York Times’ front page and when they are not. The left panel of the figure varies from 2 to 8 because 2 is the lowest possible score for a case that has been covered on the front page of the Times. The most important lesson from the left panel of Figure 4 is that three-fourths of the cases that are covered on the front page of the New York Times are not covered on the front page of at least one other paper. This is further indication that simply using the New York Times measure produces results that are very different than the results you receive when considering four papers.

Figure 4.

Distribution of the Case Salience Index (CSI)

The right-hand panel of Figure 4 presents the distribution on the CSI for cases that are not covered on the front page of the New York Times. Here, a score of 8 is impossible because none of these cases were covered on the front page of the Times. Once again, this figure suggests that consulting only the New York Times gives a very different view of salience than consulting all four papers. Approximately two thirds of the time cases are covered somewhere in at least one paper, even though the case does not appear on page 1 of the Times.

To further evaluate the efficacy of this multi-paper measure, we estimated an ordinal logistic regression model where the unit of analysis is the case and the dependent variable represents a particular case’s score on the Case Salience Index. Following Maltzman and Wahlbeck (2003), we control for a number of other factors that may influence newsworthiness. For example, any follower of the Supreme Court would expect that a case on flag burning will be more salient to the public than a case debating the intricacies of administrative law. If the new measure is truly representative, however, other more idiosyncratic factors should exert no influence on the likelihood of coverage and thus no influence on the likelihood of being considered a salient case.

Our model includes independent variables representing ten of the traditional issue areas used in the U.S. Supreme Court Database (Spaeth 2008),¹² with the expectation that traditionally newsworthy issue areas like civil rights and First Amendment will receive higher salience index values than other issues with less public resonance, such as cases dealing with economic issues. Given the journalistic norms to cover unusual political behavior (Groeling 2010), we include a variable representing whether the opinion writer takes a stance that is different than the writer’s ideology,¹³ whether the case overrules precedent, whether the case rules a law unconstitutional, and the size of the majority coalition (with the expectation that closer decisions will be more likely to be covered).

We also control for whether the case is decided in a liberal direction.¹⁴ As prior studies show that particular media sources, including the New York Times, may have an ideological bias in the stories they cover (Groseclose and Milyo 2005; Unah and Hancock 2006), we need to determine whether this new measure shares such biases. If this measure also shows an ideological bias, its usefulness as an objective measure of salience would be compromised. After all, few would argue that, ceteris paribus, cases with liberal outcomes are more salient than cases with conservative outcomes, or ones from other parts of the country. We also include a control for whether the case occurred in one of the typical coverage areas for our four papers, including a dummy variable indicating whether the case originated in one of the papers’ home states.

The results in Table 1 indicate that certain types of cases are more likely to be covered (and thus to be considered salient) than others. For example, First Amendment, civil rights, and privacy issues are more likely to be covered than economic cases (the base term). Similarly, the size of the majority coalition is negatively associated with the CSI.¹⁵ Cases dealing with constitutional issues are more likely to be covered, although cases overruling precedent are not statistically significant. We also see that a case originating in one of the papers’ home states is more likely to be covered as well. However, the influence of the home-state variable is relatively small, as the average change in predicted probability is less than 3 percent when moving from a home-state case to a case from another state.

Table 1.

Checking for Bias in the Case Salience Index

	Logistic regression coefficient	Robust SE
Liberal outcome	0.06	.05
Writer takes alternative stance	−0.11	.09
Number in the majority coalition	−0.26	.02**
Overrules precedent	0.27	.17
Rules unconstitutional	0.59	.18**
First Amendment issue	1.51	.20**
Civil rights issue	0.85	.16**
Criminal procedure issue	0.33	.15*
Due process issue	0.55	.25*
Privacy issue	2.21	.36*
Attorney issue	−0.80	.35*
Union issue	−0.00	.22
Judicial power issue	−0.56	.19**
Federalism issue	0.55	.21**
Tax issue	−0.58	.25*
Interstate relations issue	−0.61	.66
Miscellaneous (uncategorized) issue	1.30	.68
Case from covered area	0.54	.09**
N	1,876
Chi-square	4083.89**
Pseudo R²	.06
Reduction in error	4.0%

Note: Robust standard errors are clustered on the term. All case issue areas are listed except “economic activity,” which is the base term.

p < .05. **p < .01.

