"Evidence exists of modest, short-term reductions in the numbers of homicides in Texas in the month of or after executions," the study published in a recent issue of Criminology, a journal of the American Society of Criminology, said.
The study adds to decades of academic dissection of the death penalty and deterrence. Results over the years vary from capital punishment saving more lives than suggested in this study to no conclusive effect.
This study, however, is the first to focus on monthly data in Texas, where researchers said the number of executions -- 447 since capital punishment resumed in 1982 -- is statistically significant enough "to make possible relatively stable estimates of the homicide response to executions." A national deterrent effect can't be determined because "most states ... have not engaged in a sufficient level or frequency of executions per year," they said.
Kent Scheidegger, legal director of the California-based Criminal Justice Legal Foundation, which supports capital punishment, said the study "would be sufficient by itself to justify the death penalty."
But Richard Dieter, executive director of the Death Penalty Information Center, a Washington, D.C.-based organization opposed to capital punishment, said while he was not a statistics expert, "the large number of variables affecting these calculations and the relative rarity of executions make final conclusions about deterrence very suspect."
The study analyzed data from January 1994 through December 2005, during which 284 lethal injections were carried out in Texas -- about one-third of all death sentences carried out in the U.S.
The year 1994 was selected as the starting point because state and federal legislation and court rulings beginning then led to "an orgy of executions in Texas," the researchers noted.
Of the years studied, four had more than 30 executions, including a record 40 carried out in 2000.
Researchers ran mathematical models that considered homicide figures from the Texas Department of Public Safety to see if month-to-month fluctuations in executions could be associated with subsequent month-to-month fluctuations in homicide counts.
Teske told The Associated Press while the published study ended with results through 2005, the conclusions are valid for subsequent years.
David McDowall, a professor at the State University of New York at Albany and an expert in statistical analysis of crime and violence patterns, said the study appeared solid and used standard accepted research methods.
"What the study does is try to control a constant variety of factors that vary over time by chance and then try to assess whether any decreases in homicides are large enough that chance can't account for them," McDowall said.
He said additional research examining homicides in nearby states where the death penalty is less active could add to the Texas study's credibility.
The researchers said they did exactly that and found the frequent use of executions in Texas had a greater cumulative impact on homicides in Texas when compared to homicide numbers in Louisiana, New Mexico and Oklahoma. They didn't include those findings in the final paper because reviewers wanted them to narrow its focus.
Teske acknowledged some experts disliked the results. He speculated criticism came from peer reviewers opposed to capital punishment.
"I have a hard time getting people to understand that this reports a scientific analysis of an issue and is not a political statement," Teske said.
Six Texas inmates are scheduled to die this year, including one Thursday and another next week.
