> Perhaps use school records with documented evidence of bullying by a third party or something. That's actually less likely to be an accurate source, because

> Perhaps use school records with documented evidence of bullying by a third party or something. That's actually less likely to be an accurate source, because it depends on the school documenting everything, which to be quite honest, a lot of schools do not do -- I remember when I was bullied it was a literal uphill battle with the school to get them to recognize it in the first place, let alone do anything about it. How much of the bullying you saw in school, was documented by a third party? I would wager zero, but I'm up for being surprised. Psychology has a bag of tools, questionnaires are useful tools that fit a specific purpose and objective. They are relatively simple to create, perform, and get a lot of reasonably accurate data from them. And after all, what they're trying to assess here is the impact of bullying, so it doesn't actually matter what the exact bullying act is, in this case. They're simply trying to find a causal link between being bullied and negative behaviour. I think it's always amusing when people on hacker news think they know more about fields than, you know, people who have spent decades studying and working in that field. It says a lot about this website that it attracts folk who have no respect for other professions.