The Dunning–Kruger Effect: Confidence and Performance on a Task

On a new assignment, you may think, “That should be quick,” before checking all the requirements. Someone who has done similar work many times may worry about small ways the result could go wrong. It is tempting to conclude that beginners are always confident and experts always doubt themselves. But confidence alone cannot tell us how capable someone is or how well they understand their own skill.

The Dunning–Kruger effect is the name often given to a pattern in which people who perform poorly on a particular task rate their ability higher than it is relative to others. It is not a universal rule that “the less you know, the more confident you always are.” Impostor feelings describe difficulty internalizing accomplishments and fear of being exposed as inadequate. They are not an official diagnosis, and the two ideas are not opposite personality types.

Confidence and performance are not the same thing

When you are new to a skill, you may not yet know all the complicated parts. You might estimate that a task will take less time than it actually does or expect a better result. That does not mean you overestimate yourself in every area. A beginner cook might think a new recipe looks easy while accurately judging their ability to prepare a work report.

An experienced person may speak cautiously because they know the different ways a result could turn out. Caution does not necessarily mean low confidence or an inability to recognize one’s skill. Someone with a strong track record can feel anxious before a presentation, while a beginner can judge their current level accurately by preparing and using feedback.

The Dunning–Kruger effect became widely known through a 1999 study comparing participants’ performance and self-ratings on particular tasks. Later researchers have argued that patterns can vary with the task and the way ability is measured. They have also discussed how score boundaries and statistical regression could produce a curve that looks like overconfidence among people with lower scores. This does not mean every result from the original research was wrong, but a famous graph cannot rank people’s general ability. The original paper is Kruger and Dunning (1999), Unskilled and unaware of it. For a later analysis of score boundaries, see Magnus and Peresetsky (2022), A Statistical Explanation of the Dunning–Kruger Effect. That analysis uses statistics-exam data; it does not establish that psychological explanations are unnecessary for every task.

Impostor feelings are not the same as “a capable person being humble”

The phrase “impostor phenomenon” is used for an experience in which someone has trouble accepting their accomplishments as their own, attributes success to luck or outside help, and worries that others will discover they are not good enough. After a promotion or positive review, they may think, “I just got lucky this time,” and then worry about making a mistake on the next task. An early account is Clance and Imes (1978), The imposter phenomenon in high achieving women. Clinical observations from that selected group of women cannot simply be generalized to everyone.

These feelings are not limited to high achievers, and experiencing them does not prove that someone is excellent by every standard. Work and school settings, evaluation practices, belonging, experiences of discrimination, and personal anxiety may all matter. Estimates of how many people experience impostor feelings vary widely depending on the questionnaire and cutoff used. Instead of diagnosing someone from an online score—or declaring, “You’re actually talented, so don’t worry”—it is more precise to look at when self-doubt grows. Variation across questionnaires and thresholds is discussed in Bravata et al. (2020), Prevalence, Predictors, and Treatment of Impostor Syndrome: a Systematic Review. The task records and feedback examples in this article are reflection suggestions, not a validated treatment protocol.

It is also worth being careful about teasing a confident person by calling them Dunning–Kruger or labeling someone who doubts themselves as an impostor. Putting the ideas at opposite ends ignores how one person’s confidence can shift by setting. Someone may feel at ease at work but anxious in a close relationship, or confident with a familiar skill and uncertain in a new field.

Look at your ability and a colleague’s more concretely

If you want to assess your performance, replace “Am I good at this?” with a specific task and clear standards. For a presentation, you could consider accuracy, staying within the time limit, and answering questions from the audience. Putting the actual result beside your prediction can show what to practice next time.

When asking for feedback, try “Was I good?” or “What did you think?” less often. Ask, “Was the main point clear?” or “What should I work on first next time?” Specific, observable feedback is information for the next action, not a verdict on your worth. Feedback from one person is not a final answer either; consider different tasks and points of view.

If you tend to rate yourself too low and cannot accept success, write down the outcome separately from the story you tell about it. “I finished the project” is a fact; “Everyone else helped, so I did nothing” is an interpretation. The fact that someone helped does not prove you made no contribution. Identify what preparation and choices actually mattered.

Rather than trying to boost confidence all at once, choose a small task, do it, and review the result. You could ask one question in a meeting or share an early draft for comments. Write down one thing that went well and one thing to change next time. Use your feelings, actual results, and clear standards together to form a more grounded view.

When self-doubt keeps getting in the way

The Dunning–Kruger effect and impostor feelings are not diagnostic tools for classifying people. Do not decide your ability from one confident moment or one mistake. If persistent self-doubt makes you avoid opportunities or leads to anxiety and exhaustion that disrupt your daily life, you can talk with a professional about that distress itself.

A popular label cannot replace your experience. A more useful starting point is to ask what you did on a specific task, what the standards were, and what you want to learn next.

Talk about confidence can easily turn into teasing. Asking lots of questions does not prove someone is unprepared, and a firm tone does not prove they lack skill. When you work together, share the standard and the result you need instead of guessing. For example, rather than “Can you handle this?”, ask, “Could you review the user flow and error handling in this draft by Wednesday?”

Revision note

October 20, 2026: Connected the original research, later statistical analysis, and impostor-phenomenon sources to the relevant claims, with limits on samples and application.

Sources

• Bravata et al. (2020), Prevalence, Predictors, and Treatment of Impostor Syndrome: a Systematic Review
• Kruger and Dunning (1999), Unskilled and unaware of it