With three recent deadly shootings (one in Isla Vista and the second in Seattle; a third in Las Vegas as I was writing this post), I’ve once again heard many ignorant statements thrown around regarding video games and violence. Much of the ignorance comes from making blanket statements completely lacking in nuance, from both sides.
Here is what’s wrong with the current discussion:
1. The anti-video game side ignores the actual crime statistics.
Whether you look at the past decade or past two decades (when video games arose and flourished), you see that general crime, violent crime, and juvenile crime are all down significantly.
Violent juvenile crime in the United States has been declining as violent video game popularity has increased. The arrest rate for juvenile murders has fallen 71.9% between 1995 and 2008. The arrest rate for all juvenile violent crimes has declined 49.3%. 
Of course, this does not mean that (violent) video games are causing the reduction in violence. Here is a graph that goes forward by several more years :
The point is that even if a study comes out demonstrating a link between video games and aggression, it is another step to go from aggression to actual violent crime, which is hard to measure because we can’t just run experiments on violent crime. To show that video games have a strengthening effect on the crime rate, you must show that in the absence of video games, the crime rate would be decreasing faster than it already is (or something equivalent to that).
2. Both sides have a wrong assumption about overall crime.
Because our media gives plentiful attention to violent crimes—the more deaths, the better—we get a sense that the nation is becoming more violent, and we desperately look for any changes that could have caused this increase in violence.
In fact, the violence rate was fairly constant until 1994, when it began dropping steadily :
The public does not see it this way. According to the same Gallup poll :
Despite a sharp decline in the United States’ violent crime rate since the mid-1990s, the majority of Americans continue to believe the nation’s crime problem is getting worse, as they have for most of the past decade. Currently, 68% say there is more crime in the U.S. than there was a year ago, 17% say less, and 8% volunteer that crime is unchanged.
Not as relevantly, but shockingly, even our long-term historical assessment is wrong. A poll was done in the UK on perceptions of violence :
When I surveyed perceptions of violence in an Internet questionnaire, people guessed that 20th-century England was about 14 percent more violent than 14th-century England. In fact it was 95 percent less violent.
This flawed assumption significantly changes the way we approach the video games and violence discussion. Instead of asking, “What is responsible for the recent rise in crime rates?” and noting that video games exist now whereas they didn’t exist before and then drawing the facile conclusion, we should ask, “Do video games hold back an even greater decline in violence?”
3. The pro-video game side ignores the link between video games and aggression.
Just like ignoring crime statistics, one can also ignore psychological effects of violent video games.
In a meta-analysis of the psychological literature, Craig A. Anderson and Brad J. Bushman, violent video games were generally found to be associated with aggression .
One concern of violent video games is that violence is often rewarded. A study [6,7] shows a difference in player aggression between a game where violence is rewarded and one where violence is punished.
It would be nice if psychological results were not ignored by the pro-video game side. On the other hand, psychological results are often tenuous and likely to be wrong. So it would also be nice if the anti-video game side took these results with a bit more caution. After all, some studies are skeptical of the video game-aggression link [8,9].
Finally, even if we assume that violent video games definitely lead to increased aggression, this is one step removed from deducing that video games actually lead to violent crimes such as shootings.
4. Mechanisms are argued instead of statistics.
I wrote about this topic before in my blog post “Mechanisms vs Statistics,” which incidentally used video games and violence as the example.
The gist is, if you don’t use statistics or real evidence, then you can argue anything you want. If you are anti-video games, you could argue that gamers imitate the characters they play, hence they become more prone to going on shooting rampages. If you are pro-video games, you could argue that someone who otherwise would have committed a violent crime satisfied their aggression in video games instead of in real life, thus decreasing crime. Without data, it’s hard to say which of these stories is more correct, or correct at all. (And you could come up with dozens of such plausible-sounding stories for either side.)
Even with statistics, we have to make sure to interpret the data carefully. Being relaxed with statistics will lead us to believe the wrong things.
 Anderson, C.A. & Bushman, B.J. (2001). Effects of Violent Video Games on Aggressive Behavior, Aggressive Cognition, Aggressive Affect, Physiological Arousal, and Prosocial Behavior: A Meta-Analytic Review of the Scientific Literature Psychological Science September 2001 12: 353-359.
 Carnagey, N.L., & Anderson, C.A. (2005). The effects of reward and punishment in violent video games on aggressive affect, cognition, and behavior. Psychological Science, 16(11), 882-889. http://pss.sagepub.com/content/16/11/882.abstract
 Williams, D. & Skoric, M. (2005). Internet fantasy violence: A test of aggression in an online game. Communication Monographs, 72, 217-233.