Importantly for our purposes, we see that the liberal outcome is not statistically significant, indicating that the CSI does not overrepresent liberal (or conservative) outcomes. In sum, the criteria we established earlier in this article combined with the results in Table 1 suggest that the CSI closely follows those factors thought to be traditional indicators of important or salient decisions while avoiding the biases suggested concerning other salience proxies.

We next seek to determine whether our measure of issue salience influences a substantive area in the way in which we would expect. Following Epstein and Segal (2000), we expect that Chief Justices are more likely to write the opinion on salient cases. Indeed, this is what Epstein and Segal find, although the size of the effect they find is smaller than that found by using other competing measures like cases on Congressional Quarterly’s list. To test this possibility with our data, we determined the percentage of cases on which the Chief Justice self-assigns opinion-writing duties for each level of the Case Salience Index and plotted the results in Figure 5 along with a line representing results using the New York Times measure. Figure 5 suggests that the Chief Justice is more likely to write the majority opinion as the salience of the case increases. In the line representing the New York Times’ measure, we see that the Chief Justice writes the majority opinion in about 10 percent of the nonsalience cases and about 17 percent of the salience cases. However, rather than an “all or nothing” approach that we find with this dichotomous variable, our approach allows us to see that the effect here is not linear. In the second line using the Case Salience Index as our measure of salience, we see the jump is the greatest near the low end and at the highest end of the scale—as it moves from 1 to 2 and from 6 to 7 on the scale. Furthermore, the magnitude of the effect is plausible and fits with what we know about judicial behavior more generally.

Figure 5.

Chief Justice opinion self-assignments and salient cases

To determine whether the effect we find previously is merely a result of other factors, we next model whether or not Chief Justice Rehnquist self-assigns the opinion-writing duties as a function of salience as well as thirteen other factors. As Table 2 displays, even after controlling for many of the typical signs of a significant or important case, such as overturning precedent, issue areas, and the size of the majority, we see that Chief Justice Rehnquist was more likely to assign the opinion-writing duties to himself in salient cases using the index. Our analysis of the predicted probabilities displays that this effect is substantively large as well—the probability of Rehnquist writing the opinion doubles from the low end of the high end of the salience scale.¹⁶

Table 2.

Chief Justice Self-Assignments

	Logistic regression coefficient	Robust SE
Case salience	0.09	0.03**
Number in the majority coalition	0.02	0.04
Overrules precedent	0.66	0.31*
Rules unconstitutional	−0.84	0.40*
First Amendment issue	1.53	0.33**
Civil rights issue	1.13	0.31**
Criminal procedure issue	1.83	0.25**
Due process issue	1.84	0.39*
Privacy issue	2.27	0.49**
Attorney issue	0.85	0.81
Judicial power issue	1.03	0.31**
Federalism issue	1.26	0.29**
Interstate relations issue	0.80	1.18
Miscellaneous (uncategorized) issue	1.07	1.02
Constant	−3.68	0.36**
N	1,769
Chi-square	183.94**
Pseudo R²	.06

Note: Robust standard errors are clustered on the term. All case issue areas are listed except “unions” and “tax,” which were dropped due to collinearity. “Economic activity” serves as the base term for the issue areas.

p < .05. **p < .01.

Discussion and Conclusion

Patterns of media coverage of the Supreme Court are important for two reasons. First, the media are an important means through which citizens learn about the Court (Caldeira 1986, 1216). If there are biases in the types of cases that are covered, then, presumably, citizens’ knowledge of the job of the judiciary will be similarly skewed. In addition, media coverage provides a contemporaneous and easily accessible measure of case salience—a concept that has been shown to be an important influence on judicial decision making.

