You can look at hard data and form models about groups (e.g. perhaps to best market to them), but I don't think that's what we're talking about here. And of co

You can look at hard data and form models about groups (e.g. perhaps to best market to them), but I don't think that's what we're talking about here. And of course, it is often important to group and categorize people to think about them. I'm not disagreeing with that. I don't think anyone reasonably intelligent could disagree with that. Let me put it this way: I'm a classic overanalyzer and I've found while I'm very good at analyzing and theorizing, if I do not have a rock-solid foundation for my premises my theories can be shockingly wrong. If someone snubs me in public I can create quite elaborate theories about why they are the way they are, but if I simply check with them to see if my premises are true I find that maybe they just didn't see me. of course, that's kind of a silly example, but the same goes for any kind of analysis. That's why you talk to people -- to test out your premises to see if they're really true. Because otherwise, when you analyze people, what are you really basing it on? A completely neutral, objective assessment based on data with an understanding that it paints multiple pictures? Or is it a passing value judgement when the data can be interpreted many different ways? To understand a subculture, you have to first understand that subculture. You don't only have to talk to people -- you can also read and think about their beliefs in good faith. But if you don't do either, you are essentially arguing with the title of the news article without reading the explanation!