Phantom Studies

"The verdict on violent entertainment is now in," Sen. Sam Brownback (R-Kan.) declared during the most recent moral panic involving kids and popular culture. "Violent entertainment is a public health hazard." To back up that claim, Brownback touted a joint statement from the American Medical Association, the American Psychological Association, the American Academy of Pediatrics, and the American Academy of Child and Adolescent Psychiatry. "Well over 1,000 studies," it said, "point overwhelmingly to a causal connection between media violence and aggressive behavior in some children."

That sounds pretty impressive, except for one thing: It's not true. Jonathan Freedman, a University of Toronto psychologist who recently completed a review of the scientific literature, counts about 200 published studies that have tried to measure the impact of TV or film violence on aggression. What's more, he says, "The majority of studies do not find evidence that supports the notion that television violence causes aggression." Senior Editor Jacob Sullum spoke with Freedman by telephone.

Q: One of the classic studies in this area found that preschoolers who were shown a film of a man attacking an inflatable clown doll were more likely to knock the toy around than preschoolers who didn't see the film. What are the problems with that sort of research?

A: The situation is very subject to "demand factors." The showing of the violent film, which is usually a very short excerpt, almost inevitably carries a message that this is expected or allowed or wanted. It's very, very artificial research. When you get effects, it's easy to explain in other terms. Another big weakness is that very few of [the studies] make any effort to equate the violent and nonviolent films in terms of activity level, arousal, or interest. More arousal produces more of whatever it is you're measuring, which could include aggression.

Q: What about studies in more realistic settings?

A: The field experiments show almost nothing….The contrast between the way it's described, in terms of "overwhelming evidence," and the reality is just remarkable. It's not that 60 percent of them get good results and 40 percent of them don't—which would be pretty iffy from a scientific point of view but at least substantial. It's more like 25 percent get supportive results, and 75 percent either get nothing or get mixed results.

Q: If the evidence is as inconclusive as you say, why do we keep seeing these authoritative-seeming statements to the contrary?

A: I think it's purely political. This most recent statement was elicited by Brownback. [The organizations] had issued similar statements before, separately, that were equally inaccurate. "Thousands of studies" is a joke. It's sort of like the McCarthy era: "I have here a list of Communists…." You just know that no one looked at the research when they said that, because if they had they would know there aren't 1,000 studies.