Hacker News new | ask | show | jobs
by lupire 1522 days ago
Dunning-Kruger (as you are using it) is a myth, though.
1 comments

Are you referring to people often misapplying the concept, or its replicability issues (which I had admittedly come across before but don't hold in active memory)?

I'm confident I nailed the former (I thought I was better at things than I was in reality when I was younger, and am now more likely better than I think I am in those same domains today).

If you referring to the latter, it's tricky for me. It's like the famous two humped camel in programming aptitude. It seems to be useful in describing experiential observations but didn't (doesn't?) replicate. I've stopped using it at this point because I don't want to mislead people.

DK also has the appearance of explanatory power at a minimum, and utility in communicating the simple form of the concept. I don't like the replicability issue but wonder how bad the down side is if all I'm really saying is we aren't often humble enough about new things, and aren't confident enough about things we've dedicated time to?

Welp, that justification for promoting something potentially falsifiable surely isn't going to come back to bite me and expose my hypocrisy elsewhere...