Vera publishes results of study exploring prosecutorial decision making

Bruce Frederick Senior Research Fellow
Mar 15, 2013

Prosecutors enjoy broad discretion in making decisions that influence criminal case outcomes. They make pivotal decisions throughout the life of a case, from deciding what charges to file to crafting plea offers and recommending sentences. Yet there is little systematic oversight of the process by the public, the press, or the judiciary—prompting some observers to refer to the “black box” of prosecutorial decision making—and leaving open to question how consistently prosecutors’ decisions serve the goals of justice and fairness.
Prior studies have found case outcomes to be primarily associated with the strength of evidence, the seriousness of offenses, and the culpability of defendants. However, few studies have examined how and when these factors enter into the decision-making process, or how contextual factors such as policies, resource constraints, and relationships with other justice system actors influence prosecutors’ case-level decisions.
A unique feature of Vera's study, titled The Anatomy of Discretion: An Analysis of Prosecutorial Decision Making—supported by the National Institute of Justice—was the way it combined multiple quantitative and qualitative methods. For each decision point, analyses of administrative data for actual cases examined the influence of evidence, crime type, crime seriousness, criminal history, characteristics of defendants and victims, victim-offender relationships, and prosecutors’ level of experience. A general survey solicited prosecutors’ beliefs about the goals and functions of the criminal justice system, conceptions of justice and fairness, criteria for evaluating their success as prosecutors, the importance of relationships among system actors, resource and policy constraints, principles that guide their case processing strategies, and the adequacy of training and oversight. A factorial survey asked prosecutors what their screening and charging decisions, plea offers, and sentence recommendations would be for a standardized set of hypothetical cases that systematically varied offense seriousness, evidentiary strength, and criminal history. Analyses of the factorial survey responses also applied multi-level modeling techniques to incorporate prosecutors’ level of experience and measures of their attitudes and beliefs, constructed from their responses to the general survey. Finally, two waves of interviews and focus groups with selected prosecutors yielded additional information about contextual constraints that modify or override the influence of case-level factors, and  provided insights into how prosecutors balanced various case-level and contextual influences at each stage.
Prosecutors said they were committed to “doing the right thing,” and that two basic questions guided their decisions: “Can I prove the case?” and “Should I prove the case”? The study found that strength of the evidence was the primary consideration at screening, which focused mostly on the first question. In later decisions, where prosecutors were also considering the second question, strength of the evidence was balanced against other considerations—crime seriousness, criminal history, fairness to the victim and defendant, and contextual factors such as office and unit policies, resource constraints, and relationships with colleagues and other system actors.
There was considerable variation in the decisions reached by different prosecutors in similar cases. They differed in their opinions about how to define fair treatment, they differed with respect to the plea bargaining strategies they preferred, and different prosecutors may have reacted differently to contextual pressures. Thus, contextual factors could place constraints on the exercise of prosecutorial discretion yet produce systematic differences across offices, units, or individual prosecutors. Future research should incorporate objective measures of these pressures and begin to evaluate the locus, direction, and magnitude of their influence.