04 MAY 2022
The Dunning-Kruger effect is a purported bias in human thinking that leads us to misjudge our abilities in contrast to our actual skill level.
It is named after American social psychologists David Dunning and Justin Kruger, who since the late 1990s have conducted a series of investigations into our ability to self-assess our own performance at various tasks such as logical reasoning, grammar, emotional intelligence, and humor.
What does the Dunning-Kruger effect mean?
According to their results, we all overestimate our ability to succeed at particular tasks when our actual skill in that task is relatively low, and in some cases underestimate our chance of success when our skill is high.
This tendency to overestimate, they’ve found, stems not from a mistaken assessment of others’ abilities, but due to a lack of insight into our own. By improving our ability to judge the accuracy of information, we can evaluate our own knowledge better and estimate our own abilities more reliably.
How is the Dunning-Kruger effect misunderstood?
Since the dawn of Western philosophy, one of the marks of wisdom has been humility in knowledge. In the words of 20th century British philosopher Bertrand Russell, “the fundamental cause of the trouble in the modern world today is that the stupid are cocksure while the intelligent are full of doubt.”
So it comes as no surprise that the Dunning-Kruger effect would be so readily adopted in reflection of this age-old truism. The results are often interpreted as generalizations of intelligence along a dividing line – between the smart and stupid, so to speak.
It’s often misinterpreted as “overconfidence in oneself“, or as a graph portraying an expert-level confidence among the incompetent.
But Dunning and Kruger’s investigations apply less to individuals and more to discrete fields of knowledge and skill. We’re all likely to overestimate our ability to recall or identify factual information, or exercise a skill, as our actual performance in that area drops below the top quartile.
That isn’t about ‘incompetent’ folks, but about most people in any given expertise. You can have a high IQ, know all there is to know about neurosurgery, have zero handicap on your golf game, and cook a mean lasagna, and still be subject to the Dunning-Kruger effect when discussing domestic politics.
Is it possible the Dunning-Kruger effect isn’t even real?
Recent investigations into the method used by Dunning and Kruger to analyze their data have raised cause to be suspicious, if not doubtful of the strength of the Dunning-Kruger effect.
By producing pools of data in a completely randomized fashion and then separating the test and prediction figures into the quartiles (the bottom performers, the top performers and two blocks in the middle) used by Dunning and Kruger, several researchers have demonstrated it’s possible to get the same results.
This and other criticisms imply the effect might be the result of a statistical artifact called a ‘regression toward the mean‘ – a natural consequence of randomness and repeated sampling.
This doesn’t rule out such an effect in specific areas. It does leave plenty of room for debate over the nature of self-awareness in our own knowledge.
Even if the Dunning-Kruger effect was found to be a complete illusion that age-old advice of reserving doubt would still be worth keeping in mind.
