Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 27 March 2017

Expectancy Theory and the Election of Judges: Do Judicial Campaigns Really Stink?

Summary and Keywords

Even though most judges in the United States stand for election in the context of strong normative objections to the practice of electing judges, political scientists have produced a surprisingly thin theoretical framework for understanding how judicial campaigns affect voters. This paucity of research is particularly surprising given the increasingly politicized environment in which judicial elections operate. The literature on judicial campaigns is well-served to draw upon the well-trodden research about campaign effects for executive and legislative office. In some important respects, however, judicial contests differ from those for executive or legislative office. To this end, the Expectancy Theory pioneered by James L. Gibson provides an important theoretical development, emphasizing that the effects of judicial campaigns are highly conditional upon variation in voters’ willingness to tolerate different types of campaign activity. Moreover, the effects of campaigns are highly dependent on the context of both institutional design and voters’ own experiences with judicial elections.

Keywords: judicial elections, negative campaigning, judicial legitimacy, state supreme courts, Expectancy Theory


In his 2012 book on state judicial elections, Gibson reports a remarkable finding, reproduced here in Figure 1 (see 2012, p. 62). Embedded within a representative survey of the residents of Kentucky during the Kentucky judicial elections of 2006, Gibson implemented an experiment featuring three actual judicial campaign ads. The respondents were randomly assigned to hear (over the telephone) one of the three ads read to them by the interviewer. After they heard the ad, the respondent was asked how appropriate such an ad was for a judicial election. Figure 1 plots the results of that question, where higher values indicate a higher proportion of respondents felt the ad was appropriate. As the figure makes clear, the first two ads (A and B in Figure 1) were deemed appropriate by a majority of the respondents. The third ad, however, is different—less than one-fifth of those queried judged it appropriate. Something about Janet Stumbo’s attack on David Barber seems to have crossed some sort of line for most of the constituents of the Kentucky Supreme Court.

Expectancy Theory and the Election of Judges: Do Judicial Campaigns Really Stink?Click to view larger

Figure 1. Assessments of three attack advertisements broadcast by Kentucky judges, 2006.

Source: Gibson (2012, p. 62).

Note: Total N = 1,032. Individual treatment condition Ns vary from 332 to 351. Cross-condition difference of means tests (on the uncollapsed response set): η‎ = .38, p < .001.

The ads read:

Ad A. [Announcer]: In 2003, Circuit Judge Bill Cunningham tried to make six rapists eligible for parole. One had been out on parole for only 12 hours when he raped a 14 year old and made her mother watch. Bill Cunningham already had tried to reduce their sentences, but our Supreme Court said no. Bill Cunningham said it was folly and a blatant injustice to keep these rapists in prison. Judge Rick Johnson believes that a life sentence means a life sentence. Please, vote for Rick Johnson for Justice on the Supreme Court.

Ad B. [Announcer]: John Roach says he’s tough on crime but Judge Mary Noble has put thousands of criminals behind bars. John Roach, none. Judge Mary Noble has helped dozens of lives through her Drug Court Program. John Roach, none. Judge Mary Noble has been elected by the people twice. John Roach, none. Elect a real judge to the Supreme Court. Vote for Judge Mary Noble.

Ad C. [Announcer]: David Barber is confused. He’s now airing an ad that says Janet Stumbo wrote the Supreme Court opinion in the Morse Fetal Homicide Case. Barber can’t tell the boys from the girls. The Morse opinion was written by Justice Bill Cooper. More confusing is that Cooper’s opinion upheld the decision with which Barber concurred. He’s attacking the Supreme Court for agreeing with him. David Barber: confused about his own opinions. Is he a judge, or just another politician? On November 7th elect a judge: Janet Stumbo.

What is it about the third advertisement that makes it so objectionable? Gibson offered some speculation, but no definitive evidence. According to Gibson’s thoughts, the ad is confusing and hard to follow and understand, except for the concluding sentence: “Is he a judge, or just another politician?” Assuming a recency effect through which the most recent information prior to the question has disproportionate influence over the answers to the question, the respondents might have been influenced by what seemed to be the politicization of the judicial campaign.1 “Is he a judge or just another politician?” is the sort of question a politician would ask; for a judicial candidate to ask the question apparently seemed inappropriate to most Kentuckians.

Gibson and Caldeira (2009a) have documented a similar effect of the ad campaigns for and against Samuel Alito’s nomination to the U.S. Supreme Court: some ads broadcast during the confirmation fight “crossed the line,” resulting in the politicization of judicial politics and subtracting from the institutional legitimacy of the U.S. Supreme Court (see also Gibson & Caldeira, 2009b).2 The ads, most sponsored by interest groups favoring or opposing Alito’s confirmation, were not unlike the ads one sees in presidential elections.3 The ads seemed to politicize the Supreme Court, resulting in a withdrawal of support from the institution.4

Just as interesting, perhaps, is the lack of effect of the other two ads presented to the respondents in the Kentucky experiment (Figure 1). Rick Johnson’s attack on Bill Cunningham is certainly hard-hitting, claiming that Cunningham believes it a “folly” and a “blatant injustice” to keep rapists behind bars. Yet, despite this harsh rhetoric, most Kentuckians did not view the ad as inappropriate, even in a race for a judgeship.

A similar conundrum is reported in the research of Gibson, Gottfried, Delli Carpini, and Jamieson (2011) on the Pennsylvania judicial elections of 2007. Using an Internet survey, a sample of Pennsylvanians was randomly assigned to watch one of four two-minute-long videos. One concerned the public schools and had nothing to do with judges; it served as a control condition. The second video was a compilation of four fairly vigorous attacks (including the video of the Johnson attack on Cunningham), while the third was drawn from a speech by the leader of PACleanSweep urging voters to vote all judges out of office owing to their participation in the much-hated pay raise for legislators and judges in Pennsylvania. The final treatment was a traditional, low-fanfare compilation of ads promoting the candidates through lists of endorsements from various unions, interest groups, and prominent politicians.5

Expectancy Theory and the Election of Judges: Do Judicial Campaigns Really Stink?Click to view larger

Figure 2. Results from the Campaign Ads Video Experiment, Pennsylvania judicial elections, 2007.

