EMBARGOED FOR RELEASE Until 11:00 a.m. July 27, 1998
CONTACT: Michael Buckley; 301/405-8426; [email protected]
15 Proven Ways to Prevent Crime--and 23 Ways Not To-- Identified by University of Maryland
COLLEGE PARK, MD -- A Congressionally-mandated study has found scientific evidence that 15 different methods of crime prevention are effective. The study, conducted by University of Maryland criminologists for the National Institute of Justice, also found 23 programs that had been proven ineffective at preventing crime, and 30 programs for which the evidence is "promising."
The new study, Research in Brief: Preventing Crime: What Works, What Doesn't, What's Promising, released today at a Justice Department conference, is based on a review of more than 500 scientific evaluations of programs intended to prevent crime, from prenatal care to job training for older offenders. Building on a 565-page report submitted to Congress last year by the same team of researchers, the study presents the first summary lists of what works, what doesn't, and what's promising.
The list of what works includes: * nurses visiting high-risk infants at home * Head Start-type programs with weekly visits by teachers to students'
homes * extra police patrols in high-crime "hot spots" * anti-bullying programs in schools * drug treatment programs in prisons * special police units and prison for repeat offenders * rehabilitation programs focused on offender risk factors, such as
illiteracy
The list of programs that don't work includes: * gun buy backs * military-style correctional boot camps * summer job programs for youth * home detention on electronic monitoring * neighborhood watch * Drug Abuse Resistance Education (D.A.R.E.) classes taught to
school children by police officers
Lawrence W. Sherman, chair of the University of Maryland Criminology Department and lead author of the report, said that "If this country could do more of what works with money saved from not doing what doesn't, we might prevent millions of crimes annually -- from school shootings to shoplifting."
The study employed the University of Maryland Scale of Scientific Methods, ranking every evaluation on a scale of 1 to 5 for the level of certainty that could be given to its conclusions. The six members of the Department of Criminology research team then identified all programs that had at least two studies scoring a "3" or higher showing the program was effective. The same rule was used for detecting programs that had been proven ineffective.
The new report cautions that "all science is provisional," and that further research may produce different results in some areas. But the team has already begun the process of annually updating the lists of what works, what doesn't and what's promising.
Some of the 30 programs found to be promising include community-based mentoring by Big Brothers/Big Sisters of America, community-based after school recreation programs, and Enterprise Zones.
Maryland's Sherman said, "This study is an attempt to make scientific evidence--not political philosophy--the primary basis for saying what works to prevent crime. Just as medicine has increasingly moved towards the practice of `evidence-based medicine' about the choice of medical treatments, this report is an exercise in `evidence-based crime prevention.'"
According to Sherman, several states and other countries have adopted the Maryland Scale approach in making crime prevention policy. These include Ohio, Arizona, Washington, and the United Kingdom. "Now that the lists are available, we expect even more policy decisions to be informed by clearer evidence of what works," Sherman said.
Copies of both the original report and the new summary lists can be obtained free of charge on the Internet at www.preventingcrime.org.
The Maryland report summarizes what works in seven institutional settings of crime prevention, with different authors responsible for each area: schools (Denise Gottfredson), corrections (Doris MacKenzie), places/premises (John Eck), labor markets (Shawn Bushway and Peter Reuter), and police, communities, and families (Lawrence W. Sherman).
The report also concluded that "too little is known about the effectiveness" of the $3.2 billion in crime prevention programs funded by the federal government. It recommended that until more is known about what works, up to 20 percent of the federal crime prevention budget should be invested in rigorous testing of innovative programs. Otherwise, the report concluded, much of the funding may be wasted. "Crime prevention program evaluations are the only way to audit the profit and loss from the taxpayer's investment," Sherman said.
According to Sherman, true evaluations measure program results, defined as the cause and effect relationship between the program and some future level of crime and violence. While both descriptive and impact evaluations are useful, only the scientifically rigorous impact evaluations can tell us how effective a program is, or even whether it is backfiring--as some prevention programs have clearly done.
The need for more impact evaluations is shown most clearly by the longest list generated by the University of Maryland study--the list of 30 promising programs, says Sherman. He notes that if even half of these programs were found effective with one additional level 3 impact evaluation, the number of programs proven through rigorous scientific study to prevent crime would double.
Through the generosity of individual donors and private foundations, the University of Maryland has established a Crime Prevention Effectiveness Program to continue its work on crime prevention, and to communicate its findings to policymakers and the public. Philadelphia broadcaster and philanthropist Jerry Lee chairs its Advisory Board. Michael Buckley, the former associate director of the Consortium of Social Science Associations and a former Congressional staff member, serves as the program's executive director.
######
98093r, 7/27/98, Workman
RSS