Dunning-Kruger Effect: Definition, Examples, & Psychology​

The strings buzz because they are not holding them down firmly enough; the chords sound off because they have placed a finger incorrectly; their playing falls out of time with the music, but they do not seem to notice.

All of this exasperates me as a veteran guitar player; I worry how these players will learn to correct these issues with their technique. Years of playing have shown me the limits of my abilities: I recognize when I am not keeping the beat or when I have played notes with too much or too little force. I even exasperate myself with my own inconsistencies; I cannot help but notice them. I can tell you the three or four aspects of my playing that need the most work or are the most unpolished.

At the same time, an expert singer would listen to my singing and hear all kinds of subtle things I should change that are entirely beyond my awareness. Until we have built a serious foundation of knowledge in a category or skill, we run the risk of being ignorant of our own limitations. This phenomenon is called the Dunning-Kruger effect.

The Dunning-Kruger effect is important because it allows us to make four useful assumptions about people performing in domains in which they are not entirely competent (Kruger & Dunning, 1999):
​

1. Less competent people will overestimate their abilities in that domain. As we will read more about in a moment, the Dunning-Kruger effect actually originates in the repeated finding that when people perform badly on a task, they usually rate themselves as having done reasonably well, or even quite well, on the task.
2. Less competent people will have difficulty identifying skillfulness in others in that domain. If you do not know what competence entails in a given domain, you won’t be able to recognize it in other people. I think about this in regard to songwriting all the time: As a student of the craft, I believe people are often missing the most meaningful or interesting flourishes and innovations used by songwriters.
3. Less competent people will not recognize their knowledge or skill gaps. If I sing something and am pretty sure I hit all the right notes, I assume that I have sung the line successfully. But there are all kinds of aspects of making one’s voice sound pleasing to the ear of which I have little awareness; my voice is probably far less enjoyable than I imagine, even if my singing is technically on key.
4. If trained to be more aware and given more skills, people will recognize that they were less competent before. The Dunning-Kruger effect is expected to be widespread but also generally fixable: With education and training, we grow to see where the gaps were in our performance and what constitutes effective performance.​

​These characteristics of the Dunning-Kruger effect have widespread implications, as we often—if not always—base our behaviors on our perceived skill and knowledge in different life domains. To take one example that really stresses the importance of this gap, we can look at how partisanship affects our political knowledge (Anson, 2018). People do in fact regularly overestimate how politically informed they are, and to make matters worse, they do this more fervently when their political affiliations are emphasized.

For example, I might think I know a lot about the topic of abortion rights, but if you remind me that I have voted Democrat my whole life, I might be even more likely to express an inflated sense of my own knowledge about that hot-button topic. Just as troublingly, Anson (2018) found that people with little actual political knowledge see other people who seem similarly partisan, rather than people who seem well informed, as being experts. This is how we develop the political echo chambers and siloes that plague our polarized political system: We feel more comfortable around other people who are similarly confident in their polarized views than we do around people who show signs of being intellectually well versed in the topic.

Another important domain influenced by the Dunning-Kruger effect is that of medical science and the training of doctors (Rahmani, 2020). Self-assessments of skill and competence taken across many samples of doctors who are being trained in any way, shape, or form demonstrate that these trainees overestimate their competence across several domains. To make matters worse, some of the doctors who were rated as least competent by their medical peers were the ones to most strongly overestimate their own abilities. The more competent the doctors were, the more likely they were to accurately estimate their own abilities compared to other doctors. Some of the most talented doctors—and this is true across most domains and skills—were even likely to underestimate their abilities.

In some of their earliest experiments on this topic, Kruger and Dunning (1999) found that people routinely overestimate how logically they think, the quality of the grammar of their writing, and how funny they are. In fact, the people who scored in the bottom quarter of test subjects on these measures typically believed they were in fact in the top half of performers. Similar results have been found on any number of subjects, from debate team members thinking they are more knowledgeable than they actually are (Ehrlinger et al., 2003) to medical residents overestimating their interpersonal skills with clients (Hodges et al., 2001).

