Basketbawful just absolutely killed it yesterday when he put up this re-cap of an old study that reveals why people who are terrible at something think they're so much better than everyone else. I seriously want to read this to my students very, very badly today.
On top of the hilarious examples he riffs on in his post, there are serious, serious repercussions to following this thought to its logical conclusion. If this is truly the case (and it seems likely that it is) we have great reason to doubt pretty much everyone all the time. This is a great start point for anyone who wants to suggest some sort of mandatory testing for (literally) any given activity. If the people who want to do something are over-estimating their skill at that thing, and they will consequently underestimate everyone else, that means, in essence, they're a danger to everyone else doing that thing. (Although, to be fair, I suppose in some inherently un-dangerous activities this wouldn't be a life-threatening concern; it's hard for me to get worked up over old women playing Bridge and overestimating themselves...) However, it's certainly worth thinking about.
And it does explain so, so, so, so much.