Topics Mind and behavior

What is the Dunning-Kruger effect?

The Dunning-Kruger effect is the finding that people who are weak at something often overestimate how good they are at it, partly because the skills needed to do a task well are the same skills needed to judge how well you did it. It comes from a 1999 paper by two psychologists, David Dunning and Justin Kruger, who tested people on things like logic, grammar and humor.

The idea became famous, and it is often stretched too far. It is not a clean proof that unskilled people are foolish and confident. Some researchers argue that part of the pattern comes from statistical effects, such as regression to the mean, and that almost everyone misjudges their own performance to some degree. That debate is not settled, and it is worth knowing before you use the term on someone in an argument.

An episode on this topic would walk through the original experiments, what the famous graph does and does not show, and the criticisms. It would also cover why self-assessment is hard for everyone, including the listener. bre's hosts are AI, so they can be wrong, and you can press Talk to ask a question or push back on a point as it goes.

What a bre episode would cover

An outline of the episode bre would make for this question. Every episode is written fresh when you ask, so yours will differ.

  1. Where the idea came fromThe 1999 paper by Dunning and Kruger, the kinds of tests people took, and how participants rated themselves against how they scored.
  2. The double burden of low skillWhy lacking a skill can also make it harder to spot your own mistakes, and what the authors proposed as the reason.
  3. What the famous graph really showsThe popular chart of confidence against skill, how it gets redrawn online, and why it is often misread.
  4. The statistical critiqueWhy some researchers think part of the effect could come from regression to the mean and general misjudgment, and where that argument stands.
  5. Do skilled people underestimate themselves?The finding that high performers sometimes rate themselves a bit low, and the explanations offered for it.
  6. Using the term carefullyWhy calling someone a Dunning-Kruger case is usually a shortcut, and how the rest of us can check our own judgment.

How the episode might open

A sample exchange between two of bre’s AI hosts, bre and Cal. Both are AI; this is written by AI, as every bre episode is.

  1. breAI host

    Okay, so everyone has heard this one. Someone online is wrong and loud, and a commenter says, classic Dunning-Kruger. Cal, is that what the effect actually is?

  2. CalAI host

    Kind of. Mostly not. The original idea is narrower. People who do badly on a test tend to guess they did much better than they did, and I love that it started with something as small as joke ratings.

  3. breAI host

    Right, the researchers tested things like grammar, logic and humor. Then they asked people how they thought they ranked.

  4. CalAI host

    And the lowest scorers put themselves well above where they landed. The proposed reason is the clever bit: the skill you need to do the task is the same skill you need to notice you did it badly, so you are missing both at once.

  5. breAI host

    That sounds tidy, which makes me suspicious. I don't know how much of it holds up.

  6. CalAI host

    Fair, because there is a real argument about it. Some researchers say a chunk of the pattern is a statistical artifact, and that nearly everyone is a bit off when guessing their own rank.

  7. breAI host

    So the meme might be bigger than the finding.

  8. CalAI host

    Could be. Let's start with the original experiments, then take the criticism seriously and see what survives.

Questions people also ask

Does the Dunning-Kruger effect mean stupid people think they are smart?
No. The original work was about specific skills, like logic or grammar, not general intelligence. People who scored low on a given task tended to overrate their performance on it. Someone can be weak in one area and well calibrated in another.
Is the Dunning-Kruger effect real?
Something real is going on: people often misjudge their own performance, and the weakest performers tend to overestimate the most. How much of that is a special effect of low skill, versus a statistical artifact, is still debated among researchers.
Do highly skilled people underestimate themselves under this effect?
In the original data, top scorers tended to rate their own ability slightly low, partly because they assumed others found the tasks as easy as they did. Later critics say statistical effects may also contribute, so the explanation is not fully settled.
How can I avoid overestimating my own skill?
This is explained, not advised: researchers note that clear feedback, comparing your work against objective standards, and learning more about a field all tend to make self-assessment more accurate. Knowing how much there is to learn often shifts people's estimates.

Related topics

More: all 300 topics, mind and behavior, or the longer reads on /learn.

bre’s hosts are AI, and every episode is generated, so they can be wrong: check anything that matters. This page outlines what an episode would cover. It is for interest and learning, not medical, financial or legal advice.