Posted by Steven Johnson is a writer who I very much admire. I’m particularly impressed by his defense of video games and other technologies in his book Everything Bad is Good For You. However, in defending the good aspects of video games, he has also felt compelled to downplay their negative effects. For example, a recent blog post argues vehemently that video game violence does not lead to aggression. He makes his argument by examining a recent study which found a link between violent game exposure and aggression. I’d like to do something a bit unusual for Cognitive Daily. First, I’ll examine the study itself, then take a look at Johnson’s critiques and see how well they hold up.
The study, conducted by Douglas Gentile, Paul Lynch, Jennifer Ruh Linder, and David Walsh, examined survey responses for over 600 8th and 9th graders. The students were questioned not only on their exposure to violent video games, but also on a standard measure of hostility (the Cook & Medley Hostility Scale). Finally, they were asked two simple questions: how often they argued with their teachers, and if they had been in a physical fight in the past year.
Perhaps not surprisingly, the researchers found a positive correlation between hostility and both arguing with teachers and fighting. Exposure to video game violence also correlated with these aggressive behaviors, as well as with hostility itself.
Next, the students were divided into two sets of quartiles based on the data: 4 equal size groups were created according to their exposure to violent video games, and another set of groups was made for hostility ratings. Not surprisingly, the students that placed in the top quartile for both measures were most likely to get into fights. However, for students in the lowest hostility quartile, those who had the most violent video game exposure were actually in more fights than those with the highest hostility ratings and the lowest violent video game exposure. Here’s a summary of this data:
| Hostility |
Exposure to violent video games
|
|||
|
1 (lowest)
|
2
|
3
|
4 (highest)
|
|
| 1 (lowest) |
4%
|
26%
|
15%
|
38%
|
| 2 |
16%
|
33%
|
27%
|
58%
|
| 3 |
24%
|
42%
|
44%
|
54%
|
| 4 (highest) |
28%
|
37%
|
58%
|
63%
|
Next, they conducted a logistic regression analysis—essentially a way to isolate the impact of certain factors while factoring out others. After controlling for hostility, they found that exposure to violent games still accounted for a significant portion of the physical fights kids got into. Interestingly, parental involvement in regulating their children’s access to violent games also had a significant impact.
Now, let’s take a look at Steven Johnson’s specific critiques of this study.
First, Johnson complains that the study doesn’t compare its results to other activities that are potentially associated with violent behavior—playing football, for instance. In one sense, Johnson is correct to point this out: what if it turns out that football is a more significant contributor to aggressive behavior than video games? However, one might also ask just how many other activities researchers might be asked to account for—gun ownership, knife skills, karate class—the list is endless. None of these examples take away from the fact that a correlation was found between violent gaming and aggression. Perhaps all these other issues warrant further study, but they can’t contradict the results of Gentile et al.
In a separate post, Johnson argues that this kind of study can’t explain the significant decrease in violence nationwide over the past two decades—the very same decades in which video games have significantly risen in popularity. Again, in one sense, he’s correct. But consider an analogy. If a 1940 study argued that bicycle riding results in a large number of head injuries, would you accept as a refutation of that study the argument that in the years between 1910 and 1940, there was a 50 percent decline in head injuries? Not if you also knew there was a 60 percent decline in horse riding during those same decades (I’m making these numbers up, by the way—the point is that it’s entirely possible for overall violence to decline while video game related violence is increasing. It doesn’t mean we shouldn’t do something about video game related violence).
Finally, Johnson quotes the following passage from the Gentile et al. article:
In fact, statistically controlling for respondent sex, hostility, weekly amount of video game play, and video game violence exposure, the frequency with which parents monitor their adolescents’ video game habits added a significant amount of predictive power when predicting physical fights.
Johnson uses this quote to suggest that parental control (or lack thereof) accounts for all fighting among youth. But this is not what the article says at all. In fact, I’d argue that it’s a weakness of the article that it doesn’t indicate what the contribution of video game violence is after accounting for parental control.
But Johnson takes this argument one step further, making the following claim:
Think about it this way: these kids who have heavy exposure to violent video games—their parents are letting them play at least 3-4 hours of these games a day. Isn’t it reasonable to assume that there might be something else going on in a household where the kids are left unsupervised to play violent games with that frequency?
Wait a minute… I thought Johnson was arguing that video games weren’t so bad. Now he’s making the opposite argument—that any household which allowed kids 3-4 hours of gaming a day must have something wrong with it. You can’t have it both ways—either video games do or don’t promote violence. If you now make the claim that anyone who’ll let their child play video games must be neglecting them and thereby promoting violence, you’ve got some ‘splainin’ to do. And what of the many studies that show that non-violent video games don’t cause aggressive behavior? If leaving kids unsupervised to play violent video games demonstrates that something is wrong with a household, then why doesn’t leaving kids unsupervised to play nonviolent video games indicate the same thing—unless the violence in the video games itself is part of the problem?
Johnson’s conclusion is most suspect of all: “if you’re an involved parent with a kid who doesn’t have any major aggression issues, then playing some violent video games isn’t going to make much of a difference either way.” Remember, the data indicates that video game violence matters more than aggression.
None of this is to say that the Gentile et al. article is the definitive answer to the problem. No one study is. And I’m certainly not advocating the panic or censorship that Johnson seems to fear, especially given the positive aspects of many video games. But there’s little reason to suspect that video games don’t have an impact. As Johnson points out in his book, games can be powerful learning tools. Why should we suspect that when games teach us violent or aggressive solutions to problems, we don’t learn those, too?
Gentile, D.A., Lynch, P.J., Linder, J.R., & Walsh, D.A. (2004). The effects of violent video game habits on adolescent hostility, aggressive behaviors, and school performance. Journal of Adolescence, 27, 5-22.
August 30, 2005
August 30, 2005
August 30, 2005
August 31, 2005
August 31, 2005
October 17, 2005
October 26, 2005
November 30, 2005
December 3, 2005
May 15, 2006
May 15, 2006