this post was submitted on 18 Mar 2024
26 points (65.9% liked)

Psychology

316 readers
6 users here now

Come talk about psychology and related disciplines.

founded 1 year ago
MODERATORS
nonfatmatt
26
The Dunning-Kruger effect is not real (economicsfromthetopdown.com)
submitted 5 months ago* (last edited 5 months ago) by [email protected] to c/psychology
16 comments fedilink hide all child comments
 

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

top 16 comments
sorted by: hot top controversial new old
[–] dustyData 45 points 5 months ago* (last edited 5 months ago)

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.

[–] robolemmy 34 points 5 months ago (2 children)

You're in trouble now, OP. The internet loves Dunning-Kruger almost as much as it loves Keanu.

[–] [email protected] 16 points 5 months ago (1 children)

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!

[–] Wogi -3 points 5 months ago (2 children)

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.

[–] [email protected] 4 points 5 months ago

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!

[–] [email protected] 1 points 5 months ago

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.

[–] [email protected] 12 points 5 months ago

You done messed up A-A-Ron

[–] [email protected] 13 points 5 months ago (2 children)

You're going to have to explain this in a more digestible way if you want most people to understand it

[–] [email protected] 31 points 5 months ago

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.

[–] [email protected] 1 points 1 month ago* (last edited 1 month ago)

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".

[–] renrenPDX 10 points 5 months ago (1 children)

I think people in general tend to think they are smarter than they really are.

[–] yemmly 18 points 5 months ago

Maybe some people, but I really am smarter than I am

[–] [email protected] 3 points 5 months ago (1 children)

I'm gonna need more people / experts advancing this notion before I believe it.