Examining Murder Convictions and Punishment

Murder is the most serious criminal offense, and murder sentencing entails the most consequential punishment decisions in society. There has been long-standing concern about disparity in these most consequential decisions, most prominently, disparity based on the race/ethnicity and gender of defendants and victims. In addition, differences in case outcomes related to legal representation raise issues of adequate and fair representation for indigent defendants. Variation between locales in punishment outcomes are also central to a major recent direction of social science research; how discretion in the criminal justice system is shaped by larger socio-political environments. This proposed research presents a unique opportunity to delve deeper into these important theoretical and criminal justice policy concerns. This study's findings would be of interest across several disciplines, especially criminology/criminal justice studies, sociology, and legal studies. For cases charged with first, second, or third degree murder, this proposed project will examine: 1) patterns of disparity in charging, conviction, and punishment outcomes related to defendant and victim race/ethnicity, victim gender and age, and type of defense representation; 2) between-county differences in murder case outcomes and patterns of disparity in relation to court sociopolitical contexts; and 3) qualitative differences between counties in their formal and informal norms shaping murder case processing, focusing on charging and plea bargaining, prosecutorial decisions to seek the death penalty and to retract the death penalty if sought, and sentencing decisions.

The project will compile statistical data from court and prosecutor files on cases charged with first, second, and third-degree murder in 18 Pennsylvania counties that account for 79% of the criminal homicide charges from 2000-2010. These data will be used to examine individual-level differences in murder case processing outcomes, and between-county differences in murder case outcomes relative to court sociopolitical contextual features. Analyses will focus on the following dependent variables: charging decisions, mode of conviction (conviction or not, and conviction by guilty plea or trial), and sentence. For first degree murder charges, analyses will examine whether prosecutors seek the death penalty and whether they retract that filing, and whether defendants receive the death penalty. Three classes of statistical models will be estimated: 1) propensity score weighting and/or matching models to examine individual-level differences between groupings of defendants; 2) inverse propensity score weighted regression (IPSWR) models to examine between-county differences in individual-level outcomes on the dependent variables above; and 3) hierarchical logistic and linear models to examine the effects of county contextual variables on the dependent variables. In addition, open-ended, in-depth interview data will be collected to examine qualitative differences between counties in their formal and informal norms, policies, and practices shaping their murder charging, conviction processes (especially plea bargaining), prosecutorial decisions to seek the death penalty and to retract the death penalty if it is sought, and sentencing decisions. Forty to fifty individuals will be interviewed in the five counties with the most criminal homicide cases; interviews will include judges, prosecutors, public defenders, and private defense attorneys.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.