Are the least skilled really the most confident?

Only partly, and the popular version gets it wrong. The Dunning–Kruger effect, studied by psychologists David Dunning and Justin Kruger in 1999, is the tendency of people with low ability in a specific area to overrate their ability in that area. In the original studies, the students who scored in the bottom quarter had real scores around the 12th percentile, yet ranked themselves around the 62nd. But the least skilled were not the most confident: they overestimated themselves, while still feeling less confident than the top performers. And the effect is about specific skills, not intelligence in general.

It started with a bank robbery. In 1995, McArthur Wheeler robbed banks near Pittsburgh without a mask, his face covered in lemon juice, convinced it would make him invisible to cameras. He had even tested it with a Polaroid, and he wasn't in the picture. When police showed him the surveillance photos, he was stunned: "But I wore the lemon juice." The case directly inspired the research.

Why the effect appears is still debated. Dunning and Kruger's explanation is that telling good work from bad is part of the skill itself, so beginners can't see their own mistakes. Critics say much of the pattern is statistics. Self-ratings and real scores are only loosely linked, so people picked for very low scores will tend to rate themselves higher, a regression to the mean. Add the fact that nearly everyone thinks they're above average, and you get "noise plus bias", which some argue explains most findings. Dunning replies that the misjudgements are real, whatever their cause.