One of the strangest (to me) kinds of mental bias is the tendency reject a statistical argument due to statistically insignificant exceptions. Curiously, I’ve mostly encountered this with people who are seriously interested in the social sciences, a field that is pretty much entirely statistical and quantitative (or at least the useful parts are). Of course, people only do this when they know what the correct opinion is anyway so they’re not so bothered about engaging the actual analysis.
Example: I was recently talking with a friend about differences between the sexes. The conversation went something like this:
Gandalf Kensington (name changed): But are you talking about sex or gender, here?
Me: Sex, although I think gender is almost the same thing – so close that the same things apply.
GK: Well, the whole debate is kind of pointless because sex doesn’t really exist.
Me: Oh? I’d say the distinction between the sexes is pretty clear-cut.
GK: What about gender-ambiguous people? Or transsexuals? Obviously it’s not binary, it’s a continuum.
Me: But those constitute such a small part of the population that we can just throw them out and look at the rest of the data.
GK: That’s so typical of sex-normative discourse: just throw out the exceptions and get the results you want. You can’t ignore reality!
Me: I’m not ignoring reality, I’m just approximating a part of it away.
GK: Well, you can’t.
If people thought like this about everything, it would be impossible to make sense of the world. There are always exceptions and statistical noise. That’s not to say it isn’t understandable; it’s difficult to accept unpleasant things. I’m sure I do it a lot without noticing. But I don’t recommend this approach people who want to change the world. Ignoring the unpleasant parts is very unlikely to change them.
