> ... when the study didn't look at any real-world decisions whatsoever, but merely at performance on a battery of academic tests, (whose relationship with real

> ... when the study didn't look at any real-world decisions whatsoever, but merely at performance on a battery of academic tests, (whose relationship with real-world decision-making we're expected to take on faith)? These aren't even decent forms of those academic tests, either. When I read it yesterday, I was very unimpressed: https://www.reddit.com/r/Economics/comments/14nuucw/economic_inequality_cannot_be_explained_by/jqabh4z/ https://www.reddit.com/r/Economics/comments/14nuucw/economic... Every single variable here, whether their self-reported current income as relatively young respondents (who haven't had much career/lifetime to build up large SES differences) or the very brief questions about individual cognitive biases which are then discretized to throw away even more information, is poorly measured, and there's not much of a sample to begin with. So, they predictably find that none of their measurements are correlated with each other (the biases don't even correlate with each other), and report the amazing discovery that "economic inequality cannot be explained by individual choices". They fail to reject the null (perhaps because their data is so bad?) and conclude they have proven the null.