Mental Model: The Dunning–Kruger Effect

Have you ever wondered why some people come to erroneous conclusions despite all the counter evidence, overestimate their abilities, and constantly make mistakes whereas other, more intelligent people often claim ignorance and throw things on the “too hard” pile?

The reality is that not everyone in a given population can be above average despite people believing otherwise due to the Lake Wobegon effect.  For example, people overestimate their relative ranking all the time.  Billionaire Charlie Munger, a major stockholder of Berkshire Hathaway, one of the largest insurance conglomerates in the world and thus responsible for billions upon billions of dollars in policies on everything from cars to airplanes, liked to point out that one study showed 90% of Swedish drivers believed themselves to be better than other drivers, which, having seen the underwriting data himself for different risk pools in the domestic market, knew was mathematically absurd.

This pattern repeats itself all over the place.  Most people believe they are in the top half of intelligence distribution.  Most people believe they are in the top half of physical appearance distribution.

How do we explain this enormous disconnect from reality?  In some cases, it is due to a psychological phenomenon known as the Dunning-Kruger effect, which is one of the most interesting mental models.

The Dunning–Kruger effect is a cognitive bias in which an unskilled person makes poor decisions and reaches erroneous conclusions, but their incompetence denies them the metacognitive ability to realize their mistakes. The unskilled therefore suffer from illusory superiority, rating their own ability as above average, much higher than it actually is, while the highly skilled underrate their abilities, suffering from illusory inferiority. This leads to the perverse situation in which less competent people rate their own ability higher than more competent people. It also explains why actual competence may weaken self-confidence: because competent individuals falsely assume that others have an equivalent understanding. “Thus, the miscalibration of the incompetent stems from an error about the self, whereas the miscalibration of the highly competent stems from an error about others.”

In other words, the Dunning-Kruger effect is when the incompetent person is literally so incompetent that he doesn’t realize just how incompetent he is because he lacks the capacity, or standard, to process that information.


Lack of Access to Performance Standards Data, or Ranking Systems, Explain Some of the Reason the Dunning-Kruger Effect Exists

Dunning and Kruger argued that the reason people feel this way is that many people don’t have access to performance standards data.  In other words, they don’t have a way to rank themselves against other drivers, or professors, or chefs.

Personally, I think this mental model could be exacerbated by the “birds of a feather” nature of human relationships which causes people to be attracted socially, romantically, and professionally to those who are similar.  This could lead to an echo-chamber feedback that may explain things like a group of racists believing they are intellectually superior to other races despite the fact they would fail standardized tests and read at the equivalent of a third grade level.  Since they are only surrounded by others who are of comparable, limited ability, they don’t realize just how badly off they are relative to everyone else.  They literally lack the data and framework to come to this realization.

The Test for the Dunning-Kruger Effect

Kruger and Dunning believed that for any given skill, incompetent people will do four things:

  1. tend to overestimate their own level of skill;
  2. fail to recognize genuine skill in others;
  3. fail to recognize the extremity of their inadequacy;
  4. recognize and acknowledge their own previous lack of skill, if they can be trained to substantially improve.

The Cure for the Dunning-Kruger Effect

If a person has the ability to be trained, meaning the defect isn’t due to limited brain power or some other factor, the cure for the Dunning-Kruger effect is knowledge and exposure to data performance standards.

Think about the people on television shows like Hell’s Kitchen in which regular cooks compete to be the head chef at one of Gordon Ramsay’s restaurants.  Some go in believing, with all of their heart, that they are amazing, talented gourmands.  However, it is clear to everyone around them within a few days that they are terrible in the kitchen or have no palate.

If they have the inherent ability to learn, the only way for these poor fools to overcome the Dunning-Kruger effect is to realize how inadequate they are by exposing them to real chefs who are the best in the world.  They will then understand that they don’t even begin to stack up to them and can start remedial training to improve their skills.

This leads to a fantastic paradox: They must first realize they are ignorant before they can cease to become ignorant. That is the type of problem that would make Confucius smile.

Those of you who want to read the original study that resulted in the Dunning-Kruger Effect being discovered can find it here.


  • Frankie

    Old gem i just ran across. Love it. Thanks Josh.

  • dapmk

    Hi Joshua, an avid reader of your blog here. You’re making a mistake by saying that at least half of all drivers must be below average in driving skills.

    Let me illustrate by looking at a population of 10 drivers, one of which has had a accident in a given year (and let’s also define driving skill by the absence of accidents). All others drive safely. On average, everyone in the population had 0.1 accidents in that year, yet 90% actually are above average.

    It is a common misconception to understand that 50% of any given population have to be below (or above) average. In fact, it’s so common that I wouldn’t even point it out here if it wasn’t for your blog’s much-above-average level in quality and insight.

    Let’s connect this to the Pareto principle here, which states that the upper 20% of salespeople are doing 80% of turnover. Obviously then, less than 20% are responsible for just 50% of turnover. Make some assumptions about the very best members of the team, and this would arguably work out for 85-90% of any sales team to be below average.

    What you wrote would be correct if you used the term “median” instead. Full disclosure: I’m not from Sweden, and I have never driven a car there 😉

    • Mathematically, you’re, of course, right. (I hard on the mean/median difference all the time in the posts dealing with things liked economic data looking at household income as it’s one of my pet peeves.) I was lazy and basically alluded to the original quotation Charlie Munger used in the 1990’s speech in which he introduced the topic. He used them interchangeably but, of course, everyone knew what he meant (he’s Charlie Munger for heaven’s sake – the man is a genius and owned the world’s largest reinsurance empire). I click-whirled it. There’s no excuse.

      I’ll correct it and then delete both comments. Hopefully, I’ll get to it tonight but I have one more quick project I need to complete before I turn in for the evening. If you still see this in the morning, I haven’t had a chance to republish the fix. Thanks for taking the time to send me a message.