• Callander, Steven, and Tom S. Clark, "Precedent and Doctrine in a Complicated World," American Political Science Review 111(1):184-203
    Courts resolve individual disputes under the law and create principles of law to justify their decisions and guide the resolution of future disputes. Those tasks present informational challenges that affect the whole judicial process. Judges must simultaneously learn about (1) the particular facts and legal implications of any dispute and (2) the doctrine that appropriately resolves the dispute. Judges must attempt to articulate those rules in the context of a single case so that future courts may reason from past cases. We propose a model of judicial learning and rule-writing in which there is a complicated relationship between facts and legal outcomes and judges are bound to resolve cases by analogizing from past precedents to new situations. The model has implications for many of the important questions in the judicial process, including the dynamics of common law development, the path-dependent nature of the law, and optimal case selection for rule-making.

  • Lauderdale, Benjamin, and Tom S. Clark, "Estimating Vote-Specific Preferences from Roll-Call Data using Conditional Autoregressive Priors," Journal of Politics 78(4):1153-1169 (2016)
    Recent research has demonstrated that the preferences of US Supreme Court justices are not, and have never been, simply unidimensional.  At the same time, in advance of decisions on high profile cases, there has been increasing speculation about which justices are likely to be pivotal.  In this paper, we demonstrate a new approach to Bayesian preference estimation that estimates case-specific preferences for justices, using a conditional autoregressive model with citation counts determining the correlation between justices' preferences across cases.  By using citations in case briefs to identify the most relevant precedent cases, we are able to describe variation in revealed preferences across areas of the law, as well as to predict in advance of a decision how justices will align on a new case, based on their past behavior in similar cases.

  • Clark, Tom S., "Scope and Precedent: Judicial Rule-Making Under Uncertainty," Journal of Theoretical Politics 28(3):353-384 (2015)
    This paper develops a formal model of Supreme Court opinion-writing in an environment of uncertainty. In particular, the model is designed to capture how the Supreme Court will optimally design the specificity of its legal rules, balancing the tradeoff between more precise rules which are controlling in a smaller subset of cases against less precise rules, which have wider applicability but less predictive power. When the basic model is considered in a dynamic world in which the Court is able to hear multiple cases, it yields insights about how the factual representativeness of a case affects its optimal opinion-writing and willingness to hear new cases. These last implications provide theoretical foundations for case selection at the Supreme Court.
  • Clark, Tom S., and Jeffrey K. Staton, "An Informational Model of Judicial Docket Styles," Journal of Politics 77(3):589-607 (2015)
    In shaping constitutional courts' jurisdictions, societies must contend with the informational challenges associated with rule-making and the distributive politics of granting courts jurisdiction over administrative law-making. Curiously, judges are often granted jurisdiction that seems to create a tension between their ability to acquire information about appropriate rules and to clearly articulate them. In particular, many courts handle the onerous burden of resolving thousands of routine, low-stakes cases of law-application. We develop an informational model of judicial docket style that isolates a tension between information acquisition and quality rule-writing and examine how that tension manifests in the incentives concerning jurisdiction style. Dockets that include a mix of law application and rule construction promote more informed judicial rule construction at the cost of lower quality rules and a greater role of the judiciary in the day to day activity of the state. We develop implications for constitutional design in a liberal democracy. 
  • Clark, Tom S., and Drew A. Linzer, "Should I Use Fixed or Random Effects?" Political Science Research & Methods 3(2):399-408 (2015)
    Empirical analyses in political science very commonly confront data that are grouped---multiple votes made by individual legislators, multiple elections in individual states, multiple conflicts during individual years, etc. Modeling these data presents a series of potential challenges, of which accounting for differences across the groups is perhaps the most well-known. The most widely-used approach to this challenge is the use of either fixed or random effects. However, how best to choose which approach remains unclear in the applied literature. We employ a series of simulations to evaluate the commonly-used Hausman test and demonstrate that it is neither a necessary nor sufficient statistic for deciding between fixed and random effects. We use our simulations to evaluate the relative fit of fixed and random effects estimators under varying types of datasets. We summarize the results into a typology of datasets to offer guidance to the applied empirical modeler.