Source: Gibson et al. (2011, p. 553).

Note: Eta = .08, p = .029, N = 1,558.

Figure 2 reports their results. The dependent variable (Y-axis) in the graph is change in institutional support from before the election to after the election, a variable that could range from a negative number (decline in supportive replies) to a positive number (increase in supportive replies). As can be seen in the graph, average support for the Pennsylvania Supreme Court increased across all four conditions, to a statistically significant degree. However, the respondents in the control condition (who watched a video on public education) increased their support for the court at about double the rate of the respondents in the other three conditions.

Gibson and his colleagues drew several conclusions from this graph. First, the election by itself seemed to boost the legitimacy of the Pennsylvania Supreme Court. Second, direct exposure6 to judicial campaign ads cut that boost by one-half. Third, the net increase in legitimacy was still positive. Finally, and most relevant to our concerns here, little difference exists across the three campaign ad presentations. Strong “new-style” attack ads had roughly the same effect as tepid and conventional promotional ads.

How is it that a fairly innocuous endorsement ad of the type long used in judicial campaigns has no less impact than the hard-hitting attack ads? Unfortunately, extant research does not offer much clear guidance for understanding the impact on citizens of campaigns and campaigning in judicial elections. What little evidence exists suggests that there are three types of determinants of the consequence of ads: (1) the content of the ads themselves, (2) the context in which the ads are displayed, and most importantly, (3) the expectations citizens hold about what constitutes appropriate content within the context.

Our purpose in this chapter is to develop a theory of campaign impact as it pertains to the election campaigns of judges. In doing so, we will borrow from the more general literature on campaign advertisements will be borrowed from, although judicial elections are in some respects different from ordinary campaigns for public office. Legal scholars have long maintained that judicial elections are sui generis. We use that insight in a more general theoretical context specifying the nature of the normative expectations citizens hold about various types of campaigns. We acknowledge that considerable heterogeneity exists in expectations and expectations are subject to change over time with experience.7 Because expectations play such a crucial role in this theory, we begin this chapter by discussing Expectancy Theory as previously developed by Gibson (2012).

Expectancy Theory

People seem to hold expectations about what constitutes appropriate conduct within campaigns for public office. In the run-up to the presidential campaign of 2016, these expectations seemed to be especially important as one candidate, Donald Trump, acted in such a way as to violate what some thought are the norms of appropriate behavior for a presidential candidate, such as speaking largely off-the-cuff and denying some journalists credentials to cover his events in retaliation for negative coverage of his campaign (neither of which, of course, is unprecedented in American politics) (Noble, 2015). It is important to note, however, that Trump clearly did not violate the expectations of everyone. For example, Trump made a comment about fellow candidate Carly Fiorina, stating “‘Look at that face! … Would anyone vote for that? Can you imagine that, the face of our next president?! … I mean, she’s a woman, and I’m not s’posedta say bad things, but really, folks, come on. Are we serious?’” (Solotaroff, 2015). Trump’s comments were widely criticized in the public, the media, and by his fellow candidates. Still, polls indicated that only 59% of Republican voters thought that Trump’s comments about Fiorina’s appearance were inappropriate (Public Policy Polling, 2015).

This, then, leads to the conclusion that expectations of presidential candidates are probably heterogeneous, and perhaps even that no single expectation is consensually held by the electorate (and the media). Put differently, individuals differ in the extent to which they feel that contentious campaign activity—even negative campaigning about candidates’ personal appearance—is fair game in electoral politics. Thus, campaign activities may be able to “cross the line,” but the “line” different people hold differs.

Nowhere is this more true than in judicial elections, which are undoubtedly the most normatively constrained elections in American politics.8 Though the political science evidence on judging has long shown that judges vote in accordance with their policy preferences, just as legislators do, legal scholars and reformers have criticized the practice of electing judges,9 fearing that judicial elections make judges beholden to campaign contributors, reduce the legitimacy of judicial institutions, and turn judges into nothing more than ordinary politicians.10 Indeed, the U.S. Supreme Court has recently joined this chorus, permitting states to circumscribe the campaign behavior of judicial candidates more restrictively than other types of candidates “because the role of judges differs from the role of politicians” (Williams-Yulee v. Florida Bar at 10).

The Court’s logic in this decision is based on the assumption that judicial representation differs from legislative representation, stating that “[p]oliticians are expected to be appropriately responsive to the preferences of their supporters … In deciding cases, a judge is not to follow the preferences of his supporters, or provide any special consideration to his campaign donors” (10–11).11

Yet, the Supreme Court’s view of representation differs from that of the American people. While the Court’s decision is based on the claim that a good judge should not respond to political pressure, Gibson (2012) finds that the public does not hold this view nearly so strongly (see also Gibson, 2011). Indeed, according to Gibson’s 2006 survey of Kentuckians, nearly a majority (46.5%) of respondents believe that it is “Very Important” that a state supreme court justice “be involved in politics, since ultimately they should represent the majority.” Nearly one-fifth of the respondents believe that a good state supreme court justice should base her decisions on her party affiliations, and over two-fifths of respondents (43.7%) believe that a good supreme court justice should give the respondent’s ideology a voice in constitutional interpretation.12 The expectations of these Kentuckians clearly differ from those of the Supreme Court justices, although they do not seem to differ greatly from the rest of the American people (Gibson & Caldeira, 2009a).

Thus, in all types of elections (and particularly in those that select judges), people hold normative expectations about appropriate campaign activities, but people hold different expectations. What is acceptable behavior to some in campaigns is different from what others are willing to accept as appropriate behavior in other parts of the electoral arena. Likewise, citizens vary in their beliefs about “good” and “bad” behaviors for elected officials; while the Constitution created the judiciary as a fundamentally unrepresentative branch of government, Gibson’s survey evidence suggests that many citizens believe courts should be representative.

