John Cleese, the British comedian, once summarizes the idea of the Dunning-Kruger effect as, “If you are really, really stupid, then it is impossible for you to know that you are really, really stupid.” A quick news search brings up dozens of headlines connecting the Dunning–Kruger effect to everything from work on empathy and even why Donald Trump was elected president.
As a mathematics professor who teaches students to use data to make smart decisions, I am familiar with the common mistakes people make when dealing with numbers. The Dunning-Kruger effect is the idea that the least skilled people enhance their abilities more than others. It sounds convincing on the surface and makes for excellent comedy. But in a recent paper, my colleagues and I suggest that the mathematical method used to demonstrate this effect may be wrong.
What Dunning and Kruger showed
In the 1990s, David Dunning and Justin Kruger psychology professors at Cornell University and wanted to test whether incompetent people do not know their inadequacy.
To test this, they gave 45 undergraduate students a 20-question logic test and then asked them to rate their own performance in two different ways.
First, Dunning and Kruger asked students to estimate how many questions they got right — a straightforward assessment. Then, Dunning and Kruger asked the students to estimate how they compared to other students who took the test. This type of self-assessment requires students to guess how others are doing and is prone to a common cognitive error – most people consider themselves to be better than average.
Research shows that 93% of Americans think they are better drivers than average90% of teachers think they are more skilled than their peers, and this overestimation is prevalent in many skills – including logic tests. But it is mathematically impossible for most people to be better than average at a task.
After giving the students a logic test, Dunning and Kruger divided them into four groups based on their scores. The lowest quarter of students got, on average, 10 out of 20 questions right. In comparison, the top-scoring quarter of students got an average of 17 questions right. Both groups estimated that they got about 14 correct. This is not a terrible self-evaluation of any group. The least proficient overestimated their scores by about 20 percentage points, while the best performers underestimated their scores by roughly 15 points.
The results appear even more surprising when looking at how students rate themselves against their peers, and this is where the better-than-average effect is in full force. display. Students with the lowest scores estimated they did better than 62% of test takers, while students with the highest scores thought they did better than 68%.
By definition, being in the bottom 25% means that, at best, you get better than 25% of people and, on average, better than 12.5%. Estimating that you did better than 62% of your peers, while only scoring better than 12.5% of them, gives a whopping 49.5 percent overestimation.
The measure of how students compare themselves to others, rather than their actual grades, is where the Dunning–Kruger effect arises. This greatly overstates the overestimate of the lower 25% and seems to indicate, as Dunning and Kruger titled their paper, that the The least proficient students are “unskilled and uninformed.”
Using the protocol designed by Dunning and Kruger, many researchers have since “confirmed” this effect in their own field of study, which led to the feeling that the Dunning–Kruger effect is intrinsic to how the human brain works. For everyday people, the Dunning-Kruger effect seems real because the overly arrogant fool is a familiar and annoying stereotype.
Eliminating the Dunning-Kruger effect
There are three reasons that Dunning and Kruger’s analysis is misleading.
The worst test-takers will also overestimate their performance because they are the furthest from getting a perfect score. Furthermore, the least skilled people, like most people, think they are better than average. Finally, the lowest scorers were less severe in estimating their goal achievement.
Establishing the Dunning-Kruger effect is an artifact of research design, not human thought, my colleagues and I have shown that it can be done. using randomly generated data.
First, we created 1,154 fictional people and randomly assigned them a test score and self-assessment ranking compared to their peers.
Then, as Dunning and Kruger did, we divided these fake people into quarters based on their test scores. Since the self-assessment rankings are also randomly assigned a score from 1 to 100, each quarter returns to the mean of 50. By definition, the lowest quarter exceeds only 12.5% of the participants. on average, but from the random assignment of self-assessment scores they consider themselves better than 50% of the test-takers. This provides and overestimated by 37.5 percentage points no one person is involved.
To prove the last point – that the least skilled can judge their own skill – requires a different approach.
My colleague Ed Nuhfer and his team provide students with 25-question scientific literacy test. After answering each question, students rate their own performance on each question as “nail it,” “not sure” or “no idea.”
In working with Nuhfer, we found unskilled students very good at estimating their own competence. In this study of unskilled students who scored in the bottom quartile, only 16.5% overestimated their abilities. And, it turns out, 3.9% lowered their score. That means that about 80% of unskilled students are very good at overestimating their true ability – a far cry from the idea put forward by Dunning and Kruger that unskilled people often overestimate their abilities.
The original paper by Dunning and Kruger begins with the quote: “It is one of the essential features of incompetence that the person inflicted is unable to know that they are incompetent.” This idea has spread far and wide through scientific literature and pop culture alike. But according to the work of my colleagues and I, the truth is that very few people are truly unskilled and uninformed.
Dunning and Kruger’s experiment found a real effect – most people think they are better than average. But according to my team’s work, that’s just what Dunning and Kruger showed. The truth is that people have an innate ability to measure their competence and knowledge. on another’s claim suggests, wrongly, that most of the population is hopelessly ignorant.
Eric C. Look is Senior Lecturer in Mathematics, Bowdoin College.
This article was reprinted from The Conversation under Creative Commons license. Read the original article.