  • Clark, Tom S., and Jonathan P. Kastellec, "Source Cue and Public Support for the Supreme Court," American Politics Research 43(3):504-535 (2015)
    It is well known that the public often relies on cues or heuristics from elites when forming opinions. Yet, little is known about whether public opinion on the Supreme Court—which has been traditionally thought to be relatively fixed—is also dependent on elite cues. Using a series of survey experiments, we find that elite cues (in the form of partisan source cues) significantly influence the public’s support for judicial independence. We also find that the mere exposure to such cues—that is, the politicization of the Court—can affect the support for specific decisions made by the Court. These results have important implications for understanding the extent to which politicians can shape the public’s overall support for judicial independence, as well as for assessing the degree to which the public views the Court as a “political” institution.
  • Clark, Tom S., Jeffrey R. Lax, and Douglas R. Rice, "Measuring the Salience of Supreme Court Cases," Journal of Law & Courts 3(1):37-65 (2015)
    Supreme Court cases are generally considered to be atypical cases, of greater importance or salience, though some are more important, to the justices and other actors, than others.  A wide range of theoretical and empirical work in judicial politics invokes this varying salience as a factor that can explain some types of judicial behavior. Other work considers salience a variable to be explained, with judicial behavior the explanatory factor. The currently dominant measure of salience is the coverage of decisions by the \emph{New York Times}. Because coverage occurs after a decision is announced, there are limits to causal inference when conflating both the treatment effect of salience on Supreme Court decisions and the treatment effect of Supreme Court decisions on salience. We develop a measure of latent salience that relies on media coverage but has the potential to ameliorate concerns of causal inference and put research findings on sounder footing.
  • Canes-Wrone, Brandice, Tom S. Clark, and Jason P. Kelly, "Judicial Selection and Death Penalty Decisions," American Political Science Review 108(1):23-29 (2014)
    Judges face some form of election or reappointment in most U.S. state court systems. In recent decades judicial campaigns have come to resemble those for other offices. We develop predictions on how the selection system should affect judicial decisions given this development, and test these predictions on an extensive dataset of death penalty decisions by state courts of last resort. The data include over 12,000 decisions on 2000 capital punishment cases decided between 1980 and 2006 in judicial systems with partisan, nonpartisan, or retention elections or with reappointment. As predicted, the results suggest that judges face the greatest pressure to uphold capital sentences in systems with nonpartisan or retention elections. Also as predicted, judges respond similarly to public opinion in partisan election and reappointment systems. Finally, the judicial decisions reflect plebiscitary pressures only after interest groups have achieved success at unseating justices for their decisions.
  • Lauderdale, Benjamin E., and Tom S. Clark, "Scaling Politically Meaningful Dimensions Using Texts and Votes," American Journal of Political Science 58(3):754-771 (2014)
    Item response theory (IRT) models for estimating spatial preferences using roll-call voting data have provided political scientists with parsimonious descriptions of what political actors relative preferences.  However, models using only voting data tend to obscure variation in preferences across different issues due to identification and labeling problems that arise in multidimensional models.  Latent Dirichlet Allocation (LDA) is an increasingly widely applied tool for using word counts to estimate the relative degree to which each text in a corpus discusses a set of issues.  However, while LDA is powerful for discovering which issues are being discussed in which texts, models based on word counts have proven less useful for discovering what texts are saying about those issues. With an application to the U.S. Supreme Court, we combine these two models into a new tool for discovering preference variation across issues, using voting data augmented with texts describing each vote.
  • Clark, Tom S., and Jonathan P. Kastellec, "The Supreme Court and Percolation in the Lower Courts: An Optimal Stopping Model," Journal of Politics 75(1):150-168 (2013)
    One of the few stated criteria by which the Supreme Court decides to hear a case is the existence of a conflict among the regional judicial circuits. However, a circuit split does not automatically lead the Supreme Court to review a case, and the justices have often allowed circuit splits - and thereby the application of different legal standards across the country - to stand for long periods of time. What explains the Court's tolerance of circuit conflict, and its eventual decision to resolve a circuit split? We model the Supreme Court's decision to intervene in an inter-circuit conflict as an optimal stopping problem. The Court faces a strategic trade-off between allowing conflict to continue while learning about the implications of the possible policy choices and intervening to end a costly conflict between the circuits. Our model provides the first theoretical framework for understanding when and how the Court decides to resolve inter-circuit conflict.
