American Academy Of Pediatrics Claims Broad Consensus On Violent Media Effect That Doesn't Remotely Exist
from the fake-it-until-you-make-it dept
Search through all of our stories about the supposed link between violent movies and games and real world violence by those that enjoy them, and you should come away with the impression that, at the very least, the science isn't settled on the issue. The more specific impression you should get is that violent media might -- might -- have a short-term impact on behavior, but that there isn't anything like a general agreement on the long term effects, which is obviously the vastly more important question.
Yet the American Academy of Pediatrics just released a policy statement on "virtual violence" that recommends legislative action due to a "broad scientific consensus" that violent media increases aggression in children.
Although there is broad scientific consensus that virtual violence increases aggressive thoughts, feelings, and behaviors, there has been little public action to help mitigate children’s exposure to it.
As a policy position, the statement goes on to recommend federal control over media ratings and a policy of reduced screen time to be prescribed to parents by pediatricians. We could spend all kinds of time arguing over whether those recommendations are good, bad, or somewhere in between. Instead, let's go ahead and expose the whopper of a false premise that is the so-called broad consensus on the effects of violent media.
Christopher Ferguson, a Psychology professor, wrote at HuffPo about how silly the claim is.
First, the AAP presents the results on media violence as if they were consistent. They argue that “hundreds” of studies show that “…the linkage between virtual violence and aggression has been well supported and is robust.” But this claim is easily contradictedby a whole host of studies that find no effect for media violence on aggression. Other studies have examined links between media violence consumption and societalviolence and found that media violence is, if anything, associated with reduced societal violence. This is not to say evidence is consistent against effects either. Some studiesdo find some evidence for media effects (although typically small and usually for minor behaviors) yet others do not. Claiming consistency in either directly merely discredits the claimant as a credible source of information.
This claim also ignores increasing controversies about media effects’ role in psychology’s replication crisis. Scholars have identified “questionable researcher practices” in this realm, and have had difficulty replicating old studies, and studies using preregistered designs (where scholars publish their methodology in advance of collecting data, making it harder for them to monkey with analyses to get the results they want), generally found little evidence for effects.
He goes on to note that, despite the AAP's claim of broad consensus on the topic, the policy paper cited very few studies to back this claim. Those it did cite, of course, agreed with its position. But ignoring the multitude of studies that don't is. in itself, a form of citation bias, particularly given that the paper's two chief cited sources cited are generally considered controversial on the topic of media violence. Included in the post is a wonderful graph going back twenty years showing that as violent media consumption by youths has gone up, youth violence reports have gone in the opposite direction. It would be hard to square that graph with the AAP's claim, never mind with its claim of broad consensus on the topic.
But this is more than a mistake. The AAP should know better than this, and does.
The AAP claims the presence of a consensus of scholars and practitioners regarding media effects. However, most surveys of scholars and clinicians reveals this to be false. Depending on how questions are asked, surveys indicate that only 10% to 58% of scholars agree some kind of links may exist. Higher numbers are found when undefined “aggression” is used in queries (which could involve fairly minor laboratory tasks like filling in the missing letters of words...”kill” being more aggressive than “kiss” for instance in response to ki__), and drop significantly once questions ask about youth assaults or other significant behaviors. All surveys indicate significant disagreements among experts, with only a minority worried about media effects on violent behavior. Indeed, past claims of consensus have been thoroughly discredited.
Further, a group of 230 scholars, back in 2013, wrote an open letter advising against policy statements on media effects just like this one. That letter was written to the American Psychological Association (APA), but the same principle holds here. How does the AAP pretend this large group of 230 scholars don’t exist?
Combined with a later section of the paper, which laughably misreads a Supreme Court decision from 2011 as rejecting federal oversight on violent media consumption by children on First Amendment grounds when the court actually stated that lower courts have found the claim of a link between media and violence to be meritless, one wonders exactly what the motive is for all of this by the AAP. You have to blink before believing the AAP's recommendation that journalists not talk to "contrarian researchers" who don't see things its way. It's hard to think of a less scientific recommendation than censoring contradictory studies.
But more worrisome, the AAP appear to pressure journalists not to speak to anyone, including “contrarian scholars” who disagrees with their position. This arguably puts the AAP in the bizarre and aggressive position of, in effect, arguing for scientific censorship (whether or not this was their intent). They chide news media for presenting “both sides” of the debate (despite the presence of scientific evidence for both sides and the fact that most researchers do not agree violent media poses a serious risk to society) and offer as a recommendation “The news and information media should acknowledge the proven scientific connection between virtual violence and real-world aggression…”
The point of all of this isn't to fight over the link the AAP suggests. The point is that a policy paper built on a demonstrably false premise that then attempts to silence the work of opposing research isn't a prescription for good scientific output.