Have you heard of the ‘Dunning-Kruger effect’? It’s the (apparent) tendency for unskilled people to overestimate their competence. Discovered in 1999 by psychologists Justin Kruger and David Dunning, the effect has since become famous.
Except there’s a problem.
The Dunning-Kruger effect also emerges from data in which it shouldn’t. For instance, if you carefully craft random data so that it does not contain a Dunning-Kruger effect, you will still find the effect. The reason turns out to be embarrassingly simple: the Dunning-Kruger effect has nothing to do with human psychology.1 It is a statistical artifact — a stunning example of autocorrelation.
EDIT: see response from dustyData and the article they linked to https://www.bps.org.uk/psychologist/dunning-kruger-effect-and-its-discontents
I trust a political economist opinion on psychology as much as I trust my car mechanic’s opinion on cardiac surgery.
Which is funny, because that’s a perfect example of the Dunning-Kruger effect. This Blair Fix guy doesn’t know as much of psychology as he thinks he does and he is unaware of it.
Specially since Nuhfer et al made some egregious statistical mistakes themselves. Particularly as their critique is over something that was done in the 1999 paper but has not been done ever since nor in any of the subsequent studies.
Here’s David Dunning response (2022) to most critiques. The most poignant is of course the argument that even if it is just an statistical artifact, it is still a real phenomenon worthy of study and of understanding the causes. Because the statistical arguments might debunk one experimental design, that the authors were aware of and corrected. But it doesn’t debunk the experimental designs of the subsequent paper that same year and the other thousands of papers published since.
You’re in trouble now, OP. The internet loves Dunning-Kruger almost as much as it loves Keanu.
Yup! The overall tone of the text comes of as arrogant and smug…
I was writing a very long wall of text to explain how the authors comprehension of statistics is limited, but it doesn’t matter so much.
What really matters is that on the end, the weak correlation between actual knowledge and perceived knowledge results in the DK effect.
People with no real ability will overestimate their perceived ability, while the reverse is true for people with real ability.
The lack of correlation between perceived and actual ability, that is the real DK effect!
Multiple researchers have debunked the effect, and they managed to recreate the results with completely random data. Meaning the effect being described doesn’t actually exist. They would have produced the same results regardless of the information they plugged in.
Recreating the results with completely random data is a proof of the DK effect!
It means that there is no link between ability and perceived ability, so people without ability will overestimate while people with ability will underestimate.
To refute the DK effect, we would need ability and perceived ability to be exactly the same.
To me, this whole debate shows that some people without social science background think they understand social science better that the actual experts… Quite ironic in my opinion!
Yup! The overall tone of the text comes of as arrogant and smug…
I was writing a very long wall of text to explain how the authors comprehension of statistics is limited, but it doesn’t matter so much.
I read the beginning of their response and I can’t tell if the rest was supposed to be a joke.
You done messed up A-A-Ron
You’re going to have to explain this in a more digestible way if you want most people to understand it
They can’t… The article is wrong, focusing on mathematical artifacts and missing the whole point.
If one’s actual ability is 0, any minuscule amount of perceived ability is an overestimation. Same thing if your actual ability is 100%, it’s impossible to overestimate your ability (only underestimation is possible). Resulting in the DK effect.
Look at the chart in the article. Let’s say actual ability is x and perceived ability is y. The graph effectively asks us to compare the difference between perceived ability and actual ability, which we can write as y-x. Thus, the chart effectively graphs y-x over x, which can be written as -x+y over x. There is always a strong correlation unless y approximates x.
IMO, the conclusion of that analysis should be “Dunning−Kruger is a truism”, not “Dunning−Kruger doesn’t exist”.
I think people in general tend to think they are smarter than they really are.
Maybe some people, but I really am smarter than I am
I’m gonna need more people / experts advancing this notion before I believe it.
People who think they understand Dunning-Kruger, and use it as an insult, are a perfect example of Dunning-Kruger, which does not exist.
