Let's assume that Group X is responsible for nearly 50% of murder suspects according to eyewitnesses but make up a reasonably small percentage of the population.
The assumption is that this group is somehow special. It is. But the assumption carries with it a special emphasis on tracking members of that group, of suspecting that they're all bad, encouraging fear of each and every one of them.
And for many this seems reasonable. It's not. In fact, that's the actual numbers for a particular ethnic group in a Western country, and to single out every member of that group as worthy of special attention is viewed as profoundly discriminatory. Because it is. Such a view gets the stats backwards.
Given that murders aren't common, it's a very small percentage of Group X responsible for the disproportionate stats. At the individual level, the group stats say precious little about the individual and certainly doesn't merit special scrutiny.
It works for ethnic groups. It works for the usual way that we define the set of deviants that love their guns. I know lots of them. None has every so much as lifted a finger to hurt another person in the decades I've known them. This is consistent with the facts viewed accurately. Now, do some of them go and do bad things? Sure. But just as we don't like ethnic profiling because it's based on a statistically false conclusion, so we don't profile "gun humpers" because such scrutiny would also be based on a statistically false conclusion.
And, yes, some say we're stupid for opposing racial profiling. Mostly we just suck at intuitive approaches to probability and statistics. Just ask nearly anybody who's taken the class. Our intuitions tend to betray us every time.
= new reply since forum marked as read