  • Lauderdale, Benjamin E., and Tom S. Clark, "The Supreme Court's Many Median Justices," American Political Science Review 106(4):847-866 (2012)
    While unidimensional preference estimates for the U.S. Supreme Court exist in both constant and time-varying forms, estimating variation in preferences across areas of the law has been difficult because multidimensional scaling models perform poorly in the context of a nine-member court. We introduce a new approach to recovering estimates of judicial preferences that are localized to particular legal issues as well as periods of time. Using multiple indices of substantive similarity among cases, we apply a kernel-weighted optimal classification estimator to analyze how justices’ preference vary across both areas of the law and time. We find that incorporating variation in substantive area significantly improves the predictive power of estimated preference orderings over a model that only allows for variation in preferences over time. We also find substantial variation in the identity of the median justice across areas of the law during most periods of the modern court, suggesting a need to reconsider empirical and theoretical research that hinges on the existence of a unitary and well-identified median justice.
  • Carrubba, Clifford J., and Tom S. Clark, "Rule Creation in a Political Hierarchy," American Political Science Review 106(3):622-643 (2012)
    A considerable portion of the law is judge-made. Moreover, most law creation takes place in the judicial hierarchy, with the Supreme Court serving primarily as the overseer of the decisions made by lower courts. Despite the significance of the judicial hierarchy for law creation, there remain considerable gaps in our knowledge about the incentives and constraints facing lower court judges responsible for law creation. We model law creation in a judicial hierarchy characterized by principal-agent dilemmas. These models incorporate the empirical observations of previous research and outline a theoretical framework and testable predictions about the choices lower court judges will make when creating law. While the modeling exercise is focused on courts, the implications of the theoretical analyses extend to other hierarchical rule-making structures.
  • Clark, Tom S., and Benjamin E. Lauderdale, "The Genealogy of Law," Political Analysis 20(3):329-350 (2012)
    We develop a method for systematically connecting Supreme Court cases to each other to identify the genealogy of Supreme Court doctrine. We apply an original estimator to original data on the total number of citations from one case to another. Our model yields a "family tree" representation of cases which provides a systematic method for studying numerous quantities of interest to scholars of judicial politics. The model has a number of important benefits that improve upon existing network methods. First, we construct networks of doctrine that incorporate the theoretical structure of judge-made law. Second, we use the full ensemble of citations to locate opinions within a parsimonious description of existing precedents. Third, our model allows for the construction of measures of legal significance and other relevant quantities based on axiomatic definitions of those concepts. The model we develop lends itself easily to future extensions and avenues for research among scholars of judicial politics and doctrine.
  • Clark, Tom S., and Clifford J. Carrubba, "A Theory of Opinion Writing in a Judicial Hierarchy," Journal of Politics 74(2):584-603 (2012)
    We develop a model of opinion-writing in the judicial hierarchy. The model adopts a case-space approach to judicial decision-making and exploits informational asymmetries across levels of the hierarchy. In the model, a lower court writes an opinion with two features: a legal rule and a level of quality. An upper court must then decide whether to review the decision. The model yields new insights about the strategic incentives created by the judicial hierarchy. Empirical predictions are derived which demonstrate that several common tests employed in the literature may not discriminate among competing sets of first principles. Additional implications for rule-making in a bureaucratic hierarchy are also considered.
  • Canes-Wrone, Brandice, and Tom S. Clark, and Jee-Kwang Park, "Judicial Independence and Retention Elections," Journal of Law, Economics & Organization 28(2):211-234 (2012)
    Judges face retention elections in over a third of U.S. state courts of last resort and numerous lower courts. According to conventional wisdom, these elections engender judicial independence and decrease democratic accountability. We argue that in the context of modern judicial campaigns, retention elections create pressure for judges to cater to public opinion on “hot-button” issues that are salient to voters. Moreover, this pressure can be as great as that in contestable elections. We test these arguments by comparing decisions across systems with retention, partisan, and nonpartisan contestable elections. Employing models that account for judge- and state-specific effects, we analyze new data regarding abortion cases decided by state supreme courts between 1980 and 2006. The results provide strong evidence for the arguments.
