Reports of a study to be published in the October issue of the American Journal of Preventative Medicine have been making the rounds. The study looked at survey data collected from 552 subejects aged nineteen to ninety in the Seattle/Tacoma area. It was conducted by researchers at the Centers for Disease Control and Prevention, Emory University and Andrews University. Their hypothesis was that regular video game players would have a higher body mass index and more “poor mental health days” than non-gamers. The results reportedly confirmed the hypothesis.

In fairness, much of the media is covering this responsibly, pointing out that the study found only a correlation between gaming, higher BMI and depression. Unfortunately that isn’t going to deter the Jack Thompson crowd who will almost certainly latch on to this study as yet another example of the grave threat video games pose to both mental and physical health. The video game players in the study were more likely to be fat and depressed, so video games must be causing it. There is a fancy Latin name for that kind of argument; cum hoc, ergo propter hoc – with this, therefore because of this. People making this argument assume that since two events are seen together they must be causally related and this is just not the case. It is impossible to argue, based on the data, that the people who reported playing video games in the study would NOT be overweight and depressed if they weren’t gamers.

Ars Technica has a very good summary of some of the problems with drawing conclusions from this study, most of which are reported in the study itself but have been glossed over in the popular media. The main point is that there is a mantra that should accompany all studies of this kind. The mantra is “correlation does not equal causation” and it should be repeated frequently.

Edit – And, as often happens, XKCD says it best.