Vol. 26 No. 4 (August 2016) pp. 82 – 84

VOTERS’ VERDICTS: CITIZENS, CAMPAIGNS, AND INSTITUTIONS IN STATE SUPREME COURT ELECTIONS, by Chris W. Bonneau and Damon M. Cann. Charlottesville: University of Virginia Press, 2016. 163pp. Cloth $40.00 ISBN: 978-0813937595

Reviewed by Todd A. Curry, Department of Political Science, University of Texas, El Paso. Email:

Chris Bonneau and Damon Cann are familiar names for those individuals who are interested in the research on state courts, and specifically, judicial elections. They, like many of us in the field, have in part oriented their research agenda around a simple premise: when someone important or influential says something with absolutely zero empirical backing, design a study to examine if they are correct. In the sub-field of state judicial politics, with the myriad of ways states select their judges, there is no dearth of commentators, spectators and even participators who regularly make their opinion known as to the perceived deficiencies of the selection mechanism they like least. This provides good, if common, fodder upon which to build numerous research projects. However, what differentiates this book, and makes it exceedingly valuable, is that instead of relying on aggregated data which is exceedingly common when studying state courts, Bonneau and Cann examine familiar questions at the micro level, through individual data.

While aggregate level data have led us to numerous insights about state courts, there has been difficulty demonstrating causality because of the lack of micro level data. For example, researchers have demonstrated that ballot roll-off was on average, highest in states that use retention elections (when compared to partisan and non-partisan elections), but the lack of individual level data prevented them from claiming that it was retention elections that were causing ballot roll-off.1 That is, without individual level controls on individual level data, we could only examine the occurrence in the aggregate, unable to demonstrate that an institution which provides no information to the voter in the form of party identification and simultaneously denies the voter the choice of an alternative candidate systematically causes ballot roll-off.

As such, the true strength of this book is the wealth and variety of the data used to examine judicial elections. The authors use both the 2010 and 2012 Cooperative Congressional Election Study (CCES), experiments carried out with a diverse body of undergraduates, and experiments from Mechanical Turk (Mturk). These data allow the researchers to examine the individual causes of vote choice and ballot roll-off, and the effects of campaign cues, including incumbency, all under a rich level of institutional variation. It provides for a robust book, which may border on being a bit data heavy.

The book begins with an introductory chapter that describes the methods of judicial selections used within the states, with a primary focus on judicial elections. Three types of election mechanisms are used to staff high courts in the states. Partisan elections are identical to congressional elections, with candidates being chosen in a party primary to then face off during the general election. Non-partisan elections share many similarities with partisan elections, however, they lack a significant cue to voters, namely the appearance of partisan identification on the ballot. Retention elections are different. Retention elections are defined by a lack of contestation. Candidates run in what amounts to a referendum on their judgeships, with the only options available to voters are a yes or no vote to retain.

Chapter one focuses on ballot roll-off in contested election. Bonneau and Cann add a third group, beyond partisan and non-partisan elections, which they term quasi-partisan. This coding corresponds to Michigan and Ohio, which use a combination partisan and non-partisan process to elect their judges. In the general election, candidates do not appear with party identification on the ballot, but the candidates were chosen in a partisan fashion in either a primary (Ohio), or a party convention (Michigan). In a two-stage model, the authors demonstrate that individuals are more likely to roll-off the ballot when their judges are elected in non-partisan elections, even when controlling for other individual factors that may increase the likelihood of ballot roll-off. The authors interpret this to mean that when voters must make an off the cuff decision, in the absence of party identification, some voters choose to abstain.

This could be a desired effect of non-partisan elections. Removing party identification from the ballot could lead to less ideological voting. This assumption is put to test in chapters two and three. In chapter two, the authors demonstrate through CCES data that ideological voting still is highly likely to take place in non-partisan elections, though not at as high a level as in partisan elections. Chapter three demonstrates why this is likely the case. In a series of experiments, Bonneau and Cann test whether individuals can pick up ideological clues from short vignettes and correctly assign them to hypothetical candidates. Even in the absence of party identification, when ideological language is used by the candidates, they can be identified as partisans. In the era following REPUBLICAN PARTY OF MINNESOTA V. WHITE (2002), candidates are more likely to use ideological language in their campaigns. As such, the authors conclude, non-partisan elections do not reduce the level of ideological voting.

Continuing in the same vein of examining the relevance of ballot information, the authors examine two different possible effects of incumbency. First, when incumbency is listed on the ballot, does it have an appreciable effect on vote choice? Second, does the ability to campaign as an incumbent have a significant effect on incumbent vote share? Shockingly, unlike the wealth of information on the power of incumbency in congressional elections, there appears to be no affect for incumbency in either circumstance. As the authors point out, there are clearly structural advantages to being an incumbent, including name recognition and ease of raising campaign funds, however when these are controlled for, incumbency itself does not give a campaign boost.

In the last empirical chapter, Bonneau and Cann introduce the uncontested retention elections into the analysis. While these may seem like retention elections are left as an afterthought, they are not. There exist some important structural differences that need to be taken into account when attempting to compare contested elections to uncontested retention elections, and the authors in this chapter, handle these difficulties well.

Retention elections, and those who claim their benefits, have not fared well in previous empirical analyses. This chapter continues that notable trend. The two phenomenon examined are ballot roll-off and vote choice. Similar to the model in chapter one, the authors augment their model by including retention elections and find that individuals voting in retention elections are 25% more likely to engage in ballot roll-off than individuals voting in partisan elections, a finding that echoes much of the work done at the aggregate level.

Personally the most interesting part of the entire book, and also the part that seemed the most original, was the authors’ examination of vote choice in retention elections. At the aggregate level we have long known that judges hardly ever get removed from office in retention elections, minus some notable exceptions. These authors bring individual level data to bare to discover why that is the case. They demonstrate that while Democrats are less likely to vote to retain Republican judges, and Republicans are less like to retain Democrat judges, the lowest predicted probability for a “yes” vote is 50%. This means that Republicans vote to retain Democrat judges 50% of the time, and the probabilities of a “yes” vote only increase from there. Furthermore, anti-retention campaigns have no significant effect on individual vote choice. The authors conclude that while judges in retention elections have a small form of accountability, they are given a significantly longer electoral leash than contested judicial elections

While this book contributes significantly to the fields of judicial politics and electoral politics, it lacks in one prominent area: theoretical contribution. Much of the theoretical explanation or justification seems like forgone conclusions to those versed in similar literature. This is not to say the theory is ad hoc, it certainly is not, however, this book does not advance our theoretical understanding of these phenomenon. The theory used in this is book is heavily borrowed, and less well explained, than in similar books.

The major contribution that this book gives to the field is in the form of individual level data and analysis. It provides insight into how individuals make decisions and behave in these judicial elections. However, the theoretical backing is not very thorough, and heavily borrowed from other sources. For a graduate course, I would assign this book after Bonneau and Hall’s (2009) IN DEFENSE OF JUDICIAL ELECTIONS because I feel it provides a more robust theoretical review and historical context. VOTERS’ VERDICTS, however, deserves to be assigned alongside IN DEFENSE, because it is the first large scale analysis of individual level data concerning judicial elections.

1 Ballot roll-off occurs when individuals go to the polls and vote at the top of the ticket, but do not vote for positions lower down.


Bonneau, Chris W. & Hall, Melinda G. 2009. IN DEFENSE OF JUDICIAL ELECTIONS. New
York: Routledge.



© Copyright 2016 by author, Todd A. Curry.