Success of Anti-Crime Programs Is Elusive

How, and to what degree, can crime be prevented, and how may success, or failure, be reliably measured? Congressional debate on the crime bill refocused attention on these questions, which are as old as police departments.

The bulk of the legislation’s proposed spend ing is directed toward restricting criminal opportunity. More police officers, more prisons, longer periods of incarceration, broader application of capital punishment--all are believed to be effective deterrents. Logic, too, seems on our side. But objective data in support of our belief--and logic--is scarce, at best.

Most of the rancor surrounding the crime bill, however, is reserved for “crime prevention programs,” which aim to diminish crime by changing individual attitudes, thus behavior. Drug courts, the Violence Against Women Act, the Local Partnership Act, among other provisions in the bill, all promise to do this. These proposals have drawn fire, but nowhere near the firestorm of ridicule and contempt as have inner-city youth activities. Opponents mouth “midnight basketball,” “arts and crafts,” “self-esteem counseling” and “after-school programs” as if they were expletives.

Supporters of these programs claim they have evidence to buttress their case. Most of it, however, is anecdotal.

One bright spot in attempts to assess the effectiveness of such purely preventive approaches is the Los Angeles Police Department’s DARE (Drug Abuse Resistance Education) program. Because a substantial commitment of police resources is required to teach the DARE curriculum, city and police administrators demand an elaborate and aggressive evaluation effort. Professionals in the design and administration of scientific surveys were retained. That DARE has survived more than a decade of tightening budgets speaks not only to its effectiveness, but also to the value of a reliable evaluation system. Still, the DARE record yields no clue as to its effectiveness as a crime-prevention tool.

The problem is that opponents--or advocates, for that matter--want to know, “How much crime did not occur because of this or that program?” Measuring the absence of anything is always risky business. Generally, police compare recent statistics on reported crime with prior numbers, then, through deduction, judge whether the incidence of crime has increased, decreased or remained static. This seems straightforward, no?

Well, several assumptions are made in such moves from A to B, chief of which is that reports of crime consistently relate to the actual incidence of crime. That assumption, it turns out, is questionable.

Community surveys conducted in Los Angeles and in other large U.S. cities indicate that much crime is never reported to the police. When police are notified that a crime has occurred, variations in police documentation policies and practices frequently undercut any attempt to draw definitive conclusions.

Consider the situation that developed several years ago in the West San Fernando Valley. The local police-station commander detected a significant increase in burglaries. He also noted that many of these burglaries involved garages with open doors. He quietly directed that such incidents--no ostensible breaking and entering--be reported as thefts rather than burglaries. Burglary statistics plummeted.

Most variations in reporting are much more subtle. They result from fluctuations in workload levels, police response times, citizen confidence in their police department and other factors. The presumed effects of any anti-crime program, whether it be police suppression or social work, can be negated, masked or even enhanced by these reporting factors without any real effect on crime.

Does this mean we should not engage in any of the programs included in the crime bill? Of course not. What it does mean is that regardless of who pays for them, all anti-crime programs must have an evaluation component, even though that will add to administrative costs. The longevity--and success--of the DARE program urges us to do no less, even though we may doubt the “objectivity” of our crime-measuring methods.