NCJ Number
129787
Journal
Journal of the American Statistical Association Volume: 85 Issue: 410 Dated: (June 1990) Pages: 295-303
Date Published
1990
Length
9 pages
Annotation
Information from computerized death certificates from California during 1960-63 formed the basis of an analysis of whether the occurrence of executions has a short-term deterrent effect on homicides.
Abstract
The study period was chosen because it was the last period during which frequent executions took place in the State. The research focused on the homicides over a 6-week period spanning the day on which an execution took place, with analyses of the numbers of homicides during 2-week to 4-week periods immediately surrounding the execution. It also considered the race and sex of victims, to determine race-specific or gender-specific effects that would be too small to be determined from aggregate data. Poisson and compound-Poisson regression models were estimated to control for seasonal fluctuations and to measure the effect of executions on daily homicide counts. The results consistently indicated no support for the hypothesis that executions deter murder in the short term. Tables, figures, and 29 references