The Doom of Video Game Correlation
This article has been brewing for a while, but following the murder of 26 people in Sandy Hook, Connecticut, we made an editorial decision to let it sit for a while. We didn’t want it getting lost in the noise of a very important debate here in the U.S. or coming across as a knee-jerk, hypersensitve reaction to perceived unfair criticism. Let’s be fair: overreacting to perceived criticism is something gamers and the gaming industry are particularly adept at. I’d say, taking the industry and consumers as a whole, we’re second in overdoing our responses only to the NRA. (Ba-dum, tsh!). With that said, I wanted to take some time to do some in-depth analysis of society’s perception of games, violent crime and any tie between them.
There isn’t one.
If you didn’t stop reading, it means you’re not one of my sycophants who believes my every word. That’s okay. I have charts. Before I really get rolling, though, I wanted to touch on something that I think most geeks are aware of, but that (I get the feeling) a lot of other people aren’t. It’s the difference between a causative effect and a correlation. A causative effect is the result of an if/then. For example, If: you jump off of a 30 story building, Then: you are going to die. There have been lots of medical studies to back that conclusion up, so just go with me on this one. Sometimes it’s a more subtle effect, like an increase in risk factors, e.g.: If you use cocaine in the long term, Then: your risk, compared to a non-user, of experiencing psychosis and hallucinations is increased.
Correlation, on the other hand, may imply causation, but it doesn’t equate it. If we take two data points and examine their relationship to one another, in a vacuum, it might appear that there’s a causal relationship. The classic example is the relationship between the violent crime rate and ice cream. If you look at ice cream sales and violent crime on a monthly scale, the consumption of ice cream and the rate of violent crime both increase during the summer months. Taken without context, the conclusion would be that for some reason, there’s a direct causative relationship between ice cream consumption and violent crime. Taken without context, you’ve identified a spurious relationship by inferring causality, something that humans are prone to do. Another thing that humans are prone to do is be REALLY HORRIBLE at identifying risk. I mean, we are awful at it. Really awful. (I’ve mentioned stuff like this before, but seriously: the U.S. Department of Homeland Security had a budget in 2012 of $57,000,000,000. Your chances of dying in a terrorist attack are approximately 1:25,000,000…and it ain’t ’cause of the money we spend on DHS.
The reason I made you go through a refresher on Stats 101 and read about stuff that isn’t video game-related is because every time there’s a mass murder, two things happen: people freak out about how these things are escalating and then they look for a cause. Media outlets often latch onto this and proceed to scare the bejeesus out of everyone even though…
It’s safer now than it’s ever been. No, seriously.
Let’s start out with the crack epidemic, which suddenly made cocaine cheap, resulted in the 1986 Federal Minimum Sentencing Guidelines and may have had some things to do with an uptick in violent crime in the 70′s and 80′s. To be completely honest, I don’t have the time to really dig down into the data, but as you can see below, violent crime, aggravated assault and murder/nonnegligent manslaughter were relatively high until 1992/1993, at which point they started to taper off and have continued to do so since. Surprised? I wasn’t, but then again, I like statistics. More on this later.
If you look at cocaine, per the University of Maryland’s CESAR project, cocaine isn’t great for you. It strongly stimulates the production of dopamine in your brain and, over a 5-15 minute high, gives you a feeling of hyper-stimulation and euphoria, but also jacks your heart rate, makes you more aggressive and paranoid and, once you come down, leaves you depressed and craving more. In the short term, it can also kill you. Long term, it’s just…bad. You’re irritable, you have mood disturbances, are prone to paranoia, psychosis, hallucinations, stroke, sexual dysfunction and… it can kill you.
Video games on the other hand, according to my sources (me), might cause some seratonin and dopamine stimulation (I have fun and relax at the end of the day with some gaming), but there’s not much research out there supporting a causal relationship between gaming and insanity like the relationship cocaine and psychosis have. It might lack scientific rigor, but I’m going to go out on a limb and say that video games are demonstrably not as bad as crack.
There are enough people out there with some media clout, like Senator Jay Rockefeller (D. W. Va.) who are just sure video games are the equivalent of the Visigoths at the gates of modern society. Immediately following the Sandy Hook shootings, he introduced a bill to study the impact of violent video games on children. (Never mind that studies showing a deleterious effect have been shot down and there’s another that looks at 10 developed countries indicating there’s really no link between gun violence and gaming). If Senator Rockefeller were serving in the U.S. Senate in the 1950′s, it’s likely he would have been spearheading the crusade against comic books, blaming them for violence and moral turpitude of the day…and he’s probably one of the people who buy the veracity of the notably crazy person, Anders Brevik, who killed 77 people in Norway, that he used first person shooters to ‘hone his shooting skills,’ according to this not-even-slightly-biased article from the New York Daily News.
Now, I’m a gun owner. I like to think I’m a responsible one. And I found the concept of using Call of Duty: Black Ops II as a stand-in for marksmanship training a little hard to swallow. But, since I’m trying to rely on actual information as much as possible here, I thought it would make some sense to talk to someone who has a little bit more firearm training than I do. I called up a source at the Indiana State Police and asked him what he thought about this. He wrote me a pretty lengthy email in response, which is too long to quote in its entirety, but here are some comments from someone who carries a gun, deals with other people carrying guns who might object to his presence, and enjoys a relaxing evening with his console as much as I do. I’ve selectively pulled some of the email, but I’ve done so in a way that I hope does justice to his overall letter.
