Fort Hood Fallout
Illusory correlation and the Fort Hood tragedy.
Posted Nov 07, 2009
Consider the following research study: you're shown flash cards with information about individuals from two different groups, X and Y. For both groups of people, 75% of the individuals are described as having engaged in some sort of positive, expected behavior. Like tipping their waiter, holding the door for someone else, or helping a fellow shopper load groceries into her car.
The other 25% of these fictional individuals are described as having engaged in negative, deviant behavior. Skipping out on their restaurant bill. Letting the door close on someone behind them with hands full. Trying to sneak more than a dozen grocery items through the express checkout. Etc.
When you're then asked to estimate what percentage of Group X and Y individuals exhibited deviant behavior, your answer will typically depend on relative proportions of these groups in the population at large. The more distinctive a group is, the more likely you are to associate its members with distinctive, deviant behavior.
So when there are 100 Group X individuals in your flash card population, but only 20 Group Y people, it seems like more of the Group Y folks are doing negative, distinctive things. Even though the rate is the same 25% in both Groups X and Y, we tend to inflate the number for the minority group, Group Y.
It seems a legitimate worry in light of all we know about illusory correlation. Not only do we tend to overestimate the association between distinctive behaviors and distinctive groups, but this tendency is also exacerbated when we have pre-existing expectations that the variables in question go together. And the stereotype of American Muslims as violent extremists was clearly a salient one in post-9/11 America long before this week's tragedy at Fort Hood.
Again, I don't mean to suggest that Hasan's religious beliefs are irrelevant when examining his terrible actions. Time will tell, but many indications suggest that they are related. And by no means do I seek to minimize the horror of what he appears to have done. But it seems a legitimate question to ponder how his actions may impact the way Americans see his group more generally.
Timothy McVeigh was a libertarian and NRA-loyalist; Eric Rudolph cited his religious beliefs to explain his violent opposition to abortion. But as White, Christian Americans, their social category membership wasn't particularly distinctive. Thus, it's not surprising that their abhorrently deviant acts didn't have much lasting impact on the perception of their groups at-large. Research on the illusory correlation suggests that the fallout to the Fort Hood shooting could be different in that respect.
At least, that's the fear of many Muslims today, as demonstrated by the following quote in the New York Times from an ex-soldier who attends the same mosque as Hasan: "When a white guy shoots up a post office, they call that going postal, but when a Muslim does it, they call it jihad."
This isn't the first time the stress of deployment and other untold factors have driven someone to open fire on fellow soldiers. It isn't the first time a disgruntled employee has turned a weapon on co-workers–in fact, it wasn't even the only such incident of workplace violence this week. And unfortunately, it won't be the last time either. We shall see whether the tendency towards illusory correlations plays any role in how this particular tragedy is reported and responded to.