Monday, May 05, 2014

The Best of OSU/UMI Sociology

From CBS Cleveland:

Study: Violent Video Games Encourage Racist, Aggressive Attitudes Toward Blacks

Benjamin Fearnow

March 21, 2014 9:56 AM

Everything but the abstract is behind a paywall, so I’ll have to comment on the strength of the article.

“Our research suggests that people who play violent video games as violent black characters are more likely to believe that blacks are violent people,” writes a research team led by Grace Yang of the University of Michigan and Brad Bushman of the Ohio State University. “Playing a violent video game as a black character reinforces harmful stereotypes that blacks are violent.”

The study examined the effects of playing violent video games as a black avatar (versus a white character) on racial stereotypes and aggression. Games such as Grand Theft Auto V and Saints Row 2 allow players to choose the race of their character, and the study findings suggest that a player’s aggression against others is increased “immediately afterwards” in some cases, “even more than playing a violent game as white characters would.”

Multiple problems.  First, the experiment only included white test subjects.  The avatar effect of violent video games on black subject attitudes compared to white subject attitudes remains uninvestigated, but I will speculate that it would add unwanted nuance to the study’s conclusions about how racist white people are.  But nobody cares about black racism, so . . .

Second, gamers apparently chose their own avatars rather than had them assigned randomly.  So the door is open to selection bias.

After playing the games for 20 minutes, participants who played as black avatars were more likely to link photos of black faces with weapons, while those who played as white characters associated white faces with objects such as mobile phones.

Okay, but if your point is to show the effect of the video game, then you should show the associations with blacks by both groups of players.  In other words, ask your control group (playing white avatars) the same question asked of the test group (playing black avatars).  As reported, this study doesn’t really measure the effect of video game avatar at all.

They were also asked to respond to statements measured on a “symbolic racism scale” such as, “If blacks would only try harder, they could be just as well off as whites.”

Okay, so this sentiment is pretty dumb, but it strikes me as the exact opposite of racist.  In fact, if you believe the races are equal, then equal effort should produce equal results.

Participants were also tested on a seemingly unrelated food preferences test in which they tested hot sauce, and then were asked how much another person would like the spicy food.

Among players of the violent game, those who used a black avatar gave their “partner” more hot sauce compared to those who used a white avatar. The black avatar participants gave the hypothetical food “partner” more than double (115 percent) the amount of chili sauce than participants who played as white avatars.

The researchers suggested this element of the study was “particularly noteworthy,” because “this increase in aggression occurred over and above any increase in aggression among participants playing the violent game as a white avatar.”

This finding is interesting, but I would be reluctant to use hot sauce as a proxy for aggression in these circumstances.  It would be more useful if they were asked to add some arbitrary amount of actual hot sauce to actual food about to be eaten by an actual person who had actually told them already that they didn’t like hot sauce.  That would be a credible aggression proxy.  But making it all hypothetical isn’t.

In addition to negative attitudes held toward black people, the study noted that women and police were depicted negatively in a way that may have aggressive effects following gameplay.

“Police are portrayed as brutal. Players witnessing or enacting these violent actions may develop a distrust of police,” write the researchers. “Other violent games portray women in a sexualized and stereotypic way” that may impact male attitudes toward women in their real lives.

This is sheer speculation, not a study.

According to FBI data on violent crime and murder rates, in 2012, an estimated 1,214,462 violent crimes occurred nationwide, an increase of 0.7 percent from the 2011 estimate. Of the 12,765 murder victims in 2012 for which supplemental data were received, most (77.7 percent) were male.

Concerning murder victims for whom race was known, 51.1 percent were black, 46.3 percent were white, and 2.6 percent were of other races. Race was unknown for 130 victims. Of the offenders for whom race was known, 52.4 percent were black, 45.2 percent were white, and 2.4 percent were of other races.  The race was unknown for 4,077 offenders.

As these numbers should make obvious, Hispanics (i.e. Mestizos) are lumped in with the white numbers


Anonymous said...

So what does that say about people who play games that feature killing zombies?

heresolong said...

And what did all the crime stats at the end have to do with the "study"? Nothing, apparently, as it was just a recitation of violent crime rates.

Dr. Φ said...

PH: The sensitive term is "Undead Americans".