Biases and effectsDunning and Kruger, 1999
The less skilled you are, the more you overrate yourself. Famous — and the "mountain" chart everyone shares is not in the paper.
Participants took tests of logic, grammar and humour, then estimated how well they had done.
The bottom quarter rated themselves around the top third. The top performers slightly underrated themselves.
The claim: the unskilled lack the very knowledge needed to recognise that they are unskilled, because spotting the error requires the same skill as avoiding it.
The viral chart — "Mount Stupid → Valley of Despair → Slope of Enlightenment" — appears nowhere in the paper. Someone drew it later and it spread.
Worse, the same pattern shows up in random numbers. Everyone rates themselves "a bit above average", so the gap between self-rating and score is automatically larger at the low end. Part of the effect is arithmetic, not psychology.
The effect has not been thrown out, but "incompetent people are confident" is not what the evidence says.
Even so, one lesson holds: you can only see how much you did not know once you know more.
If something feels "surprisingly easy" on day three, it is usually not because it is easy but because the hard parts are not visible yet.
The reverse holds too: getting anxious after improving is what seeing the whole picture feels like. Anxiety at that point is not a step back.
"This is easy" on day three
Cooking, an instrument, coding. The first week's "I have got this" is almost always said just before the hard part — which you need some skill to even see.
How many drivers rate themselves above average
In some surveys over 80% say they are better than average, which is impossible. But note: this is everyone drifting a little upward, not the unskilled specifically. That is exactly the disputed part.
The anxiety that arrives with skill
Six months in, feeling "I am nowhere near good" is proof you are far better than in week one — you can now see gaps that were invisible before.
If you take one thing
Anyone using the mountain chart to mock someone is a decent illustration of the effect itself. The more widely a story circulates, the more it is worth checking the source.
Related