To this end, Gibson (2012) has relied upon these observations in developing his Expectancy Theory. He argues that “[l]egitimacy-stripping dissatisfaction with courts is hypothesized to flow from the intersection of citizens’ expectations and perceptions, with those believing that certain activity is inappropriate for judicial campaigns withdrawing support from the institution” when they are exposed to the objectionable activity (2012, p. 106). In other words, citizens who view as normatively appropriate highly politicized campaigning are unbothered by it when they see it in practice and, by extension, do not update their evaluations of the institution after seeing a contentious, politicized campaign. Yet, for those who view such campaigning as inappropriate, exposure to these symbols of politicization and “politics as usual” is associated with a withdrawal of support for the institution. The contributions of Expectancy Theory are that (a) it stresses normative expectations of how candidates ought to behave, (b) it recognizes heterogeneity in the expectations people hold, and (c) it posits that expectations interact with perceptions of activity to produce judgements of courts. Nevertheless, an important part of these perceptions is the content of the ads broadcast in judicial races.

Ad Content: Negativity in Judicial Elections

Perhaps nothing in politics is as emblematic of “politics as usual” as the attack advertisement; thus, it is unsurprising that political scientists have paid a great deal of attention to the content of ads run in judicial elections and their effects on voters (e.g., Champagne, 2002; Hall, 2015; Iyengar, 2002). Most studies, relying on data collected by Campaign Media Analysis Group (CMAG) and made available by the Brennan Center for Justice, characterize ads as either “promoting,” “contrasting,” or “attacking” (Caufield, 2007).

According to Hall (2015), the plurality of advertisements aired in judicial elections between 2002 and 2008 were promote ads (which discuss a candidate’s strengths without reference to her opponent), airing in 69% of contested state supreme court contests in 2008. Contrast advertisements (in which a candidate is compared to his opponent without criticizing her) were aired in 31% of contested state supreme court elections during that election cycle, while only one in five (20.7%) contested state supreme court contests featured attack advertisements (in which one candidate criticizes her opponent). Indeed, in a strong contrast to claims that state supreme court elections are dominated by highly negative campaigning, Hall shows that only about one in five contested judicial elections feature even a single negative advertisement.

The traditional concern with attack advertising, which has been subject to much scrutiny by political scientists studying presidential elections, is that it demobilizes the electorate and so depresses turnout in elections (e.g., Ansolabehere & Iyengar, 1995). Much political science research challenges this hypothesis, suggesting that the effects of negative advertising are highly conditional on timing (Geer, 2006; Krupnikov, 2011). Indeed, in a recent study, Mattes and Redlawsk (2014) find that citizens are generally accepting of negative campaigning; though voters may not love negative ads, they look past the negativity and learn from the issue content of the ads (see also Sides, Lipsitz, & Grossmann, 2010). Gibson (2012, p. 119) presents some evidence of this fact in the judicial context: citizens shown an attack ad, rather than a promoting ad, viewed that ad as significantly more negative and more partisan, but not less fair. While citizens may not love negative advertisements, they, on balance, do not view them as inappropriate for a campaign context.

The potentially deleterious effects of negative advertising may be particularly acute in lower-ballot races, where this negative information may make up a higher proportion of the information readily available to voters. This concern is compounded by recent legal developments, particularly the U.S. Supreme Court’s 2002 ruling in Republican Party of Minnesota v. White, which held that prohibitions on the ability of judicial candidates to announce their policy views were unconstitutional because they violated the free speech rights of judicial candidates. Critics (including the minority in the decision) howled, claiming that the decision would greatly harm the judiciary.13

Hall and Bonneau (2013) took these complaints as a challenge and sought to investigate rigorously the effect of the decision. While they do not address legitimacy specifically, they examine the effects of attack ads and liberalized speech codes on participation in judicial elections, finding that participation in these elections is enhanced by liberalized speech codes and negative advertising.14 Hall (2015) explores this hypothesis further, examining whether attack advertisements are associated with higher levels of turnout in state supreme court elections based on ballot type. She finds that negative advertising is unrelated to citizen participation in partisan elections, though such advertisements actually increase participation in nonpartisan elections. In short, there seem to be no evidence of demobilizing consequences of negative campaigning on voter participation in judicial elections.

In the context of Expectancy Theory these results are not surprising. Because a sizable proportion of voters see courts as political institutions, they are not off-put from voting because of negative television advertising. Expectations condition the impact of perceptions of campaign activities.

Of course, the larger concern of the White detractors was institutional legitimacy, rather than voter participation. As it turns out, citizens vary enormously in the extent to which they view different types of campaign activities as acceptable. Gibson (2012, p. 13) reported the results from a national survey on whether various types of campaign activities were appropriate for judges. We reproduce those findings here as Figure 3.

Expectancy Theory and the Election of Judges: Do Judicial Campaigns Really Stink?Click to view larger

Figure 3. The campaign expectations of the American people.

Source: Gibson (2012, p. 13). These results are from the 2008 Freedom and Tolerance Survey, Weidenbaum Center, Washington University in St. Louis.

The questions read:

In many states, the justices of the state Supreme Court are elected and they therefore have to campaign to get people to vote for them. Do you strongly agree strongly, agree, are uncertain, disagree or disagree strongly with the following statement: When state supreme court judges run for election they should make their views known about controversial issues, such as abortion, gun control, the death penalty, and gay marriage.

Do you agree strongly, agree, are uncertain, disagree, or disagree strongly with the following statement: When state supreme court judges run for election they should be allowed to attack their opponents’ positions on issues that might come before the court.

Please tell me if you agree strongly, agree, are uncertain, disagree, or disagree strongly with the following statement: When state supreme court judges run for election, they should NOT accept campaign contributions from any lawyer or law firm.

Figure 3 shows the judgements of a national sample on the appropriateness of three types of campaign activities. Clearly, a very large proportion of the American people judge it appropriate for candidates for judicial office to “make their views known about controversial issues, such as abortion, gun control, the death penalty, and gay marriage.” Just as clearly, people do not think it appropriate for candidates to “accept campaign contributions from any lawyer or law firm.” In-between are attack ads, which a considerable majority of Americans find acceptable. These normative expectations guide how people evaluate the campaign activities they observe in judicial elections.

