The Dunning-Kruger Effect

| November 25, 2020 | 0 Comments

In 2000, social psychologists David Dunning and Justin Kruger published a research paper that made them famous. The paper was titled “Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments.” It details several studies the authors conducted, and how they reached the conclusion that incompetent individuals are often incapable of recognizing their own incompetence. At first, their paper was not widely noticed, but in recent years it has gained significant popularity. The human tendencies Dunning and Kruger describe have been referenced in explaining numerous errors in human judgement, even significant events like wars and stock market crashes. Let’s take a brief look at these studies and see what we can learn.

The Studies

Dunning and Kruger conducted four different studies on a combined total of 334 undergraduate students at Cornell University in New York. One study tested a social skill (how good a group of students were at recognizing humor), one tested English grammar skills, and two tested logical reasoning abilities. In every instance, the subjects were given a test with right and wrong answers, and then were asked some questions afterward. The questions were roughly: (1) How well do you think you did on the test? and (2) How do you rate your skill in this field as compared to the average student? The conductors of the study graded the students’ tests and observed how well the test scores correlated with the answers the students had given. They found that the students tended to be quite inadequate at judging their own abilities, especially in relation to their peers. They were somewhat accurate at predicting their own test scores, but not accurate at all in assessing how their scores compared with others. The more incompetent a person was (the lower their actual test score was) the farther off the mark they were in their estimation of their own abilities. These findings were remarkably consistent across all four studies that were conducted.

Figure 1 (below) is an example of the results of these studies. The data in this chart is taken from the fourth study (testing logic abilities), but the data from the three prior studies was remarkably similar. The orange line (with the squares) tracks the actual test scores of the students. The blue line (with the diamonds) marks the ability these same students perceived themselves as having in relation to the average student, average being a 50 percentile ranking. The students that performed in the bottom quartile, with an average percentile ranking of 12, thought that they ranked at the 55 percentile instead. Students in the second quartile had average test scores with a percentile ranking of 32, but thought they scored around 59. Third quartile students performed at 63 and perceived themselves as performing at 67. Finally, top quartile performers underestimated their abilities, assessing themselves to be at 78 when they really scored in the 90th percentile.


Chart created by author with data taken from Dunning & Kruger (2000)

It is interesting that the incompetent students did not rate themselves higher than the competent students rated themselves. There is a direct correlation between the students’ abilities and how they ranked themselves. However, incompetent students were more likely to overestimate their abilities, and the more incompetent they were, the more pronounced this overestimation was. In addition to not being able to recognize their own incompetence, the incompetent ones were not able to recognize competence in other people. In another portion of the study, Dunning and Kruger gave groups of bottom-scoring students and top-scoring students a packet of other students’ tests to “grade.” The bottom-scoring students were able to see tests of people who had done much better than they, but seeing these better-scoring tests did nothing to change their perception of how they themselves had performed. On the flip-side, top-scoring students who were able to see the tests of some of their less-competent peers then correctly revised their perceived ability upward to more closely match their actual ability.

Dunning and Kruger knew that this apparent tendency of incompetent people overestimating their ability and competent people underestimating theirs could simply be a result of “regression to the mean.” That is, the majority of people may tend to perceive themselves as “around average” and their self-assessment reflects this, whether they were actually competent or incompetent. In order to ensure that “regression to the mean” was not entirely responsible for the phenomenon they were observing, they conducted a follow-up to study 4. Students who had been tested on logic in Study 4 were then given a crash course in logical reasoning which would have improved their performance had they received the information prior to the test. After the crash course, students were asked to retrospectively rate the logic ability they had when they took the test. Students who had been quite incompetent on the test were able to correctly revise their abilities downward because they now realized how badly they had done. The new skills they had just gained in the logic crash course were some of the tools they needed to recognize how badly they had done on the test. This newly-earned competence allowed them to recognize their prior incompetence. Thus, the researchers proved that regression to the mean was not entirely responsible for erroneous self-assessment.

The Conclusions

Dunning and Kruger’s conclusion was that incompetent people have a tendency to self-assess their abilities as being much higher than they actually are, giving them a false confidence in their knowledge and performance. Furthermore, they are incapable of recognizing their own incompetence and incapable of judging competence in others. Incompetent people lack the metacognition (thinking about thinking, knowing about knowing) abilities required to be a good judge of one’s own abilities or of another’s abilities. This shortfall in self-assessment has come to be known as the Dunning-Kruger effect, and numerous studies conducted since have confirmed the findings of these original researchers.

Dunning and Kruger ended their paper with a few hypotheses about this interesting self-assessment behavior. They thought that in order for people to overestimate their capabilities they had to satisfy a minimum threshold of knowledge or experience in a particular field. Feeling empowered by what little they know, they inflate the knowledge in their possession, not realizing how much more they have yet to learn. But these same people would not misjudge their capabilities in an area in which they were completely incompetent. So a beginning psychology student might have an inflated view of his knowledge and abilities in the field of psychology, but he would not suffer the same delusion in the field of engineering where he has absolutely no experience. Additionally, Dunning and Kruger hypothesized that the effect they observed was most pronounced in fields of expertise where knowledge about the field makes one capable in the field. Knowing a lot about math makes you good at math. Knowing a lot about grammar makes you good at grammar. Knowing a lot about logic makes you good at logic. It is these fields where the Dunning-Kruger effect comes into play. The same delusions do not occur in fields where knowledge is disconnected from skill, such as sports.

Dunning and Kruger noted that the solution to inflated self-assessment was increased competence. Incompetent people could not judge their abilities correctly, while competent people were much better at it. Therefore, the way for incompetent people to become better at judging their own abilities is to increase their competence. This is supported by the study they performed in which individuals incompetent at logic became better at judging their own ability after a crash course in logical thinking.

