What I would like to see, to demonstrate whether this is correlation and causation, is stats from different states. Bear with me here. If different states have different penetrations of gaming then it would be interesting to see how violent crime drops between states with lots of gamers vs states with low amounts of gamers. You could also check against different state wealth vs amount of gamers vs drop in violent crime rates. I add population wealth as wealth per citizen increase could also be the thing that decreases violent crime, and also happens to allow people to buy gaming systems. So it's worth check that it is videogames rather than affluence that has the link.DefiningReality said:Bad form Escapist and Honorof. Greg Tito posted about this same article not two months ago and he at least had the integrity to post a link to it and not pretend like it was a new story. Since you couldn't be bothered to link to his article, please allow me...
http://www.escapistmagazine.com/news/view/111191-Less-Crime-in-U-S-Thanks-to-Videogames
And the article itself...
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1804959
As for the article itself, it has three big problems.
1. I'm not saying that someone "outside the guild" of professional psychology or sociology can't make a contribution to a field outside their expertise, but when *economists* working for an *economic* institute put out a study of other people's studies, people who are "in the guild," that will, if believed, improve the *economic* prospectus of a given industry, we might need to engage in a little word counting and ask some questions about the veracity of their claims.
2. I'm outside of both the Economic and Sociological guilds so I am not myself familiar with the SSRN but it seems as if it is not a peer-reviewed publication. Furthermore, the article's status on the SSRN as a "working paper" makes it seem even less credible. (If someone is familiar with the SSRN please feel free to correct me on this. As I said, it's outside of the regular databases I use so it may be perfectly respectable within its field.)
3. As stated above, the article is frequently confusing correlation and causation. Using correlation in academic work *isn't a bad thing* unless you stop your academic work with simply showing a correlation and then go on to theorizing as if you had just shown a causation. This is exactly what the article does.
Finally, just in way of examining our community's reaction to this article, read the twenty or so comments surrounding this one and then recall this Critical Miss...
http://www.escapistmagazine.com/articles/view/comics/critical-miss/8903-Critical-Miss-Gamer-Science
Now ask yourself if most of those comments aren't exactly what the comic predicts about Gamers and gamer science.
The argument that games cause violence as a necessity is laughable, trite, and put forward largely by the culturally ignorant, but the way to combat a lie isn't with bad science and humming loudly while we cover our ears with questionable journal articles. The way forward is to honestly admit problems in our community where they exist and face them head on and without fear. If we become that kind of community, then we just might actually make the world a better place in reality and not just in theory.
If you find that states with high gaming populations have the highest drops in violent crime and that state population wealth does not affect the amount of decrease in those crimes vs amount of gamers, I would think that would go a long way to proof causation rather than correlation.
If you read the Freekonomics books they make a very good statistical case that the crime drops have been due to the legalisation of abortion 20 years before the crime drop....