Political Versus Politicized

One suggestion from this analysis is that it may be useful to characterize judicial ads as “politicized” versus “not,” with perhaps the key distinction being between issue-oriented attacks and those waged on non-policy grounds, associating judges with much-hated politicians. There seems to be a distinction between ads that implicitly claim that judges “are politicians in robes” and ads that address, either honestly or not, issues about which a judicial candidate might reasonably weigh in. By this logic, Rick Johnson’s attack on Bill Cunningham seems to be fair game because it addresses an issue clearly relevant to selecting judges (even if the facts of Johnson’s claim may not be crystal clear). Janet Stumbo’s attack on David Barber (“is he a judge or just another politician?”) seems to have backfired for Stumbo because it crossed the line by politicizing the judiciary.15 It seems that the American people distinguish between hard-hitting discussions of policies before the courts and ads that confuse the boundaries of the judiciary, failing to recognize that courts are not just another political institution. The latter seem to have consequences for institutional legitimacy.

The simple hypothesis suggested by the current literature is that policy discussions—even of the attack variety—seem to have few negative consequences for the judiciary as long as they focus on policies relevant to courts. Invoking “judges as politicians in robes” is different because it moves the discussion away from the legal issues and toward the boundary between judicial and other political institutions. Moreover, it seems likely that candidates presenting these ads do not profit from them, while the institution they hope to serve is injured, to at least some degree, by them.

The Effects of Campaign Context

Political scientists have recently conclusively documented that both institutional context and citizens’ experiences with those institutional contexts have a great deal to do with the effects of judicial campaigns (Bonneau & Cann, 2015). Institutional context dramatically structures judicial campaigns. From the point of view of the complainers about the negative consequences of judicial elections, the problem turns out to be nonpartisan elections. When partisan labels are removed from the ballot, voter participation declines (Bonneau & Hall, 2009), voters are less likely to cast votes for candidates who share their party affiliation (Klein & Baum, 2001), the incumbency advantage is weakened (Hall, 2014), challengers are less likely to emerge (Nelson, 2011), and the influence of much-ridiculed campaign ads increases (Canes-Wrone & Clark, 2009).

The institutional context also shapes the types of advertisements that candidates choose to run. Hall (2015) documents that ads were aired in 78.1% of all contested open seats, 68.9% of contested nonpartisan races featuring an incumbent, but only 41.0% of contested partisan elections involving an incumbent. Among types of advertisements, the differences are similarly stark, with promote ads broadcast in 62.3% of contested nonpartisan elections with an incumbent but only 41.0% percent of contested partisan elections with incumbents (p. 77). The differences are smaller for attack advertisements, though. Attack advertisements were seen in only 15.4% of partisan elections involving an incumbent, but a slightly higher 21.3% of nonpartisan races involving an incumbent. Just over one-third (34.4%) of open seat races involved attack advertising.

Current research suggests that partisanship may inoculate against many of the most scandalous ads broadcast in judicial elections. When voters are given information about the parties of the candidates, the value of additional information is reduced. When party information is harder to acquire, campaign activities matter more. Partisanship also likely provides a motivated reasoning heuristic for viewing the world of politics and campaign activities. In the absence of readily available party cues, the door is opened to the influence of scurrilous attacks through advertising. Ad content interacts with the institutional context, especially in nonpartisan elections.

Another aspect of context pertains to institutions more generally, particularly the extent to which campaign advertisements in judicial elections affect voters differently because these elections are judicial, rather than legislative or executive. Indeed, many of the complaints against judicial elections could just as well be directed at many types of elections in the United States. Difficulties in recruiting quality candidates to run for office, gaining and keeping voter attention, and mobilizing voters to cast ballots are inherent in any electoral context. Similarly, just as not all state supreme court races feature advertisements, not all congressional elections do, either. Indeed, Jacobson (2013) reports that only about two-thirds of House campaigns use paid television advertising (p. 99). Likewise, Jacobson notes that, though negative attacks in legislative campaigns were once seen as “a sign of desperation, ineffective, and likely to backfire,” negative advertisements have become commonplace in legislative campaigns since the mid-1980s (p. 106). Similarly, Geer (2006) demonstrates that negativity has been gradually on the rise in presidential campaigns since 1960. It is hardly surprising that judicial campaigns would follow this trend.

Still, it is unclear that the politicization of legislative elections has negative consequences for that institution. If negative advertising has strongly deleterious effects, then turnout should have dropped precipitously with the advent of negative advertising in those contests. Yet, Jacobson (2013) notes that participation in legislative elections has been fairly stable over time, declining noticeably between 1960 and 2000 and increasing slightly since then.

The broader effects of negative campaigning on trust in government are also less clear: Lau, Sigelman, and Rovner (2007), conducting a meta-analysis of studies, suggest that negative campaigning slightly lowers trust in government, while Lau and Pomper (2004) find that negative campaigning has no effect on trust in government. Jackson, Mondak, and Huckfeldt (2009) extend these findings, demonstrating that exposure to negative advertisements has no effect on approval of Congress or perceptions of congressional performance. Likewise, Lau and Rovner (2009) suggest that negative advertisements may actually stimulate interest in campaigns and learning about candidates and issues. Examining the analysis to include incivility, rather than negativity, the results are similarly conflicted: Mutz and Reeves (2005) present experimental evidence that incivility harms political trust while Brooks and Geer (2007) find no evidence that incivility is associated with lower levels of political trust. Thus, it seems that the increased politicization of these elections has not resulted in a major backlash against the institution.

Implications for Judicial Legitimacy

Of course, neither turnout nor trust is the same thing as legitimacy; the question remains of how politicized campaigning affects the legitimacy of legislative institutions. Gibson has taken up this challenge. In his analysis of Kentucky, Gibson (2012) often compared judicial and legislative elections (see also Gibson, 2008a). Perhaps surprisingly, his findings indicate that elections for the two branches of government had many similarities. For instance, Gibson documents the harmful effects of campaign contributions. However, the effects (from the point of view of the institution’s constituents) are not lesser or greater for courts as compared to legislatures. Both institutions are harmed by current policy on campaign contributions and independent expenditures. Still, a great deal more needs to be learned about how citizens fashion their expectations of different legal and political institutions.