I, like you, was raised with both firearms and video games, and have never once felt the desire to go on a rampage. I knew that guns were not toys and that they were to be treated with respect. I believe, through my experience and observations that…broken families, abuse, education, and mental issues, all contribute to someone’s propensity for violence. The fact that there are media choices that allow for the expression of violence is not the cause of someone committing a violent act in real life.
As far as using video games to prepare to commit a violent act…is an interesting issue. I would agree that [a video game] can desensitize a player’s brain to violence [but] being desensitized …and being able to commit the same act are two different things.
[G]ames also depict other aspects [of firearms] poorly, or not at all. [I]n real life, ammo is heavy, takes time to reload and is not in unlimited supply.
Playing a sniper in CoD does not make someone the next Chris Kyle any more than jumping behind the wheel of an F1 car in Gran Turismo makes someone the next Michael Schumacher.
To summarize: experiencing firearms and violence in a game does not make it easier to utilize the same in real life.
So, you’re following me, but you’re probably saying, “Yeah, but still…even if it’s not going to turn kids into raving lunatics, I’m pretty sure there’s some stuff out there in video game town that’s just not appropriate for kids.” Know what? I agree with you. Fortunately, we have the ESRB. Sure, legislators all over the place keep trying to tack fines and criminal charges and create revenue, but it turns out that the ESRB is doing a really, really good job. The ESRB says they’re at about 85% penetration with parents understanding the ratings and 70% of those parents using the ratings to inform their gaming-related purchases for their children most of the time. They do such a good job that the Federal Trade Commission has gone out of its way to praise the ESRB’s work and if you’re a kid gamer without your parental unit, you’re better off trying to buy explicit CD’s (they still make cd’s?), DVD’s, or getting into an R-rated movie than you are picking up the newest M-rated title.
Even with those successes, the Entertainment Software Association (a lobbying / interest group representing the gaming industry) recently launched a ratings/parental controls education campaign intended to use Public Service Announcements through a variety of electronic media to further educate parents how they can keep their kids playing age-appropriate games. The only other industry I know who works that hard to educate its consumers about appropriate use is the Tobacco Industry, and it took a multi-billion dollar settlement agreement to get them to do that.
Seriously. Trying to tie video games to violence is tired. It’s untrue. It’s expensive and harmful and, for the most part, if you’re sitting in your house gaming, you’re not robbing convenience stores or shooting someone in a nightclub.
Has anyone seen the numbers?
Since a lot of legislation and media commentary falls under some form of the ‘Save the Children’ headline, it’s critical to look at demographics. Although the figures come from the ESA, which is not an unbiased source, they appear to be the best (or perhaps only) data available. Specifically, sourcing from their 2012 data, the average gamer is 30, has almost a 50% chance of being a female and the 17-or-younger male set only makes up 17% of the gameplaying consumer base. That’s hardly in line with the image popularly presented of M-rated games selling millions of copies because they’re bought by 14 year olds.
Demographics are one thing, but there’s also the issue of real research (the rigorous, peer-reviewed kind) like a 2009 meta-analysis of the impact of media violence on public health published in the Journal of Pediatrics (you can read it in full here, but it’s $30), whose authors conclude, “[t]his analysis does not find support for either a causal or correlational link between violent media and subsequent aggression in viewers. Why the belief of media violence effects persists despite inherent weaknesses of research is somewhat of an open question.” There’s actually a whole host of research out there (Here’s another one from Criminal Justice & Behavior in 2008, which stated, “playing violent video games does not constitute a significant risk for future violent criminal acts.”) There’s a rather large body of research on the topic, most of which reaches a similar conclusion, raising the question: why is this still being perpetuated? It certainly doesn’t help when you have killers blaming the games themselves, but it’s likely an echo of our weak ability to identify and quantify risk. When mass murders like in Aurora, Colorado or Sandy Hook take place, perpetuated by an angry, mentally disturbed young man, it’s hard to find a reason but (!) what do male loners do in their spare time? They play video games! Eureka. Suddenly, talking heads and politicians have a correlation they make causal and bob’s your uncle.
In the Washington Post’s analysis of ten countries where video games are also popular (China, Australia, Germany, Canada, France, South Korea, the Netherlands, the United Kingdom and Japan), there was a decreasing trend-line between video game consumption and violent crime. Canada, which has similarly lax gun laws as compared to the United States, had the largest number of gun-related murders per 100,000 citizens, which suggests that, rather than video games, one might be better off addressing issues like availability and funding for mental health and availability and accessibility of firearms instead of making the faulty (and debunked) conclusion that playing violent video games has a direct relationship with violent behavior. It’s like assuming Senator Rockefeller’s childhood participation in games of Cops and Robbers means he’s more likely to shoot up the U.S. Senate than the next guy. Which might be a fair assumption, I don’t know enough about Sen. Rockefeller’s current mental health status, but I doubt it.
I don’t have a fix for violent crime in the United States, and I’m not sure about what kind of restrictions on firearms is appropriate (or if it would even be effective – I have a sneaking suspicion that it wouldn’t be.) I don’t think that 10 year-olds should be playing M-rated titles and, when I have kids, they’ll have access to age-appropriate games. And I definitely, absolutely, positively don’t think there’s any tie between video games and violence. Except maybe this one:
If you look up at those charts regarding violent crime per-capita, you’ll notice the long declining tail begins in approximately 1993. This was id Software’s heydey, and Doom, it’s first insanely popular title which really ushered in the era of the adult-oriented FPS, was released the same year violent crime started its decline.
Now there’s a correlation I’d like to falsely identify as causal.
US News and World Report
University of Maryland CESAR Project
Schneier on Security
New York Times
New York Daily News
National Public Radio
US Department of Justice / FBI Uniform Crime Statistics
The Entertainment Software Association