Interesting, maybe the fact that it is replicated by random noise is actually confirming the theory though. The article claims "completely random guessing replicates the DK graph". But that is compatible with "the ability to assess one's skill is not correlated with measured skill". If random guesses reproduce DK, it implies that people might be randomly guessing at their level of skill.
As a result people with little knowledge over-estimate their ability because a random dice roll will tend to be higher than their low ability (and the converse for people of high ability).
If you look at the DK graph, its close to random, maybe a very slight positive slope -- but ultimately the DK theory would be even stronger if it was truly random, because it means people have NO ability to assess their skill (0 correlation). I think the claim in DK is slightly weaker, that people at least have some ability, just not as much as we like to think.
As a result people with little knowledge over-estimate their ability because a random dice roll will tend to be higher than their low ability (and the converse for people of high ability).
If you look at the DK graph, its close to random, maybe a very slight positive slope -- but ultimately the DK theory would be even stronger if it was truly random, because it means people have NO ability to assess their skill (0 correlation). I think the claim in DK is slightly weaker, that people at least have some ability, just not as much as we like to think.