The Source for Crime and Justice Data

Re-examination of the Criminal Deterrent Effects of Capital Punishment in the United States, 1978-1998 (ICPSR 20040) RSS

Principal Investigator(s):

Summary:

The purpose of this study was to estimate the deterrent effect of capital punishment by employing a methodology that accounted for model uncertainty by integrating various studies into a single coherent analysis. First, this study replicated the results from two previous studies, Dezhbakhsh, Rubin and Shepherd (2003) and Donohue and Wolfers (2005), that draw on the same data. Second, the researchers implemented model averaging methods using standard frequentist estimators to take a weighted average of the findings across all possible models that could explain the effect of the difference in crime rates under alternate laws. Each model's effect was weighted based on its ability to explain the data. Variables used in this study included deterrence variables as well as various demographic and economic control variables.

Access Notes

  • These data are freely available.

Dataset(s)

Dataset - Download All Files (0.8 MB)
Documentation:
Data:

Study Description

Citation

Cohen-Cole, Ethan, Steven Durlauf, Jeffrey Fagan, and Daniel Nagin. Re-examination of the Criminal Deterrent Effects of Capital Punishment in the United States, 1978-1998. ICPSR20040-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2008-01-31. http://doi.org/10.3886/ICPSR20040.v1

Persistent URL:

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote XML (EndNote X4.0.1 or higher)

Funding

This study was funded by:

  • United States Department of Justice. Office of Justice Programs. National Institute of Justice (2005-IJ-CX-0020)

Scope of Study

Subject Terms:   capital punishment, crime prevention, deterrence, homicide, murder, policy analysis

Smallest Geographic Unit:   county

Geographic Coverage:   United States

Time Period:  

  • 1978--1998

Unit of Observation:   county

Universe:   All persons sentenced to death within the United States between 1978 and 1998.

Data Types:   program source code

Data Collection Notes:

(1) Users can obtain data used for this study in Stata format from Justin Wolfers' Death Penalty Data Web site. From this Web site, users may either download Donohue and Wolfers' (2005) county-level homicide and execution data plus controls, 1977-1996 Stata 9 data (CPcounty5_send.dta), or Dezbakhsh, Rubin and Shepherd's (2003) county-level data 1977-1996 (DRS.zip). Both downloads contain the exact same Stata dataset, however Dezbakhsh, Rubin and Shepherd's (2003) county-level data 1977-1996 download, when unzipped, also contains additional Stata Do-files. (2) While this project's final report and Justin Wolfers' Death Penalty Data Web site refer to the data as covering the years 1977-1996, the data that are available from Wolfers' Web site actually cover the years 1978-1998. (3) The files in this collection are provided in a WinZip archive with 16 computer program code files for use with the MATLAB software program. Additional information on MATLAB is available from The MathWorks, Inc.. (4) Users are encouraged to refer to the final report, available from the National Criminal Justice Reference Service (NCJRS) for more detailed information regarding the study design and model averaging methodology and for complete references to publications mentioned in this description.

Methodology

Study Purpose:   The fundamental problem that underlies the disparate findings on deterrence effects of death sentencing is that individual studies reflect specific assumptions about the appropriate data, control variables, model specification, etc., on the part of the researcher. These assumptions can have major effects on the conclusions of a particular data analysis. The existing research on this topic comes to sufficiently differing conclusions, predicated upon one or more underlying assumptions, to call into question the ability of any single model to explain the impact of execution laws. Such dependence on the specifics of research design, from data cleaning to aggregation to model choice, forms the basis for the use of averaging techniques. That is, since relatively minor variations in model or variable choice can lead to dramatic changes in conclusions, one suspects that inclusion of the information content of all of these models would lend itself to conclusions upon which policymakers could be more confident. The purpose of this study was to estimate the deterrent effect of capital punishment by employing a methodology that accounted for model uncertainty by integrating various studies into a single coherent analysis. The research project sought to utilize model averaging, a method that enables a researcher to take a weighted average of findings across possible models, to develop inferences that are not dependent on the assumption that one of the models is true.

Study Design:   This study replicated the results from two previous studies and implemented model averaging methods to analyze deterrence laws. First, this study replicated the results from two previous studies, Dezhbakhsh, Rubin and Shepherd (2003) and Donohue and Wolfers (2005), that draw on the same data. Both papers estimated some version of a deterrence regression drawn from Ehrlich (1977). That is, the murder rate is a function of three principal deterrence variables: the probability of arrest, the probability of receiving a death sentence conditional of being arrested, and the probability of being executed conditional on receiving a death sentence. Second, the researchers implemented model averaging methods using standard frequentist estimators to take a weighted average of the findings across all possible models that could explain the effect of the difference in crime rates under alternate laws. Each model's effect was weighted based on its ability to explain the data. This frequentist approach to model averaging is described in Sala-i-Martin, Doppelhofer, and Miller (2004) and Brock, Durlauf, and West (2003).

Sample:   The sample for this study included county-level data for the post-moratorium period (1978-1998).

Weight:   The researchers used theory to estimate the weighted average coefficients in a variety of cases as part of their model averaging analysis approach.

Mode of Data Collection:   record abstracts

Data Source:

Data for this study were obtained from Donohue and Wolfers (2005) and Dezhbakhsh, Rubin and Shepherd (2003). The original sources for these data were the Department of Justice's Bureau of Justice Statistics, the FBI's Uniform Crime Reports, and the Bureau of the Census. Users should reference the COLLECT.NOTE section in this description for further information regarding data used in this study.

Description of Variables:   Variables used in this study included deterrence variables as well as various demographic and economic control variables. Deterrence variables included the probability of arrest, the probability of receiving a death sentence conditional of being arrested, and the probability of being executed conditional on receiving a death sentence. Demographic and economic control variables included controls for the aggravated assault rate, the robbery rate, the population proportion of 10-19-year-olds, 20-29-year-olds, demographic percentages of Blacks, percentage of non-Black minorities, percentage of males, the percentage of NRA members, real per capita income, real per capita income maintenance payments, real per capita unemployment insurance payments, and the population density. Other variables were voting variables, year variables, and arrest rate variables.

Response Rates:   Not applicable.

Presence of Common Scales:   None

Version(s)

Original ICPSR Release:  

Related Publications ?

Utilities

Metadata Exports

If you're looking for collection-level metadata rather than an individual metadata record, please visit our Metadata Records page.

Download Statistics