Korva is a scroll of the internet’s best ideas.
Students offered him the chance to concede
Dunning's question
For David Dunning, a psychologist, the question running through his research was how well people know themselves. He noticed it in students in his office, in faculty meetings, and on C-SPAN, where people said things they had to know couldn't be true.
The experiments with Kruger
In his first year of graduate school, Justin Kruger said he wanted to work with Dunning. Dunning put a question to him: how much do people know that they're getting something wrong? They built studies testing logic, grammar, and whether people could spot funny jokes, to see how aware poor performers were of their results.
The first data
The first data set to come in was a revelation. People were going to overestimate themselves, but Dunning and Kruger hadn't realized it would be at that magnitude, and Dunning was absolutely astonished.
How do you grade a sense of humor?
In the humor experiment of the 1999 paper, Kruger and Dunning had a panel of professional comedians rate a set of jokes for funniness. Participants then rated the same jokes, and their scores measured how closely they agreed with the comedians. Participants also estimated how their sense of humor ranked against others. Even the lowest scorers placed themselves above the middle.
Why were logic and grammar tests used?
Both subjects have answers that can be scored as right or wrong, which gave a clean measure to compare with each person's guess. For the logic experiments, Kruger and Dunning drew questions from a Law School Admission Test preparation guide. Participants estimated how many questions they got right and how they ranked among their peers. Comparing those two estimates with actual scores let the researchers tell overestimating a score apart from misjudging a rank.
More from Matthew Hutson
