Topics / Think clearer
The Dunning-Kruger effect
Published November 20, 2026. 2 sources at the foot of the page.
What is the Dunning-Kruger effect?
The Dunning-Kruger effect is the tendency for people with the least skill in an area to overestimate how well they are doing. The gaps that cause the mistakes are the same gaps that hide the mistakes, so a beginner can feel well informed and be far off at the same time.
Where you might meet it
- A new driver, three weeks after the license arrives, feels smoother in traffic than most of the cars around him.
- A first-time investor picks two stocks, both rise, and he starts describing a feel for the market.
- A colleague who read the documentation last week reviews your code and calls an ordinary pattern a mistake.
None of these people are pretending. The Dunning-Kruger effect is about missing information rather than bluffing, and it visits everybody in the areas they have not practiced yet.
Humor, logic and grammar
Kruger and Dunning (1999) asked people to take a test and then estimate how they had done. Across their studies the test covered recognizing humor, logical reasoning and grammar, so the skill being measured changed while the question about it stayed the same.
What each participant gave them
- A score on the test itself, marked against the same key for everyone who took it.
- That person’s own estimate of their ability and of their test performance, given before any marks came back.
Participants who scored in the bottom quartile placed both their ability and their performance well above average. Their estimates sat further from their real standing than any other group’s.
The strongest performers leaned the other way and placed themselves a little lower than their scores deserved. Kruger and Dunning explain part of that as assuming other people had done about as well as they had.
Nobody in these studies was guessing a stranger’s score. Each person was judging work they had just finished, which is what makes the size of the gap worth a second look.
The skill and the judgment come from the same place
Dunning (2011) describes this as a metacognitive problem. The knowledge you need to produce a good answer is largely the knowledge you need to recognize a good answer, so when the first is missing the second is usually missing with it.
A double burden
Somebody who has not learned the rule for a comma cannot use that rule to check a sentence. The error and the blind spot arrive together, which is how confidence stays high while the work stays weak.
Everyday life rarely settles the question for you. Dunning notes that feedback is often missing, late, or too polite to be clear, so a beginner can go a long time without meeting the evidence.
It cuts the other way too. As you learn an area you also learn how much of it you have not covered, so your sense of your own skill can fall while your skill is climbing.
The common misreading
The Dunning-Kruger effect is not “stupid people think they are smart.”
The studies compared quartiles of performance inside one skill at a time, and the bottom quartile of a grammar test is not the same set of people as the bottom quartile of a humor test.
Read it as a statement about a skill you have not built yet. Everybody is a beginner in almost everything, so this describes a position you move in and out of rather than a kind of person.
Two more things worth keeping straight
- It is a pattern measured across groups of participants. One confident person you know is not an instance of it, and one accurate self-rating does not make anybody immune.
- It is not a score. Nothing here measures you, and the Dunning-Kruger effect is not a label to hand a colleague in the middle of an argument.
Researchers have also argued about how much of the pattern comes from measuring self-estimates this way. Dunning (2011) reviews those accounts next to the metacognitive one, which is a fair way to hold any single finding.
Three ways to find out where you actually stand
Calibration is the work of bringing your sense of your own skill closer to the evidence. It asks for something outside your own head to compare yourself against, not a lower opinion of yourself.
- Ask for feedback you can act on. “Which two lines would you write differently?” gets you further than “How did I do?”
- Test yourself against a key. Predict your score, write the prediction down, then mark the work and compare the two numbers.
- Compare notes with people who have done the job longer. Their corrections show you the parts of it you did not know were there.
Kruger and Dunning (1999) ended with a training study: once people were taught the reasoning their test was built on, their view of their own earlier answers moved closer to how those answers had really scored. Learning a skill is also how you learn to judge it.
Back to the three people from the first chapter. The new driver books a lesson with an instructor who rides along, the first-time investor writes dated predictions and checks them against the market, and the reviewer says which parts of the file he cannot judge yet and asks somebody who can.
Sources
- Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing one’s own incompetence lead to inflated self-assessments. Journal of Personality and Social Psychology, 77(6), 1121-1134.
- Dunning, D. (2011). The Dunning-Kruger effect: On being ignorant of one’s own ignorance. Advances in Experimental Social Psychology, 44, 247-296.
Practice it on Learn
Learn teaches the Dunning-Kruger effect in short games and lessons, with the same research as this page.
Practice the Dunning-Kruger effectSee what Learn is
Your first lesson is free and needs no account. For adults 18 and over.
More topics
- The framing effect
The framing effect is the tendency to decide the same choice differently depending on whether its outcomes are described as gains or as losses.
- Fast and slow thinking: System 1 and System 2
Cognitive reflection is the habit of checking the answer that comes to mind first before you give it.
- The fundamental attribution error
The fundamental attribution error is the tendency to explain other people’s behavior by their character while overlooking the situation they were in.