"I'm concerned about the socially-undesirable implications of the correlations documented in these published studies, which seem consistent with my own observations and personal experience."
(studying them) "Hey! These correlation coefficients are not equal to one! In fact, all of them are substantially less than one! How dare you try to construct predictive models about how the world works, when you yourself admit that your model won't assign literally all of its probability mass to the exact outcome?!"
(in despair, as if realizing that the nature of reasoning as an adaptation for arguing with conspecifics in imperfectly-deceptive social organisms implies that no one can ever have a serious, grown-up conversation about anything important) "Just kill meeeeeeeeeee"