The final item to be noted from Dunning and Kruger’s report is the phenomenon of competent people underestimating their own abilities. Dunning and Kruger attribute this to the false-consensus effect. The competent people knew they had performed well and assumed that everyone else had found the task equally performable. It was only after they had been presented with evidence to the contrary—the inaccurate answers of their peers—that they recognized how outstanding their own performance had been.

Does This Matter?

So is there any benefit in knowing about the Dunning-Kruger effect? Should knowledge of it affect my behavior? I find it interesting, but is it only a few fascinating but useless findings about human behavior? Or can I learn something beneficial from it? As I ponder the Dunning-Kruger effect, I can think of several ways I should live differently knowing I am susceptible to inflated self-assessment.

It would be easy to take these findings and say something like, “Well, that explains why there are so many idiots in the world. They go on being idiots because they don’t even recognize their own idiocy.” This is a wrong response. One thing that Dunning and Kruger noted and I believe to be true is that everyone has areas in which they are competent and areas in which they are incompetent. It’s not that the world population is divided between the competent and the incompetent; each person is a mix of competency and incompetency. That means every single one of us could fall prey to the Dunning-Kruger effect—you and I and Sam down the road. Those same people that I view as incompetent may very well be viewing me as the same.

Knowing that I have areas of incompetence should make me cautious about how capable I view myself as being. Hopefully there are areas in which I am competent, but, as Dunning and Kruger demonstrated, I am not the best person to make that judgement. If I am incompetent in a certain skill or field of knowledge, I should not be letting on that I am competent. This is surprisingly easy to do. I can think of times when I have read a book introducing me to a particular field of knowledge. At the end, I feel empowered! I have learned “so much” from this book and now I have above-average knowledge in this field and can hold my own in a discussion about the topic! But really? Of course not! There is no way that one book has made me competent in a field. I have crossed a minimum threshold of knowledge in the field and feel competent, but I don’t realize how much there is yet to be learned or how much more knowledgeable other people may be about this subject than I. In fact, the Dunning-Kruger effect ensures that I have difficulty even discovering who the knowledgeable people on the subject are.

Think of the negative potential for this. I read a few news articles and suddenly I am an expert at politics, know exactly who the next president should be and why, and try to convince others why I am right. I take a two-hour short course and immediately feel empowered to counsel my friend who has struggled with depression for years. I read a book and listen to a sermon, and now I know the answers to the problems our church brotherhood has been struggling with recently. So I go around feeling, acting, and talking like an expert when, in reality, I am sorely incompetent. Wouldn’t it be better for me to keep quiet and let the truly experienced and knowledgeable individual speak into the situation?

Does this mean we should always live in doubt of our own competence? Can we “know that we know” anything? There is certainly a place for confidence in our knowledge and abilities. However, this confidence must be established in the correct way.  Dunning and Kruger suggest that the reason incompetent people believe in their supposed competence is because they use a top-down approach in arriving at this conclusion. They start with the preconceived notion “I am knowledgeable” and then use this belief to estimate how well they are doing in a specific field. Instead, a bottom-up approach should be used. I should ask myself, “How have I performed in this area in the past? How do I do when my knowledge in this area is tested? Do I have a solid basis for assuming my competence?” Only after sufficient evidence has been gathered and analyzed should I make the statement, “I am competent.”

A second thing the Dunning-Kruger effect should teach me is to have patience with other people. Maybe I am a truly competent person (this having been established by objective analysis), but I am continually annoyed by people who act like they know what they are talking about, but, from my vantage point of competence, I can see quite clearly that they don’t. It is easy to attribute this to arrogance on the part of the incompetent person, but that may be an unfair judgement. There is a maxim that goes like this: “Never attribute to malice that which is adequately explained by stupidity.” I realize that this is a bit harsh, but there is some truth to it. It is easy to assume bad intent when someone is running his mouth without really understanding the subject he is talking about. However, it is more likely that he is ignorant of his ignorance. He needs someone to gently point out the truth to him, not judgement and criticism.

Confucius once said, “Real knowledge is to know the extent of one’s ignorance.” Confucius had plenty wrong, but he’s on to something here. Maybe you are one who is considered competent in a field. If you are, you probably realize how much more can be learned. Albert Einstein said, “As our circle of knowledge expands, so does the circumference of darkness surrounding it.” There are vast fields of knowledge that you and I have never explored. And as we push the limits of our personal knowledge further, we realize our ignorance afresh as we look at what knowledge we have yet to gain to truly become an expert.

Some Takeaways

We are all incompetent in something. Even if we are competent in numerous areas, there is so much more we can learn. This should encourage humility in all of us. This humility should characterize our interaction with others, whether we are doing something that demonstrates our competence or keeping quiet about something we know little about. True humility will be an antidote to erroneous self-assessment.

Keep learning! A second antidote to inaccurate self-assessment is increased competency. Dunning and Kruger demonstrated this in their study and I have found it to be true for myself. My time is limited and should be used wisely, but if any subject is truly worth discussing with others, then it is worth increasing my knowledge on that subject. Keep reading good books. Listen to knowledgeable people talk about the subject. Get some hands-on experience. Learn from those who already have experience. Grow in competence in the skills and knowledge that matter to you and that will further your service to God and His Kingdom.

~Leonard Hege

Sources:

Kruger, Justin, and Dunning, David. (2000). “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. 1121-34. 10.1037//0022-3514.77.6.1121.

Category: Public

Leave a Reply