Beyond institutions, citizens’ experiences with judicial elections both (a) vary and (b) may affect the extent to which they are affected by judicial campaigns. Remarkable variability exists in the degree to which judicial elections have the attributes of real elections. Hall (2015) notes that 53.4% of nonpartisan state supreme court elections were contested between 1980 and 2010, while 72.8% of partisan elections were contested over this time period, though contestation rates increased dramatically beginning in the mid- to late 1980s. In recent years, contestation rates in partisan elections have climbed to over 90%, though the similar statistic for nonpartisan elections has gradually declined, with only 47.6% of nonpartisan elections contested in 2010 (pp. 45–47). Even on the simple criterion of whether ads are broadcast on television, only 62% of contested judicial elections in 2012 involved any ads, and even this is a relatively new phenomenon; in 2002 only two-in-five contested elections (42.9%) involved television advertising (Kritzer, 2015; Hall, 2015). In short, even when institutional rules (e.g., whether the elections are partisan or nonpartisan) may be similar among states, there is much more variability in what actually happens during elections than might be realized. This research suggests that scholars need to move beyond institutional rules toward an emphasis on citizens’ experiences to understand how these elections shape citizens’ views of the judiciary.

A natural hypothesis is that experience with vigorous campaigning structures both citizens’ expectations and, by extension, the effects that these campaigns have on citizens’ support for the institution. This may happen both in the short- and long-term. Perhaps over the course of an election, churlish ads may be judged as less shocking as citizens begin to view them as part of a larger back-and-forth between candidates about which candidate is better suited to sit on the state supreme court.16

Similarly, citizens may, over the course of election cycles, become more accustomed to seeing hostile television advertisements and therefore begin to expect them. That is, over time, “shocking” campaign ads may lose their shock value. Thus, one obvious empirical hypothesis is that people living in states in which vigorous judicial campaigns are commonplace (e.g., Ohio; see Baum, 1987; Rock & Baum, 2010; Nelson, Caufield, & Martin, 2013) have perhaps become inured to these ads and, as a result, that these advertisements play a more minor role in citizens’ evaluations than they do in states where such campaigning is less commonplace. Citizens with experiences with churlish ads may draw information from them about candidates and issues, but not information about whether courts are legitimate. In contrast, people living in states in which judges do not face the electorate are more likely to find the idea of judicial campaigns unsavory, and therefore individual instances of campaign activity may have larger effects on evaluations among citizens in these states. All of this suggests that expectations of judicial campaigns may vary significantly across types of selection systems, and that even among those states using elections, considerable variability may exist, just as the states vary in the use of vigorous elections.

Gibson (2009) finds some suggestive evidence of this kind of effect, comparing the deleterious effects of policy statements by judicial candidates on institutional legitimacy between citizens who live in states that use judicial elections and those living in states that do not use judicial elections. He finds that, in states that use judicial elections, such policy talk has no deleterious effects on judicial legitimacy, but that policy talk harms legitimacy among those who reside in states that do not experience judicial elections. Importantly, however, Gibson’s findings do not hold for other types of campaign activity, namely attack ads or campaign contributions. His conjecture is that attack ads are widely accepted, campaign contributions are widely rejected, but policy talk is new and how it is evaluated depends upon one’s expectations for judicial candidates, which in turn depend on experiences with prior judicial campaigns.

Moreover, citizens may vary in the way in which they categorize campaign activities; not all negative ads are equally negative to all voters. Sigelman and Kugler (2003), studying information processing and negative campaigning in three 1998 gubernatorial races, find that citizens vary immensely in the extent to which they perceive a campaign as negative. Ridout and Fowler (2012) extend these findings further, demonstrating that perceptions of campaign negativity are associated with both exposure to those advertisements as well as exposure to local news media (see also Fridkin & Kenney, 2011). Gibson (2012, p. 112) notes a similar finding in his study of Kentucky judicial elections, with respondents almost evenly split on whether their assessments of an advertisement presented to them in the 2006 judicial contest were negative (43.1%) and partisan (50.1%) (p. 118). Thus, it seems clear that negativity is a subjective, rather than objective, characteristic; the same citizens can perceive the same campaign (or even the same advertisement) and come away with different feelings.

Finally, most research on the effects of campaign ads focuses on the words in the ads, not the images and symbols that so often accompany the words. This is especially important in judicial elections inasmuch as television advertisements often associate candidates with the legitimizing symbols of judicial authority (e.g., judicial robes, Lady Justice). Gibson, Lodge, and Woodson (2014) have shown that exposure to judicial symbols can significantly influence public attitudes toward courts. Other aspects of the election videos may convey information to observers as well. More research into all aspects of campaign communications—verbal and nonverbal—is essential.

Hence, there are two potential sources of variation here: (1) Expectancy Theory suggests individual-level variation in the extent to which voters are willing to accept various types of campaign activity, and (2) the political psychology literature suggests that citizens vary in their perceptions of whether those campaign activities are occurring. Without question, more research is necessary to determine both the short- and long-term consequences of television advertising on both citizens’ expectations about campaigns and their overall evaluations of institutions.

Concluding Thoughts

The clearest and most obvious conclusion to emerge from this discussion is that we actually know precious little about how judicial elections are perceived and evaluated by the courts’ constituents. It seems clear that voters’ perceptions of the tone of campaigns and of individual advertisements vary widely, as do citizens’ expectations for the role of a judge in the American political system. What one voter perceives as an appropriate attack may be seen by her neighbor as an unfair assault on a candidate or the institution. A failure to account for these differences may account for some of the confusing findings in the larger literature on the effects of negative campaigning on voter behavior.

Second, voters’ expectations about appropriate campaign behavior vary widely, with some more comfortable with vigorous campaigning than others. These individual-level differences in normative expectations about appropriate campaigning interact with the content of advertisements in a way that clouds the effects of advertisements on their own. We hypothesize that the most deleterious advertisements have effects among voters who view those advertisements as inappropriate, while the same advertisements will have less harmful consequences on the evaluations of citizens who view those attacks as appropriate.

