A Review on Dunning-Kruger effect
We are all familiar with the “Dunning-Kruger effect,” also known as the DK effect or the “frog in a well” phenomenon, which is considered to be a cognitive bias. Simply put, the Dunning-Kruger effect suggests that people with a lack of ability have an illusory sense of superiority, mistakenly believing they are more competent than they really are. This theory was proposed by social psychologists David Dunning and Justin Kruger from Cornell University in their 1999 paper in the Journal of Personality and Social Psychology.
Although I am neither a psychologist nor a behavioral scientist, I have observed less knowledgeable people brimming with confidence, while experts tend to be very humble. However, I have always found it somewhat unbelievable: are the most ignorant people really the most confident?
A recent study has found that it is actually people with a medium level of knowledge who are most likely to be overconfident.
This finding seems more reasonable to me, as there’s a Chinese saying, “a full bottle doesn’t make much noise, but a half-full bottle does.” If we reinterpret the Dunning-Kruger effect in layman’s terms, it would be “a full bottle doesn’t make noise, but an empty bottle does,” which seems a bit contrary to the wisdom of life as we know it.
Therefore, I looked up the original paper on the Dunning-Kruger effect. Surprisingly, the 1999 paper is still not openly accessible, but I managed to get hold of the document (don’t ask me how!).
So, what did they do in their study?
Dunning and Kruger explored their predictions through four studies, assessing abilities in humor, logical reasoning, English grammar, and metacognitive skills.
The results are well-known: people lacking in ability have an illusory sense of superiority, mistakenly thinking they are more competent than they really are.
But who were the subjects of these studies?
The research consisted of four stages, each involving a different number of participants. Here are the details:
1. Humor: 65 Cornell University undergraduates.
2. Logical reasoning: The exact number of participants is not specified, but at least 44 (since the average ranking of 44 participants is mentioned).
3. English Grammar, first stage: 84 Cornell University undergraduates.
4. English Grammar, second stage: Invited back the bottom and top quartile performers from the first grammar study; the exact number is not specified but is certainly less than 84.
5. Advanced logical reasoning: 140 Cornell University undergraduates.
Thus, the study involved at least 333 (65 + 44 + 84 + 140) Cornell University undergraduates.
Notice anything? The number of people is so small, and they are all college students. Can Cornell University undergraduates represent the whole world? Surely not.
The recent study, however, “used four large-scale survey datasets covering a 30-year period in Europe and the United States, proposing a new method of measuring confidence. This method does not rely on self-reporting or peer comparison, but rather defines (over)confidence as a tendency to give incorrect answers rather than an ‘I don’t know’ response.”
So, is it really the most ignorant who are the most confident, or is it a case of “the half-full bottle making the most noise”?
I believe it’s the latter. What about you?
Reference:
Kruger, Justin; David Dunning. Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments. Journal of Personality and Social Psychology. 1999, 77 (6): 1121–34. PMID 10626367
Lackner, S., Francisco, F., Mendonça, C. et al. Intermediate levels of scientific knowledge are associated with overconfidence and negative attitudes towards science. Nat Hum Behav 7, 1490–1501 (2023). https://doi.org/10.1038/s41562-023-01677-8
