Self-assessed intelligence is a self-esteem variable, not a test proxy measure: The relationship between personality, self-estimated and test-derived intelligence

Are people good at estimating their own intelligence? A new study says no.

Not only are people bad at judging how smart they are, their estimates are more strongly related to personality traits, like conscientiousness and risk approach, than IQ. Also, men rate themselves more highly than females, and college degree holders rate themselves more highly than non-degree holders.

The authors’ conclusion is clear: “…people are not good at estimating their test-derived intelligence, which is partly a function of their demography and personality” (p. 5). Based on this study and prior research, the authors argue that asking someone how smart they are measures their self-confidence—and not their IQ.

Link to original post: https://x.com/RiotIQ/status/2012267332588953629?s=20

Link to full article: https://doi.org/10.1016/j.paid.2025.113457

This makes total sense honestly. People’s self-ratings reflect confidence way more than actual ability. The fact that men rate themselves higher and educated people rate themselves higher shows it’s about self-perception, not accuracy. I’ve met plenty of average-IQ people who think they’re geniuses and genuinely smart people who underestimate themselves. Personality matters more than actual cognitive ability when you’re judging yourself. Kind of humbling to realize most of us have no idea where we actually stand.

Yeah, people are terrible at this. Self-estimated intelligence is basically just measuring how confident you are, not how smart you actually are. The gender difference is interesting, men overestimate more which tracks with general confidence patterns. And of course having a degree makes you think you’re smarter even if IQ stayed the same. This is why actual testing matters. You can’t just ask someone “how smart are you?” and expect a useful answer. It’s all filtered through ego and self-esteem.