Third, scholars need to refine their understanding of negativity. In addition to varying at the level of the voter, rather than the ad, there are different types of negativity. Perhaps, for example, the advertisement that opened this article had such a deleterious effect not because it was a negative advertisement but because it was a particular type of negative advertisement: one that likened courts to other political institutions rather than debating issues on policy grounds. Further research needs to delineate the effects of different types of attacks, finding how different negative messages interact with voters’ expectations to affect their behavior.

Fourth, voters’ experiences with judicial elections may matter, both in the short- and long-term. Voters who have long been exposed to negative advertisements may both be more likely to view those advertisements as appropriate and be less influenced by formerly “shocking” claims by virtue of desensitization via repeated exposure. Future research needs to address how, over repeated electoral cycles, voters become accustomed to negative advertising and how, in judicial races specifically, short-term variation in campaign tone interacts with voter expectations to affect voters’ perceptions of their state courts and those who seek to sit on the bench.

Much more research is needed on the nonverbal aspects of campaign ads, especially in the context of judicial elections, where the symbols of judiciary authority are so often invoked. Brader (2006) has shown conclusively that these types of attributes have enormous effects on voters in other types of elections, while Gibson, Lodge, and Woodson (2014) have shown that the symbols of judicial authority have powerful effects on individual-level evaluations of judicial institutions. Yet, perhaps surprisingly, we know little about how the symbols of judicial authority, as evoked in judicial ads, affect voter behavior and individual-level support for the institution.

Additionally, we know little about how voters’ evaluations of judicial candidates affect individual-level evaluations of judicial institutions. Gibson and Caldeira (2009a) suggest that this may be possible, demonstrating that feelings about Alito as a judicial nominee affected views of the U.S. Supreme Court. However, the mechanism behind this effect is unclear and requires additional study.

Finally, institutional context simultaneously does and does not matter. This context matters in that institutional rules (such as the presence of party labels on the ballot) affect the types and quantity of advertisements that candidates choose to run and the effectiveness of these advertisements on voter participation. Context does not matter in that voters seem to perceive legislative and judicial campaigns very similarly, thus suggesting that voter reactions to judicial campaigns are not totally different from their responses to a presidential or congressional campaign. Taken together, our discussion of institutional context underscores why the study of state judicial campaigns provides political scientists the best of both worlds: the ability to study the effects of varied institutional rules in a context that is highly generalizable to other electoral arenas.


This article relies in part on survey data collected with the assistance of the Weidenbaum Center at Washington University in St. Louis. We are indebted to the Center and its director, Steve Smith, for making this research possible.

Cases Cited

Republican Party of Minnesota, et al. v. Suzanne White, Chairperson, Minnesota Board of Judicial Standards, et al. 536 U.S. 765 (2002).Find this resource:

Lanell Williams-Yulee v. The Florida Bar. 575 U.S. __ (2015). Slip Op.Find this resource:


Ansolabehere, S., & Iyengar, S. (1995). Going negative: How political advertisements shrink and polarize the electorate. New York: Free Press.Find this resource:

Barabas, J., & Jerit, J. (2010). Are survey experiments externally valid? American Political Science Review, 104(2), 226–242.Find this resource:

Bartels, B. L., & Johnston, C. D. (2013). On the ideological foundations of Supreme Court legitimacy in the American public. American Journal of Political Science, 57(1), 184–199.Find this resource:

Baum, L. (1987). Explaining the vote in judicial elections: The 1984 Ohio Supreme Court elections. Western Political Quarterly, 40(2), 361–371.Find this resource:

Bonneau, C. W., & Cann, D. M. (2015). Voters’ verdicts: Citizens, campaigns, and institutions in state supreme court elections. Charlottesville: University of Virginia Press.Find this resource:

Bonneau, C. W., & Hall, M. G. (2009). In Defense of Judicial Elections. New York: Routledge.Find this resource:

Bonneau, C. W., Hall, M. G., & Streb, M. J. (2011). White noise: The unrealized effects of Republican Party of Minnesota v. White on judicial elections. Justice System Journal, 32(3), 247–268.Find this resource:

Brace, P., & Boyea, B. D. (2008). State public opinion, the death penalty, and the practice of electing judges. American Journal of Political Science, 52(2), 360–372.Find this resource:

Brader, T. (2006). Campaigning for hearts and minds: How emotional appeals in political ads work. Chicago: University of Chicago Press.Find this resource:

Brooks, D., & Geer, J. G. (2007). Beyond negativity: The effects of incivility on the electorate. American Journal of Political Science, 51(1), 1–16.Find this resource:

Canes-Wrone, B., & Clark, T. S. (2009). Judicial independence and nonpartisan elections. Wisconsin Law Review, 2009(1), 21–65.Find this resource:

Canes-Wrone, B., Clark, T. S., & Kelly, J. P. (2014). Judicial selection and death penalty decisions. American Political Science Review, 108(1), 23–39.Find this resource:

Caufield, R. P. (2007). Campaigning for judge: How the tone and tenor of judicial election campaigns is changing. In M. Streb (Ed.), Running for judge (pp. 34–58). New York: New York University Press.Find this resource:

Champagne, A. (2002). Televised ads in judicial campaigns. Indiana Law Review, 35(3), 669–689.Find this resource:

Christenson, D. P., & Glick, D. M. (2015). Chief Justice Roberts’s health care decision disrobed: The microfoundations of the Supreme Court’s legitimacy. American Journal of Political Science, 59(2), 403–418.Find this resource:

Driscoll, A., & Nelson, M. J. (2012). The 2011 judicial elections in Bolivia. Electoral Studies, 31(3), 628–639.Find this resource:

Driscoll, A., & Nelson, M. J. (2015). Judicial selection and the democratization of justice: Lessons from the Bolivian judicial elections. Journal of Law and Courts, 3(1), 115–148.Find this resource:

Fridkin, K. L., & Kenney, P. J. (2011). Variability in citizens’ reactions to different types of negative campaigns. American Journal of Political Science, 55(2), 307–325.Find this resource:

Geer, J. G. (2006). In defense of negativity: Attack ads in presidential campaigns. Chicago: University of Chicago Press.Find this resource:

Gibson, J. L. (2008a). Challenges to the impartiality of state supreme courts: Legitimacy theory and “new-style” judicial campaigns. American Political Science Review, 102(1), 59–75.Find this resource:

