The methodology seems highly questionable. From the actual study (https://www.apa.org/pubs/journals/releases/psp-pspi0000187.pdf https://www.apa.org/pubs/journ

The methodology seems highly questionable. From the actual study (https://www.apa.org/pubs/journals/releases/psp-pspi0000187.pdf https://www.apa.org/pubs/journals/releases/psp-pspi0000187.p...) "Social class was measured in three ways. To indicate subjective social class, participants saw an image of a ladder that 'represents where people stand in your country.' and selected the rung on which they felt they stood, relative to other people in their country" "However, when difference scores were used (or the alternative joint-testing procedure), the conclusions were not so straightforward. Here, we found support for Hypothesis 1 only in the case of subjective social class. The difference score approach also showed that education and income were negatively associated with overconfidence, directly contrasting the results generated by the residual score approach." So it's not so much that people from a higher social class are overconfident but that overconfident people appear to be from a higher social class due to their methodology and downstream assessment confirms their methodological bias. It's quite incredible this received as much attention as it did. Also this sample is limited to Mexican small-business owners applying for loans.