The Overhyped Data on Video Games and Aggression

When anti-video game researchers sell their science at the expense of truth

Posted Oct 08, 2018

Creative Commons
Source: Creative Commons

This essay is coauthored with Patrick Markey at Villanova University.

“If your kids are playing these games, either these games are having a warping effect on right and wrong or they have a warped sense of right or wrong and that’s why they are attracted to these games.”  This extremely strong claim was not made by a politician or a moral crusader worried about children playing violent video games like Fortnite.  Instead it was said by Jay Hull, an author of a recent study from Dartmouth University. This study, and its accompanying press release, claimed to have found new evidence linking violent video games to youth aggression.  However, a reading of research paper itself, instead of the press release, suggests that the Dartmouth study is actually a better argument against the belief that violent games are dangerous.  It turns out that this is just another example of how social science is often oversold to a public without the statistical knowledge to evaluate problematic claims. 

In the Dartmouth study, the authors conducted a meta-analysis of studies of youth to see to what degree violent video games contributed to physical aggression.  When most people hear that researchers investigated something like “overt physical aggression” they might assume they examined things like actual physical aggression – aggravated assaults, fights, or even homicides.  However, in the meta-analysis they focused on studies of youths self-reports of behaviors or thoughts which most people would not necessarily consider to be particularly dangerous.  For example, numerous studies included in their analysis examined responses to items such as “I have become so mad that I have broken things” and “If I have to resort to violence to protect my rights, I will.” While these might be negative behaviors it seems like a stretch to characterize such responses as “overt physical aggression” in the sense of considering criminal assaults or schoolyard fights.  

Even with the overselling of the outcome as physical aggression, the results of the study itself suggest that violent games are associated with less than 1% change in youth self-reports of aggression.  Put another way, if the only thing you knew about a group of kids was their gaming habits, your ability to predict which kid would say they are aggressive would be essentially no better than a coin toss. Effects this low are usually the result of artifacts of social science research itself.  If you ask people if they play violent games, then ask them if they have broken things, their answers will tend to drift toward one another a little bit.  In other words, one set of questions biases the responses to the second set of questions, and this can result in small correlations that don’t reflect reality.  That’s one reason the small effects like those seen in the Dartmouth study are typically considered “trivial.”

In fact, the statistics from the Dartmouth study are largely identical to a meta-analysis conducted back in 2015, only in that case results were interpreted as evidence against video game effects, rather than for them.  The Dartmouth study actually provided little new evidence, other than how scholars can fail to put trivial effects into proper context and oversell their findings to the general public. 

In contrast to the Dartmouth study, as we document in our book Moral Combat: Why the War on Violent Video Games Is Wrong, when people are playing or consuming violent video games acts of real-world aggression – aggravated assault, homicides, and school shootings – have been found to decrease.  Countries that consume more video games have lower levels of violent crime then those devoid of this media.  Months when people are playing violent video games are safer than month people are not at home playing video games.  Even when violent video games like Grand Theft Auto are released there are observable decreases in homicides and assaults.  Such findings have been replicated by criminologists, psychologists, and economists at different universities while taking into account numerous potential other variables.  Such consistent results are why about 90% of scientists who study video games or crime are skeptical concerning the links between video games and real-world violence. 

How the authors of the Dartmouth study chose to “sell” their research to the public is reflective of a larger issue for social science.  Psychology has been experiencing a replication crisis, wherein it is now known that many overhyped findings are, in fact, difficult to replicate.  Much of what we, as research psychologists, tell the public simply isn’t true.  But methods of psychological science can be cleaned up and made more transparent.  The overhyping of miniscule findings is a different issue, one that reflects the culture of psychological science.  Too often that culture resembles a company marketing a product rather than an honest appraisal of too-often weak evidence.

So, in fact, the Dartmouth study provides little by way of new evidence.  These data have basically already been available, and are the basis for why most scholars, in fact, reject the notion that violent games or media contribute to serious real-world aggression.  The American Psychological Association’s own media psychology and technology division released a statement in 2017 cautioning policy makers and news media to stop linking violent games to serious real-world aggression as the data is just not there to support such beliefs.  The Dartmouth study, by finding that the effects for violent games on self-reports of aggression are near zero, in fact, confirm this position.  Statements by these scholars to parents that their children are “warped” if they play Fortnite or some other video game are not back by the findings from this study or any other study that has been conducted.  These false claims do nothing but manufacture fear and distract us from the more pressing causes of societal violence. 

Christopher J. Ferguson, a professor of Psychology at Stetson University, and Patrick M. Markey, a professor of Psychological and Brain Sciences psychology at Villanova University, and are the authors of “Moral Combat: Why the War on Video Games Is Wrong.”