This investigation clearly shows that media coverage of the courts varies across outlets. Almost 75 percent of Supreme Court cases are covered by one of four papers, but the coverage by paper differs from a low of 34 percent of cases in the Chicago Tribune to a high of 63 percent of cases in the New York Times. If newspapers provide political information, then given these different patterns of coverage, we would expect New York Times readers to have more basic knowledge of the courts than readers of the Chicago Tribune. This could even contribute to regional differences in knowledge and public opinion of the High Court. We also find that media coverage of Supreme Court cases varies considerably over time. In some years, almost 85 percent of cases receive coverage in some papers while in other years the number hovers around 65 percent. Future research should explore the causes and effects of these temporal differences.

In addition to adding to our substantive knowledge of media coverage of the Supreme Court, this article adds to our knowledge of case salience. There are myriad ways to measure case salience, with media sources serving as perhaps the most popular. Not only is salience useful in Supreme Court studies, but research on other courts is also examining the importance of case salience using similar techniques. For example, Vining and Wilhelm (2010) have begun using coverage of state court decisions in the state’s most circulated newspaper as a measure of state case salience. It appears clear that media coverage will continue to serve as an important proxy of case salience.

In this article, we build on the use of the front-page New York Times salience measure and improve on that commonly used proxy. First, the new index allows for more variation in the salience measure, which is consistent with our theoretical and empirical expectations for issue salience. Second, the new measure includes coverage anywhere in the paper, as reporting anywhere notes that a case is at least somewhat salient and allows for a more nuanced measure of case salience. This also corrects for the “big-news-day bias” that could lead to the improper categorization of an otherwise salient case (Brenner and Arrington 2002, 103-04). Thus, an important case will still have some level of salience in our measure even if it is bumped from the front page by catastrophic events or even another highly salient Supreme Court ruling.

Third, examining multiple papers with varying ideological perspectives (Segal and Cover 1989) from multiple regions of the country provides face validity of less potential bias. This new measure, which we call the Case Salience Index, combines coverage in four papers, includes coverage anywhere in a paper, but weights prominent, front-page coverage higher. As noted earlier, we do see differences between what is covered in the Times and what is covered in other papers, collectively.

Of course this measure is not perfect. It does not consider coverage on television—which would clearly mark a salient case. Given that cases from the four papers’ home states are more likely to be covered, we improve upon but do not solve the localism problem. Furthermore, our investigation also rests on the examination of one era of the Court. It remains to be seen whether this measure would stand up to more longitudinal analysis. Despite these limitations, however, this new measure includes many characteristics to endorse its future use as a measure of case salience. It is contemporaneous, replicable, nondichotomous continuous; measures mass salience; and avoids many potential sources of systematic bias. In sum, this new Case Salience Index includes all of the advantages of the existing media proxies while reducing the negative characteristics.

Even though no measure of an abstract concept such as salience is without potential problems, this measure will provide more choice for researchers wishing to assess the influence of salience. While some scholars may choose to continue using the New York Times measure, the CSI will provide an option to those scholars that seek more nuanced analyses than those available from a dichotomous measure. Although it is difficult, if not impossible, to prove one proxy for an abstract concept is “better” than another, this article’s main focus was to show that this new measure is significantly different. With the inclusion of more information, multiple media assessments of a case’s importance from around the nation, and the ability to perform more innovative statistical techniques than available with a dichotomous measure, we are confident that the CSI will be a useful measure for future studies to advance our understanding of salience and judicial behavior.

Footnotes

The authors declared that they had no conflicts of interests with respect to their authorship or the publication of this article.

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

Notes

References

Abramowitz

Alan I.

1995. “It’s Abortion, Stupid: Policy Voting in the 1992 Presidential Election.” Journal of Politics 57:176-86.