  • Clark, Tom S., and Benjamin E. Lauderdale, "Locating Supreme Court Opinions in 'Doctrine Space,'" American Journal of Political Science 54(4)871-890 (2010)
    The development of powerful theoretical models of intra-court bargaining and judicial hierarchy has created a need to measure the doctrinal content of Supreme Court opinions. Such theories yield empirically-testable predictions, but the analysis of these predictions has been limited by the ability of scholars to measure judicial policy. This paper develops an original scaling model to estimate opinion locations and justice ideal points along a common, continuous dimension using the citations between opinions as data. We assume that each opinion has a fixed location in this unidimensional doctrine space and that the probability of a citation that affirms rather than disputes the doctrine of the precedent decreases as the doctrinal distance between them increases. This proximity citation model is applied to original data based on freedom of religion opinions written by the Warren, Burger and Rehnquist Courts. We use the resulting estimates of opinion content to evaluate median and non-median voter theories of Supreme Court bargaining and opinion writing. We find striking empirical support for theoretical models that predict the majority opinion will fall at the ideal point of the median member of the majority coalition as opposed to the court's median or the opinion author's ideal point.
  • Clark, Tom S., and Aaron B. Strauss, "The Implications of High Court Docket Control for Resource Allocation an Legal Efficiency," Journal of Theoretical Politics 22(2):247-268 (2010)
    A key source of institutional variation across judicial systems is the degree of control that the highest court has over its docket. Despite this variation, the consequences of various institutional designs in judicial hierarchies remain relatively unexplored by the theoretical literature. In this paper, we develop a formal model of high court resource allocation. We analyze the model under two institutional designs: (1) the Court must allocate at least some minimum effort to all cases; (2) the Court has complete discretion over which cases to hear. We analyze the model to identify the optimal allocation of resources across cases as the institutional design varies. We then consider the conditions under which the various institutional rules increase or decrease the legal efficiency, or performance, of the judicial system. Our analysis reveals the complex relationship among the institutional rules governing high court dockets, the design of the judicial hierarchy, and the performance of the legal system. The model informs substantively policy debates among judicial reformers and scholars concerned with institutional design as well as disciplinary debates about case selection and judicial decision-making.
  • Clark, Tom S., "The Separation of Powers, Court-curbing, and Judicial Legitimacy," American Journal of Political Science 53(4):971-989 (2009)
    Various literatures indicate that partisan labels increase the accountability of elected officials.  Correspondingly, advocates of nonpartisan elections claim that this procedure helps liberate officials from political influence. These arguments have been prominent in recent debates regarding the selection of judges in U.S. state courts. We argue that on salient issues judges in nonpartisan systems actually have greater incentives than ones in partisan systems to issue popular decisions, particularly given developments in judicial campaigns. To test this hypothesis we have assembled a new dataset that revolves around state supreme courts' decisions on abortion cases between 1980 and 2006. The analysis--which controls for a variety of factors and uses the U.S. Senate as a comparative tool--provides strong support for the hypothesis. An earlier version of this paper was presented at the Annual Meeting of the American Political Science Association, Philadelphia, PA, September, 2006.
  • Caldarone, Richard P., Brandice Canes-Wrone, and Tom S. Clark, "Partisan Labels and Democratic Accountability: An Analysis of State Supreme Court Abortion Decisions," Journal of Politics 71(2):560-573 (2009)
    Various literatures indicate that partisan labels increase the accountability of elected officials.  Correspondingly, advocates of nonpartisan elections claim that this procedure helps liberate officials from political influence. These arguments have been prominent in recent debates regarding the selection of judges in U.S. state courts. We argue that on salient issues judges in nonpartisan systems actually have greater incentives than ones in partisan systems to issue popular decisions, particularly given developments in judicial campaigns. To test this hypothesis we have assembled a new dataset that revolves around state supreme courts' decisions on abortion cases between 1980 and 2006. The analysis--which controls for a variety of factors and uses the U.S. Senate as a comparative tool--provides strong support for the hypothesis. An earlier version of this paper was presented at the Annual Meeting of the American Political Science Association, Philadelphia, PA, September, 2006.