Troublingly, the Dunning-Kruger effect applies to attitudes as well as skills. For example, two studies by West and Eaton (2019) showed that the people who were highest in racial and gender bias were the strongest overestimators of how egalitarian their beliefs were. Perhaps this finding can help us understand how people caught saying blatantly racist or sexist things can protest that they are not in fact racist or sexist: They are ignorant of where they really stand on the spectrum of racial and gender bias.

As noted above, Dunning and colleagues (2003) describe their effect as coming about because people who are unskilled in a particular way cannot accurately observe the effects of being unskilled. In the absence of useful and accurate data, they cannot recognize how their behavior or performance compares to other people’s—and they default to assuming they are doing just fine, or are even doing well. This can be seen as a deficit in metacognition, or the ability to observe one’s own thinking with some psychological distance (Dunning et al., 2003).

From the start, Kruger and Dunning (1999) hypothesized that making people aware of their lack of skill, and amending that skill through training that makes them more competent, would reduce the severity of the Dunning-Kruger effect. In other words, if people can become accurate judges of competence, they will accurately assess their own skill level. Indeed, they found that being educated about one’s lack of knowledge led to much more accurate self-judgments, or even unfairly harsh ones, where there had formerly been an overly rosy perspective.​

David Dunning was a Cornell University psychology professor who read about a particularly foolish mistake a would-be bank robber had made, and he began to wonder how universal the bank robber’s overestimation of his own intelligence was. With his graduate student at the time, Justin Kruger, he developed a series of studies to test whether people’s self-perceptions matched their actual performance relative to others. Those studies, which we have already looked at, launched what became known as the Dunning-Kruger effect.

A common criticism of the Dunning-Kruger effect is that people often see themselves through rose-colored glasses. For example, in one study of players of the card game called bridge, some of the bridge players continued to overestimate their own skills even after getting expert and accurate feedback about their performance (Simons, 2013). According to this criticism, the Dunning-Kruger effect reflects a particular instance of this positive bias toward the self.

The other chief criticism of the Dunning-Kruger effect comes from people who see it as an instance of what is called regression to the mean (Burson et al., 2006). This is the idea that most people and most behaviors will tend to be near the mean, or average, for the overall population. Regression to the mean would suggest that even people who perform poorly on tasks would be more likely, on average, to see themselves as performing close to average than to see themselves as performing poorly. Statistically speaking, it is unlikely for people to be that far from the average on both measures.​

I don’t really buy these criticisms; at least, I don’t think they eliminate the usefulness of the Dunning-Kruger effect. Perhaps they explain why the Dunning-Kruger effect exists, but they do not justify throwing out the idea itself.

• Anson, I. G. (2018). Partisanship, political knowledge, and the Dunning-Kruger effect. Political Psychology, 39(5), 1173–1192.
• Burson, K. A., Larrick, R. P., & Klayman, J. (2006). Skilled or unskilled, but still unaware of it: how perceptions of difficulty drive miscalibration in relative comparisons. Journal of Personality and Social Psychology, 90(1), 60–77.
• Dunning, D., Johnson, K., Ehrlinger, J., & Kruger, J. (2003). Why people fail to recognize their own incompetence. Current Directions in Psychological Science, 12(3), 83–87.
• Ehrlinger, J., Johnson, K., Banner, M., Dunning, D., & Kruger D. (2003). Why the unskilled are unaware: Further explorations of (absent) self-insight among the incompetent. [Unpublished manuscript.] Cornell University.
• Hodges, B., Regehr, G., & Martin, D. (2001). Difficulties in recognizing one’s own incompetence: Novice physicians who are unskilled and unaware of it. Academic Medicine, 76(10), S87–S89.
• 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.
• Muller, A., Sirianni, L. A., & Addante, R. J. (2021). Neural correlates of the Dunning–Kruger effect. European Journal of Neuroscience, 53(2), 460–484.
• Rahmani, M. (2020). Medical trainees and the Dunning–Kruger effect: when they don’t know what they don’t know. Journal of Graduate Medical Education, 12(5), 532–534.
• Simons, D. J. (2013). Unskilled and optimistic: Overconfident predictions despite calibrated knowledge of relative skill. Psychonomic Bulletin & Review, 20, 601–607.
• West, K., & Eaton, A. A. (2019). Prejudiced and unaware of it: Evidence for the Dunning-Kruger model in the domains of racism and sexism.  Personality and Individual Differences, 146 , 111–119.