Ang

Adrain U-Jin

Peksen

Dursun

. 2007. “When Do Economic Sanctions Work? Asymmetric Perceptions, Issue Salience, and Outcomes.” Political Research Quarterly 60: 135-45.

Bailey

Michael A.

Kamoie

Brian

Maltzman

Forrest

. 2005. “Signals from the Tenth Justice: The Political Role of the Solicitor General in the Supreme Court Decision Making.” American Journal of Political Science 49:72-85.

Baird

Vanessa A.

2004. “The Effect of Politically Salient Decisions on the U.S. Supreme Court’s Agenda.” Journal of Politics 66:755-72.

Barrow

Deborah J.

Zuk

Gary

. 1990. “An Institutional Analysis of Turnover in the Lower Federal Courts, 1900-1987.” Journal of Politics 52:457-76.

Biskupic

Joan

. 1996. “Police May Use Traffic Stops for Drug Probes, Court Says.” Washington Post, June 11, Sec. A, 1.

Brenner

Saul

. 1993. “The Chief Justice’s Self Assignment of Majority Opinion in Salient Cases.” The Social Science Journal 30:143-210.

Brenner

Saul

Arrington

Theodore S.

2002. “Measuring Salience on the Supreme Court: A Research Note.” Jurimetrics 43:99-113.

Caldeira

Gregory A.

1986. “Neither the Purse nor the Sword: Dynamics of Public Confidence in the Supreme Court.” American Political Science Review 80:1209-26.

10.

Canes-Wrone

Brandice

de Marchi

Scott

. 2002. “Presidential Approval and Legislative Success.” Journal of Politics 64:491-509.

11.

Cook

Beverly

. 1993. “Measuring the Significance of U.S. Supreme Court Decisions.” Journal of Politics 55:1127-39.

12.

Edwards

George C.

III Mitchell

William

Welch

Reed

. 1995. “Explaining Presidential Approval: The Significance of Issue Salience.” American Journal of Political Science 39:108-34.

13.

Epstein

Lee

Knight

Jack

. 1998. The Choices Justices Make. Washington, DC: Congressional Quarterly.

14.

Epstein

Lee

Martin

Andrew D.

Segal

Jeffrey A.

Westerland

Chad

. “The Judicial Common Space.” Journal of Law, Economics, & Organization 23:303-25.

15.

Epstein

Lee

Segal

Jeffrey A.

2000. “Measuring Issue Salience.” American Journal of Political Science 44:66-83.

16.

Gans

Herbert J.

Deciding What’s News: A Study of CBS Evening News, NBC Nightly News, Newsweek and Time. New York: Random House.

17.

Giles

Micheal W.

Blackstone

Bethany

Vining

Richard L.

Jr.

2008. “The Supreme Court in American Democracy: Unraveling the Linkages between Public Opinion and Judicial Decision Making.” Journal of Politics 70:293-306.

18.

Giles

Michael W.

Hettinger

Virginia A.

Peppers

Todd

. 2001. “Picking Federal Judges: A Note on Policy and Partisan Selection Agendas.” Political Research Quarterly 54:623-41.

19.

Groeling

Tim

. 2010. When Politicians Attack: Party Discipline and the Media. Cambridge: Cambridge University Press.

20.

Groseclose

Timothy J.

Milyo

Jeff

. 2005. “A Measure of Media Bias.” Quarterly Journal of Economics 120:1191-237.

21.

Harris

David A.

1997. “‘Driving While Black’ and All Other Traffic Offenses: The Supreme Court and Pretextual Traffic Stops.” The Journal of Criminal Law and Criminology 87: 544-82.

22.

Hutchings

Vincent L.

1998. “Issue Salience and Support for Civil Rights Legislation among Southern Democrats.” Legislative Studies Quarterly 23:521-44.

23.

Hutchings

Vincent L.