Gibson, J. L. (2008b). Campaigning for the bench: The corrosive effects of campaign speech? Law and Society Review, 42(4), 899–928.Find this resource:

Gibson, J. L. (2009). “New-style” judicial campaigns and the legitimacy of state high courts. Journal of Politics, 71(4), 1285–1304.Find this resource:

Gibson, J. L. (2011). Judging the politics of judging: Are politicians in robes inevitably illegitimate? In C. G. Geyh (Ed.), What’s law got to do with it? What judges do, why they do it, and what’s at stake (pp. 281–305). Stanford, CA: Stanford University Press.Find this resource:

Gibson, J. L. (2012). Electing judges: The surprising effects of campaigning on judicial legitimacy. Chicago: University of Chicago Press.Find this resource:

Gibson, J. L., & Caldeira, G. A. (2009a). Citizens, courts, and confirmations: Positivity theory and the judgments of the American people. Princeton, NJ: Princeton University Press.Find this resource:

Gibson, J. L., & Caldeira, G. A. (2009b). Confirmation politics and the legitimacy of the U.S. Supreme Court: Institutional loyalty, positivity bias, and the Alito nomination. American Journal of Political Science, 53(1), 139–155.Find this resource:

Gibson, J. L., & Caldeira, G. A. (2012). Campaign support, conflicts of interest, and judicial impartiality: Can the legitimacy of courts be rescued by recusals? Journal of Politics, 74(1), 18–34.Find this resource:

Gibson, J. L., & Caldeira, G. A. (2013). Judicial impartiality, campaign contributions, and recusals: Results from a national survey. Journal of Empirical Legal Studies, 10(1), 76–103.Find this resource:

Gibson, J. L., Gottfried, J. A., Delli Carpini, M. X., & Jamieson, K. H. (2011). The effects of judicial campaign activity on the legitimacy of courts: A survey-based experiment. Political Research Quarterly, 64(3), 545–558.Find this resource:

Gibson, J. L., Lodge, M., & Woodson, B. (2014). Losing, but accepting: Legitimacy, positivity theory, and the symbols of judicial authority. Law & Society Review, 48(4), 837–866.Find this resource:

Gibson, J. L., & Nelson, M. J. (2014). The legitimacy of the U.S. Supreme Court: Conventional wisdoms, and recent challenges thereto. Annual Review of Law and Social Science, 10, 201–219.Find this resource:

Gibson, J. L., & Nelson, M. J. (2015a). Is the U.S. Supreme Court’s legitimacy grounded in performance satisfaction and ideology? American Journal of Political Science, 59(1), 162–174.Find this resource:

Gibson, J. L., & Nelson, M. J. (2015b). Too liberal, too conservative, or about right? The implications of ideological dissatisfaction for Supreme Court legitimacy. Paper Presented at the 2015 Meeting of the American Political Science Association, San Francisco, CA.Find this resource:

Hall, M. G. (2001). State supreme courts in American democracy: Probing the myths of judicial reform. American Political Science Review, 95(2), 315–330.Find this resource:

Hall, M. G. (2014). Televised attacks and the incumbency advantage in state supreme courts. Journal of Law, Economics, and Organization, 30, 138–164.Find this resource:

Hall, M. G. (2015). Attacking judges: How campaign advertising influences state supreme court elections. Palo Alto, CA: Stanford University Press.Find this resource:

Hall, M. G., & Bonneau, C. W. (2013). Attack advertising, the White decision, and voter participation in state supreme court elections. Political Research Quarterly, 66(1), 115–126.Find this resource:

Iyengar, S. (2002). The effects of media-based campaigns on candidate and voter behavior: Implications for judicial elections. Indiana Law Review, 35, 691–699.Find this resource:

Jackson, R. A., Mondak, J. J., & Huckfeldt, R. (2009). Examining the possible corrosive impact of negative advertising on citizens’ attitudes toward politics. Political Research Quarterly, 62(1), 55–69.Find this resource:

Jacobson, G. C. (2013). The politics of congressional elections (8th ed.). Boston: Pearson.Find this resource:

Klein, D., & Baum, L. (2001). Ballot information and voting decisions in judicial elections. Political Research Quarterly, 54(4), 709–728.Find this resource:

Kritzer, H. M. (2015). Justices on the ballot: Continuity and change in state supreme court elections. New York: Cambridge University Press.Find this resource:

Krueger, K. (2015, July 8). Trump on retweet about Jeb’s wife: “No, I don’t regret it.” Talking points memo. CNN.Find this resource:

Krupnikov, Y. (2011). When does negativity demobilize? Tracing the conditional effect of negative campaigning on voter turnout. American Journal of Political Science, 55(4), 797–813.Find this resource:

Lau, R. R., & Pomper, G. M. (2004). Negative campaigning: An analysis of U.S. Senate elections. Lanham, MD: Rowman & Littlefield.Find this resource:

Lau, R. R., & Rovner, I. B. (2009). Negative campaigning. Annual Review of Political Science, 12, 285–306.Find this resource:

Lau, R. R., Sigelman, L., & Rovner, I. B. (2007). The effects of negative political campaigns: A meta-analytic reassessment. Journal of Politics, 69(4), 1176–1209.Find this resource:

Mattes, K., & Redlawsk, D. P. (2014). The positive case for negative campaigning. Chicago: University of Chicago Press.Find this resource:

Mutz, D. C., & Reeves, B. (2005). The new videomalaise: Effects of televised incivility on political trust. American Political Science Review, 99(1), 1–15.Find this resource:

Nelson, M. J. (2011). Uncontested and unaccountable? Rates of contestation in trial court elections. Judicature, 94(5), 208–217.Find this resource:

Nelson, M. J. (2014). Responsive justice? Public opinion and the criminal justice system. Journal of Law and Courts, 2(1), 117–152.Find this resource:

Nelson, M. J., Caufield, R. P., & Martin, A. D. (2013). OH, MI: On empirical examinations of judicial elections. State Politics & Policy Quarterly, 13(4), 495–511.Find this resource:

Noble, J. (2015, July 24). Trump barring Des Moines Register from campaign event. Des Moines Register.Find this resource:

Peters, C. S. (2009). Canons of ethics and accountability in state supreme court elections. State Politics & Policy Quarterly, 9(1), 24–55.Find this resource:

Public Policy Polling. (2015, September 22). Trump continues to lead in Iowa; Clinton in good shape [Press Release].

