<< The Grotesque Results of Politically Correct Law Enforcement | Main | News Scan >>

Intervention lowers drug, crime and behavior problems, but not by much

The title of this post is the headline of a story by Geoffrey Mohan in the Los Angeles Times.  He reports on "a 10-year study that [was designed to] compare how children would fare under prolonged therapy and tutoring aimed at improving social and cognitive skills, and whether their adult fates would differ from similar children who did not participate."
These kinds of studies in social science are called "longitudinal studies," and they have a special, important place.  They answer questions that can't be answered by snapshots in time.

The problem with longitudinal studies, of course, is that that they take so darn long to do. There is a saying among landscapers:  "When is the best time to plant a tree?  Twenty years ago."  Turns out that's a pretty good time to start a longitudinal study.  Absent any assistance from Doctor Who, we are limited to results of studies designed and funded many years ago.

On this particular study, Mohan reports (emphasis added):

Dodge's team, which included researchers from Penn State University, the University of Alabama, Simon Fraser University and Tufts University, screened nearly 10,000 children to identify 891 kindergartners who displayed aggressive behavioral problems in school and at home. They came from elementary schools in poor and crime-ridden communities near the universities.

Over the years since the study's 1991 launch, researchers have published several progress reports. But this time, they wanted to see how subjects had fared eight years after the program ended, based on interviews with the participants and those who knew them well, and a review of court records and other public documents.

Results, published online Tuesday in the American Journal of Psychiatry, show modest improvement in psychological markers that predict long-term antisocial behavior and criminality. Program graduates also had fewer legal problems, substance abuse issues and risky sexual behaviors. The data suggest that intervention can work, and effects can persist over many years.

Overall, the likelihood of psychological, criminal, sexual and behavioral problems dropped by about 9 percentage points from those of nonparticipants, the study found.

"It's not miracles; it's not huge impacts," said Dodge, who directs Duke's Center for Child and Family Policy. "We weren't successful with every child, but on average we have been able to prevent some of those [negative] outcomes."
The cost, $58,000 per child over 10 years, could prove problematic, particularly because participants were no more likely to have graduated high school or gained full employment 18 years later. And incarceration rates differed by only 1.3 percentage points between participants and the control group.

"It's a hefty sum, making it daunting to finance a program like this nationwide, or even school district-wide," Dodge said. "On the other hand, we know that the kids otherwise are going to grow up to cost society an awful lot."

A career criminal with a variety of other behavioral problems, the study noted, can cost society upward of $2.6 million over a lifetime.

As I noted in a post last week, changes of a few percent do mean real improvements in real lives of real people and are not to be brushed off lightly, but anyone who claims these programs will make massive changes in crime, incarceration, or recidivism rates is selling snake oil.

Oh, and in any well-written study, there is always the Limitations section.  That's like the footnotes in corporate financial disclosures.  Look there first.  That's where the bodies are buried.

The study did not compare the proprietary curriculum, developed by Dodge and other authors, with other interventions, and could not discount effects from other factors, such as enrollment in special education and receiving medical treatment for such diagnoses as attention deficit disorder.

Leave a comment

Monthly Archives