2001. “Political Context, Issue Salience, and Selective Attentiveness: Constituent Knowledge of the Clarence Thomas Confirmation Vote.” Journal of Politics 63:846-68.

24.

King

Gary

. 2007. “An Introduction to the Dataverse Network as an Infrastructure for Data Sharing.” Sociological Methods & Research 36:173-99.

25.

Levit

Janet Koven

. 1996. “Pretextual Traffic Stops: United States v. Whren and the Death of Terry v. Ohio.” Loyola University Chicago Law Journal 28:145-87.

26.

Maltzman

Forrest

Wahlbeck

Paul J.

2003. “Salience or Politics: New York Times Coverage of the Supreme Court.” Paper presented at the annual meeting of the Midwest Political Science Association, Chicago, IL. April 3-6.

27.

Martin

Andrew, D.

Quinn

Kevin M.

2002. “Dynamic Ideal Point Estimation via Markov Chain Monte Carlo for the U.S. Supreme Court, 1953-1999.” Political Analysis 10:134-53.

28.

McAtee

Andrea

McGuire

Kevin T.

2007. “Lawyers, Justices, and Issue Salience: When and How Do Legal Arguments Affect the U.S. Supreme Court?” Law & Society Review 41:259-78.

29.

McGuire

Kevin T.

Stimson

James A.

2004. “The Least Dangerous Branch Revisited: New Evidence on Supreme Court Responsiveness to Public Preferences.” Journal of Politics 66:1018-35.

30.

Niemi

Richard G.

Bartels

Larry M.

1985. “New Measures of Issue Salience: An Evaluation.” Journal of Politics 47:1212-20.

31.

Norpoth

Helmut

Segal

Jeffrey A.

1994. “Comment: Popular Influence on Supreme Court Decisions.” American Political Science Review 88:711-16.

32.

Pritchett

C. Herman

. 1941. “Divisions of Opinion among Justices of the U.S. Supreme Court, 1939-1941.” American Political Science Review 35:890-98.

33.

Segal

Jeffrey A.

Cover

Albert

. 1989. “Ideological Values and the Votes of U.S. Supreme Court Justices.” American Political Science Review 83:557-65.

34.

Segal

Jeffrey A.

Spaeth

Harold J.

1996. “The Influence of Stare Decisis on the Votes of United States Supreme Court Justices.” American Journal of Political Science 40: 971-1003.

35.

Segal

Jeffrey A.

Spaeth

Harold J.

2002. The Supreme Court and the Attitudinal Model Revisited. New York: Cambridge University Press.

36.

Sklansky

David A.

1997. “Traffic Stops, Minority Motorists, and the Future of the Fourth Amendment.” The Supreme Court Review 1997:271-329.

37.

Smith

Mark A.

2001. “The Contingent Effects of Ballot Initiatives and Candidate Races on Turnout.” American Journal of Political Science 45:700-6.

38.

Spaeth

Harold J.

2008. The Original United States Supreme Court Judicial Database, 1953-2007 Terms. Ann Arbor, MI: Inter- University Consortium for Political and Social Research.

39.

Spaeth

Harold J.

Segal

Jeffrey A.

1999. Majority Rule or Minority Will: Adherence to Precedent on the U.S. Supreme Court. New York: Cambridge University Press.

40.

Spriggs

James F.

II Maltzman

Forrest

Wahlbeck

Paul J.

1999. “Bargaining on the U.S. Supreme Court: Justices’ Responses to Majority Opinion Drafts.” Journal of Politics 61:485-506.

41.

Unah

Isaac

Hancock

Ange-Marie

. 2006. “U.S. Supreme Court Decision Making, Case Salience, and the Attitudinal Model.” Law & Policy 28:295-320.

42.

Vining

Richard

Wilhelm

Teena

. 2010. “Issue Salience in State Courts of Last Resort.” Political Research Quarterly. Published electronically August. doi: 10.1177/106591291036122210.1177/1065912910361222.