What I mean about "methodology" is the current trend of quantifying human personality traits, like that damned "agreeableness". I understand the aspiration to quantify the human personality (although I also personally find it a little off-putting) but it seems to me that it is very difficult to do with the tools with which it is being attempted. In fact, I think it's impossible to do what psychologists are trying to do, i.e. to do statistics on the human personality. Because the things you are trying to measure are impossible to define clearly and observe objectively since they have no physical substance.

So why do it at all? Why does psychology _need_ to have statistics on personality traits, when they are not really personality traits but complicated proxies for things we think may be representative of personality traits? Wouldn't the same effort be better spent on a different project? Can't psychology be taken seriously as a science without trying to do maths with things that don't really exist?