Unskilled individuals overestimate their competence; experts may underestimate.
Part of the pattern is a statistical artefact — noisy self-assessment plus regression to the mean produces the same graph with no metacognitive deficit at all. The narrower claim, that skill is needed to assess skill, survives.
In practice
- A new manager three weeks into the role explains the department's problems to the person who has run it for eleven years.
- An expert declines to give a firm answer where a novice would be certain, and the novice sounds more competent in the meeting.
- Possible explanations
- One explanation is metacognitive: the skills needed to do a task well overlap with those needed to judge one's own performance at it. Another is statistical, since noisy self-assessment plus regression to the mean produces the same chart with no deficit at all. A third is the better-than-average effect, which raises almost everyone's self-rating.
- In its favour
- The calibration gap between confident novices and cautious experts is found repeatedly across domains, and the narrow claim that judging skill requires skill is well supported.
- Against it
- Simulations reproduce the classic chart from noise and regression alone, so the popular version, that incompetence causes confidence, goes well beyond the evidence.
- Standing
- Part of the pattern is a statistical artefact of noisy self-assessment; the metacognitive claim is narrower than popular use. Compare illusory superiority.
- Further reading
- Wikipedia
- Subject
- Psychology & Behavior
- Essay
- A century and a half of measuring the mind