A new study conducted by Professor Richard Berk of the UCLA Department of Statistics has identified significant statistical problems with the data analysis used to support recent studies claiming to show that executions deter crime in the United States. In “New Claims about Executions and General Deterrence: Deja Vu All Over Again?,” Professor Berk addresses the problem of “influence,” which occurs when a very small and atypical fraction of the available data dominates the statistical results of a study. He found that this statistical problem is found in a number of recent studies claiming to show that capital punishment deters violent crime. The UCLA study conducted by Berk found that in many instances the number of executions by state and year is the key explanatory variable used by researchers, despite the fact that many states in most years execute no one and few states in particular years execute more than five individuals. These values represent about 1% of the available observations that could have been used by researchers to draw conclusions for earlier studies claiming to find that capital punishment is a deterrent. In Professor Berk’s study, a re-analysis of the existing data shows that claims of deterrence are a statistical artifact of this anomalous 1%. (Published on UCLA’s Web site, July 19, 2004). Read the study. See Deterrence and Resources.