  • Clark, Tom S., "A Principal-Agent Theory of En Banc Review," Journal of Law, Economics & Organization 25(1):55-79 (2009)
    This paper adds to the existing literature on en banc rehearings in two ways. First, I incorporate theoretical results from the literature on Supreme Court certiorari decisions and argue that the ideological direction of panel decisions should influence the probability of en banc rehearing only in conjunction with the panel's ideological predisposition. Second, I build upon existing theories of en banc review by incorporating the multiple levels of the judicial hierarchy into the context in which the circuit decides to hear a case en banc. From these insights, I develop and test three hypotheses about the determinants of en banc review. Specifically, I contend that the ideological relationship between a three-judge panel, the full circuit, and the Supreme Court should all interact with the ideological orientation of the panel's decision when the circuit decides whether or not to review the panel en banc. Original data including all en banc rehearings between 1986 and 1996 are then used to test the theoretical predictions. The empirical analysis provides considerable support for the hypotheses. The findings represent two important advances in the study of the judicial hierarchy: They highlight the strategic interaction of ideological composition in the en banc review process and demonstrate the incentives created by the multiple levels of the federal judiciary. More broadly, the theory and findings developed here have implications for strategic auditing in a political hierarchy. This project grows out of a seminar paper I co-authored with Justin Crowe and David Glick. An earlier version of this paper was presented at the Annual Meeting of the Southern Political Science Association, January, 2006.
  • Clark, Tom S., "Measuring Ideological Polarization on the U.S. Supreme Court," Political Research Quarterly 62(1):146-157 (2009)
    The study of ideological polarization is an important topic in research ranging from individual behavioral-level analyses to institutional studies of Congress and the executive branch. Polarization, however, has received little attention in the context of the Supreme Court, even while popular press and legal commentary suggest ideological heterogeneity on the Court are consequential for the Court's policy outputs. In this paper, I apply a method for measuring polarization developed by Esteban and Ray (1994) to ideological heterogeneity on the Court to develop a "polarization statistic." I compare this method with other common polarization measures. I then use the various measures of polarization to demonstrate that more polarized Courts produce more dissenting opinions. I argue that the study of polarization and its implications for judicial outputs is an important area of study that should be further investigated by political scientists.
  • Canes-Wrone, Brandice, and Tom S. Clark, "Judicial Independence and Nonpartisan Elections," Wisconsin Law Review 2009(1):21-65 (2009)
    This article develops a theoretical argument and, using social scientific methods, presents empirical evidence to challenge an implicit assumption that permeates both legal and political science scholarship on the consequences of partisan and nonpartisan judicial elections. In part because progressive judicial reformers argued that nonpartisan elections would serve to remove political influence which characterized partisan judicial elections of the early-twentieth century from the judicial selection process, students of judicial elections have often implicitly assumed that nonpartisan elections lead to greater judicial independence from political considerations. We argue that in the context of modern judicial elections where interest groups advertise judges positions and where judges themselves often take positions on issues, nonpartisan elections can give rise to greater political pressure on judges than partisan elections.  Partisan judges have a party label on which to rely as a signal to voters about their policy preferences. Nonpartisan judges, by contrast, communicate their policy positions through, among other means, their decisions on the bench. In this context, nonpartisan judicial candidates will be particularly sensitive to making decisions that run against public opinion. This pressure is likely to be particularly strong on the sorts of hot button issues that attract attention from interest groups and voters.

    To test this prediction, we examine the votes of sitting judges at the highest court in states with partisan and nonpartisan statewide judicial elections. Using social science methods that are standard in political science, we examine patterns in decision-making by judges in cases involving abortion between 1980 and 2006. Analyzing 597 votes in 85 cases across 16 states, we demonstrate, counter-intuitively, that judges in states with nonpartisan elections seem to be influenced by public opinion about abortion in their states, while there does not seem to be any such relationship between public opinion and judicial decision-making in states with partisan elections. This finding turns on its head much of the received wisdom about the consequences of nonpartisan judicial elections for judicial independence. Consequently, this article advances an important and on-going debate in both the legal and political science literature about judicial selection.

  • Clark, Tom S., "Judicial Decision-making During Wartime," Journal of Empirical Legal Studies 3(3):397-419 (2006)
    The separation-of-powers becomes increasingly stressed during wartime, as power is traditionally accumulated by and consolidated in the executive. This paper asks to what degree the separation-of-powers collapses by examining judicial deference to the executive during wartime. By analyzing a set of cases in the Courts of Appeals from a 100 year time period, this paper demonstrates that while judicial preferences undergo a fundamental shift with respect to criminal cases, there is no evidence of heightened deference to the executive during wartime. These findings suggest that a state of war has a preference-altering effect on judicial treatment of criminal defendants. They further suggest that concerns about judicial deference to the executive during times of war may not be as serious as conventional wisdom suggests.