Hacker News Reader

Comment by smsm42 | Hacker News Reader

But, judging from the graph in the article, the top quartile actually under-estimates their ability? Only lower 3 quartiles overestimate it, so for experts it doesn't seem to be true? Unless, of course, I misunderstand DK effect completely and am not aware of it...

View on HN

Yes, the point is that in your area of expertise you are likely to be more humble relative to your actual ability (but not too far off, on average).

However outside of your specific area(s) of expertise your confidence in your ability is likely to far exceed your actual ability (on average).

So experts are still susceptible to DK effect for areas that lie outside their expertise. Even (or perhaps especially) if they think their expertise gives them special insight into another field.

This makes sense to me. If you're an expert you are comparing yourself to other experts in your field. On the other hand, if you know anything about a topic but only interact with people who know nothing at all about it, you would rate yourself highly.

For example, I read a lot about modern physics but am not a physicist. I would give myself an above average understanding but would likely fail an exam.

It’s true for experts outside of their field of expertise.

It's true for experts (above ~70th percentile) in their area of expertise, too, just in the opposite way as it is true for people below that level.

The more you know, the more you know that you don't know...

Reply on news.ycombinator.com