Double-Dipping in Dunning—Kruger

Originally posted at sandymaguire.me.

I have to get you to drop modesty and say to yourself, “Yes, I would like to do first-class work.” Our society frowns on people who set out to do really good work. You’re not supposed to; luck is supposed to descend on you and you do great things by chance. Well, that’s a kind of dumb thing to say. I say, why shouldn’t you set out to do something significant. You don’t have to tell other people, but shouldn’t you say to yourself, “Yes, I would like to do something significant.”

Richard Hamming

I want to talk about impostor syndrome today. If you’re unfamiliar with the term, it’s this phenomenon where people discount their achievements and feel like frauds—like at any moment, others will realize that they don’t belong. People with impostor syndrome are convinced that they’ve somehow pulled the wool over others’ eyes, and have conned their way into their jobs or social circles. And it’s very, very common.

On the other end of the spectrum, there’s another phenomenon known as Dunning—Kruger. This one is widely reported in the media as “the average person believes they are better than 70% of drivers.” They can’t all be right—by definition, 50% of people must be worse than the median driver. This too is very common; it seems to hold true in most positive-feeling, every-day domains.

While this is indeed supported by Dunning—Kruger’s data, the paper itself focuses on the fact that “the worst performers in a task are often so bad that they have no idea.” For example, the least funny people are so tragically unfunny that they wouldn’t know humor if it bit them. As a result, unfunny people are entirely blind to their lack of humor:

Participants scoring in the bottom quartile on our humor test not only overestimated their percentile ranking, but they overestimated it by 46 percentile points.

A less well-known finding of Dunning—Kruger is that the best performers will systematically underestimate how good they are, by about 15 percentile points. The proposed reason is that they found the task to be easy, assume others must have also found it easy, and go from there. In other words, top performers are so good that they don’t notice many challenges.

It’s unfortunate that Dunning—Kruger has been popularized as “most people think they are better than they are.” Not only do high-performers already underestimate themselves, but those who know about Dunning—Kruger in its popularized form are likely to double-dip, and further adjust down to compensate for this. For example, if you are in fact in the top 90% for some skill, due to Dunning—Kruger you will likely estimate yourself to be at 75%. But, if you know that people routinely overestimate their abilities by 20%, you might drop your estimate down to 55% in compensation—significantly lower than your true skill level.

If this is true, it would suggest some of the world’s best people will estimate themselves to be worse than the self-evaluations of the worlds’ worst. The takeaway is this: if you’re the kind of person who worries about statistics and Dunning—Kruger in the first place, you’re already way above average and clearly have the necessary meta-cognition to not fall victim to such things. From now on, unless you have evidence that you’re particularly bad at something, I want you to assume that you’re 15 percentile points higher than you would otherwise estimate.

The mathematician Richard Hamming is said to have often ruffled feathers by asking his colleagues “what’s the most important problem in your field, and why aren’t you working on it?” Most people, I’d suspect, would say that the most important problems are too hard to be tackled by the likes of them. That it would take minds greater minds than theirs to make progress on such things. They give up before having even tried.

Instead, the smartest people I know join Google or go work at B2B startups and are simultaneously bored in their day jobs, feel like they’re frauds while they’re there, and don’t have enough energy to work on personal projects after hours. But at least they’re making wicked-big paychecks for themselves. And for their less-qualified leadership.

The best minds of my generation are thinking about how to make people click ads.

Jeff Hammerbacher

Here’s an interesting thought experiment. If you randomly swapped places with someone for a week—you doing their job and they doing yours—how would it turn out? If you’re a competent programmer with passable social skills, I suspect you would be a lot more successful in that week than your replacement. Most things just aren’t that hard, and a good percentage of the people doing those jobs are phoning it in anyways.

The bar on competency is tragically low. And yet the world revolves.

If you agree with this line of reasoning, it means the world is just oozing with potential, ready for the taking. Most of the world’s most competent people are unaware of just how good they are. Most things really aren’t as hard as getting through that graph-theory class you took, and don’t take nearly as much effort. The world is being run by people who are too incompetent to know it; people who are only in power because they’re the ones who showed up, and because showing up is most of the battle.

Which leads us to the inescapable conclusion that this world we live in is particularly amenable to change. That if you’re willing to trust in your own instinct and tackle hard-seeming problems, you’re going to experience literally unbelievable amounts of success. Everyone else is deferring to better minds. So be the change you want to see in the world, because we’ve been waiting for you.