Fallacies of Connections between Violent Videogames and Violence in Real Life

Jack Thompson continues to insert himself in the news with his claims that playing violent videogames engenders violence in the real world. I went up against Thompson last June on CNBC's Squawkbox about this issue, and was repeatedly interrupted not only by Thompson but the clueless host Becky Quick.

There is a lot misinformation floating around about the connection between violent videogames and violence in real life, and scientific studies that allegedly demonstrate this. In what follows, I'll just briefly highlight the two main kinds of studies, and why they fail to demonstrate a connection, despite what Thompson and many other people say.

1. Experiment: Two groups of kids, randomly selected so both groups are equivalent. One group plays violent videogames, the other group plays nonviolent videogames. The group that plays violent videogames is deemed to have become more violent, as indicated by a variety of factors measured in the experiment. For example, both groups are given violence profile exams, which ask, among other questions, what you would do if someone bumped into you. The group that played the violent games is more likely to say they would haul off and punch the person, etc.

Does that show that playing violent video games increases your propensity for violence in the real world?

No, it does not. It shows only that playing violent videogames makes you score higher on various laboratory measures of potential for violence. Whether these subjects then go on to actually commit or get involved in violent acts in the real world is outside the scope of such kinds of experiments - as, indeed, it must be, since once the subject leaves the lab or premises of the experiment, he or she is open to all kinds of other influences.

2. Survey: The number of real-life violent episodes, defined by whatever criteria (hitting, pulling hair, etc) are tallied for individuals in a survey, in which the playing of videogames is also tallied. Conclusion: there is a statistically significant correlation between playing violent videogames, and getting involved in violent episodes in the real world.

Does that show that playing violent video games increases your propensity for violence in the real world?

No, it does not. Correlation is not the same as causation. Correlation means A and B happen at the same time. Causation means A causes B to happen. How could A and B be correlated, and not also exist in a cause-and-effect relationship? Well, maybe B causes A (in this case, kids who are already more violent enjoy violent video games because it ratifies their behavior - in contrast to nonviolent kids, who enjoy videogames for other reasons). Or maybe a third factor, C (for example, problems in family life), causes both A and B to happen.

The above is not meant to be an exhaustive assessment of every single study ever conducted about violent videogames. But next time you see or hear a study authoritatively cited, ask whether it has avoided the above fallacies and pitfalls.

Listen to 10-minute podcast on this issue: Ask Lev



========================================
Digg   del.icio.us
reddit
=================================================


Following comments were made in response to original PaulLevinson.net post of 20 April 2007. Please feel free to your current comments here.

Comments

I agree with you on this one. My memory of "middle school" (jr high school back then) was that violent encounters were not infrequent. Sort of how status was determined on the primate tree when it came to males. Needless to say video games didn't exist in the very early 50's and the hero shot the gun out of the hand of the villain in the movies. One can come to suspect that these tendencies are "hard wired" into our reptile brain.....hmmmmm
Posted by: George at April 20, 2007 04:10 PM

I'm sure glad there are people like you, Paul, who understand science. Keep it up!
Posted by: Rat Fink at April 20, 2007 11:42 PM

Thanks, George and Rat Fink. One good thing about Jack Thompson is that he brings most rational people into complete agreement - on the absurdity of his claims.
Posted by: Paul Levinson at April 21, 2007 02:50 AM

Thanks for being the voice of reason. After watching your squawkbox showdown, I could not believe how rude both Thompson and the host were. Keep up the good work!
Posted by: Anonymous at April 23, 2007 08:59 PM
Hi
You are The Best!!!
G'night
Posted by: Terabanitoss at May 7, 2007 04:18 AM