Game Violence: Correlation Is not Causation

Ars Technica notes an intriguing paper by Karen Sternheimer over at USC in the American Sociological Association’s publication, Contexts. The article makes for an interesting read, and supports many of the points I’ve made here. Basically, Sternheimer asserts an oft repeated (but rarely acknowledged) point that correlation studies simply do not prove causation. Just because the Columbine killers played Doom does not mean that Doom caused them to go on a shooting rampage. One could just as implausibly point out that most murderers brush their teeth; ergo, teeth-brushing makes people want to kill. Sternheimer notes that studies showing links between violent behavior and video games don’t tend to consider other contributing factors to violence such as home life and socio-economic backgrounds.

Sternheimer offers a neat summary of news articles appearing since 1997 that overwhelmingly decry video games as fostering school violence. The only problem is that statistically, school shootings are way down; in fact they are such rare occurrences that the FBI does not have enough data to create a profile of likely school shooters. So, while a purported link between video games and school shootings have been beaten into the public consciousness, the exact opposite has been taking place.

Video games became what Sternhemier labels as “folk devils,” or convenient scapegoats for politicians and the press to use in explaining otherwise inexplicable acts of violence. Alas, it is too easy for politicians and media types to paint with a broad brush. Their thinking might go something like this: “Some shooters played video games. Therefore, all video games are bad. And now, some idiots out there are trying to use video games for educational purposes …”

References:

Sternheimer, K. (2007, Winter). Do video games kill? Contexts 6(1). 13-17.
[Online]. Available:
http://caliber.ucpress.net/doi/pdfplus/10.1525/ctx.2007.6.1.13?