Ridout, T. N., & Fowler, E. F. (2012). Explaining perceptions of advertising tone. Political Research Quarterly, 65(1), 62–75.Find this resource:

Rock, E., & Baum, L. (2010). The impact of high-visibility contests for U.S. state court judgeships: Partisan voting in nonpartisan elections. State Politics & Policy Quarterly, 10(4), 368–396.Find this resource:

Sides, J., Lipsitz, K., & Grossmann, M. (2010). Do voters perceive negative campaigns as informative campaigns? American Politics Research, 38(3), 502–530.Find this resource:

Sigelman, L., & Kugler, M. (2003). Why is research on negative campaigning so inconclusive? Understanding citizens’ perceptions of negativity. Journal of Politics, 65(1), 142–160.Find this resource:

Solotaroff, P. (2015, September 9). Trump seriously: On the trail with the GOP’s tough guy. Rolling Stone.Find this resource:

Tate, C. N., & Vallinder, T. (Eds.). (1995). The global expansion of judicial power. New York: New York University Press.Find this resource:

Woodson, B. (2015). Politicization and the two modes of evaluating judicial decisions. Journal of Law and Courts, 3(2), 193–221.Find this resource:


(1.) We define “politicization” as legal institutions and actors taking on the attributes of ordinary politics and political institutions. See Tate and Vallinder (1995) for an early usage of the term in this context.

(2.) For a review of legitimacy theory and the U.S. Supreme Court, see Gibson and Nelson (2014).

(3.) For example, one ad urged: “The Right Wing has already taken over the West Wing. Don’t let them take over your Supreme Court.” (ISSUE/IC%20RIGHT%20WING%20TAKEOVER,, Creative id #4176684.)

(4.) Gibson and Nelson (2015b) have shown that perceived politicization is an important determinant of institutional support for the Supreme Court, having far more impact than factors such as ideological disagreements with the Court’s rulings (see also Woodson, 2015; Gibson & Nelson, 2015a). However, not all research agrees—see, for examples, Bartels and Johnston (2013) and Christenson and Glick (2015).

(5.) These video compilations can viewed at

(6.) We are mindful that the effects of ads when presented in an interview (in the lab or in the field) are undoubtedly greater than when they are viewed in nature (with screaming babies, boiling pots, and all of the other distractions of living). Gibson’s research documents this in Kentucky (2012, pp. 120–121). As Barabas and Jerit (2010, p. 238) note, “[a]lthough the real world does not look so different as to throw into doubt the validity of survey experiments, there is drop-off in terms of both the size of treatment effects and the population experiencing those effects” when comparing real-world treatment effects to experimental effects. This attenuation could happen for any number of reasons; to take one example, every respondent in the treatment group of a survey experiment is exposed to the exact same stimulus whereas, outside of the “laboratory,” the public differs widely in the extent to which it pays attention to (and therefore is exposed to) the effects that serve as experimental stimuli. In short, Barabas and Jerit’s (2010) research indicates that experiments capture the maximum possible treatment effect.

(7.) We acknowledge that several political scientists (e.g., Hall, 2001) have noted many similarities in the features of judicial and legislative campaigns, including contestation rates and rates of incumbent defeats. We recognize these similarities, and discuss cross-institutional comparisons in more detail below.

(8.) Gibson (2012, p. 110) makes this argument (see also Gibson, 2008b). These normative constraints are occasionally reinforced by legal restrictions, as in prohibitions about what a candidate can say about public policies and restrictions on fund-raising activities (e.g., Peters, 2009). Many campaign activities that would be expected from candidates for other public offices have been legally banned if done by candidates for judicial office.

(9.) Only other one other country in the world (Bolivia) uses judicial elections to select judges on a widespread basis, and concerns about campaigning harming the legitimacy of the judicial institutions in that country were so strong that candidates in the inaugural Bolivian judicial elections were barred from campaigning entirely, relying instead on a government-run information campaign (Driscoll & Nelson, 2012, 2015).

(10.) Though we do not discuss these threats in detail here, we note that elections may create other threats to the legitimacy of state supreme courts, such as those created when a campaign contributor appears before a judge (Gibson & Caldeira, 2012, 2013).

(11.) Despite the Supreme Court’s normative assumptions, empirical evidence demonstrates that elected judges do follow public opinion more closely than appointed judges, and the extent to which judges are congruent with public opinion depends on the type of election they face (Brace & Boyea, 2008; Canes-Wrone, Clark, & Kelly, 2014). Nelson (2014) demonstrates that elected judges also respond to signals of constituency-level public opinion, changing their sentences to fall into line with the expressed preferences of their constituents.

(12.) We use the term “state supreme court” to mean the court of last resort in a state, acknowledging that, in some states (e.g., New York) the supreme court is a lower court and in other states the highest court has a different name, for example, the Court of Appeals.

(13.) An interesting question arises as to what the dependent variables ought to be in investigations of the influence of campaigns. Hall (2015) considers both voting for the incumbent and voter demobilization (see also Hall, 2014). Gibson (2012) addresses judicial legitimacy. The value of legitimacy and mobilization seems clear; that incumbent judges need greater protection seems less clear.

(14.) Additionally, in a comprehensive study, Bonneau, Hall, and Streb (2011) examine the effects of White on other outcomes of interest, such as candidate entry, voter participation, and incumbent support. Their study demonstrates that, on these outcomes as well, White’s effects have been nonexistent.

(15.) We say “backfired,” but we actually have practically no information about the efficacy of various ads in terms of convincing voters to vote for the person running the ad. Campaign managers may have insights into this question; political scientists seem not to.

(16.) Such research into the short-term effects of negative advertisements on judicial legitimacy would fit well with the literature on the effects of timing of negative advertisements in presidential elections (Krupnikov, 2011).