The Dunning–Kruger effect is often used online to describe someone who knows little but is loudly confident. That shorthand is much broader—and harsher—than the research can justify. The original studies examined how people’s self-assessments related to their performance on specific tasks, not a permanent type of person.
What the original studies examined
In 1999, Justin Kruger and David Dunning reported that participants who scored lower on certain tests tended to overestimate their performance relative to their peers, while higher-scoring participants sometimes underestimated theirs. The researchers proposed that some skills needed to perform well also help people recognize errors in their own performance.
That finding has generated debate. Later methodological work has argued that some apparent patterns can be produced or exaggerated by statistical features such as measurement error, regression toward the mean, and how groups are compared. Other work continues to examine when calibration differences occur. So the effect should be treated as a research question with nuance, not as a shortcut for dismissing someone.
A label is less useful than feedback
People at every skill level can misjudge what they know. Confidence depends on experience, task difficulty, feedback, and the way a question is framed. Calling someone “Dunning–Kruger” rarely helps them learn; specific evidence does.
- Compare your work with clear criteria or examples.
- Ask for feedback on one observable part of the task.
- Check whether your confidence matches your accuracy over several attempts.
- Update your estimate when new evidence appears.
Use calibration without shame
If you are new to a skill, uncertainty is expected. If you have experience, confidence can be useful but still deserves checks. A better question than “Am I overconfident?” is “What evidence would help me estimate my current skill more accurately?”
For a related look at confidence and action, read self-efficacy: building confidence to take the next step. If success still feels undeserved despite evidence, our article on impostor feelings explores a different confidence-related experience.
Takeaway
The Dunning–Kruger label is not a reliable diagnosis of another person’s intelligence or character. The useful lesson is more modest: self-assessment can be imperfect, and careful feedback helps people calibrate.
Sources: Kruger and Dunning (1999), “Unskilled and Unaware of It”; Gignac and Zajenkowski (2020), “The Dunning–Kruger Effect Is (Mostly) a Statistical Artefact.”
Selected for this topic
Useful next reads
Four editorially reviewed guides chosen for relevance, not at random.
Habits & decisionsHow to Make a Difficult Decision When You Keep Going Back and Forth
Habits & decisionsHow to Tell Your Manager You Have Too Much Work: A Practical Plan
Habits & decisionsHow to Restart a Habit After a Break: Resume, Resize, or Let It Go
Habits & decisionsHow to Adjust a Goal When Your Circumstances Change