Mental model: The Dunning-Kruger effect


What do you know [about x]?

How do you know that?

How confident are you that you’re right [about x]?

I don’t find these questions easy to answer. In fact, they cause me a considerable amount of hemming, hawing, head-scratching and “um”-ing.

Paradoxically, I used to find these questions easy to answer, but now that I’m older I consider prior-Jeffrey to be ridiculously naive and arrogant. I used to have the all the (wrong) answers; now that I’m a bit wiser I struggle to come up with any answer!

Nowadays, when my old(er) brain is pondering these questions I find it helpful to consider the Dunning-Kruger effect. From Wikipedia:

The Dunning–Kruger effect is a cognitive bias wherein people of low ability suffer from illusory superiority, mistakenly assessing their cognitive ability as greater than it is.

You can find this effect plotted in various ways, like this:

Dunning-Kruger effect

or this:

Dunning-Kruger effect

or my favourite:

Dunning-Kruger effect

The gist is that as one starts out in a particular domain, the tiniest bit of knowledge they gain gives them a vastly overinflated view of their abilities. They think they know more than they do and that the domain itself doesn’t have the depth that would require years of practice to master.

Eventually, they start to learn more and realize how arrogant they were. A feeling of hopelessness ensues, and if they decide to continue (which isn’t a given), it’s a long time until their confidence begins to rise alongside their abilities.

As time goes on, the confidence they have in their abilities and domain experience rise in step until finally, after years of deliberate practice, they have a level of confidence in their knowledge and skills that is honest, hard-won, and accurate.

A useful mental model

I have quite the backlog of old journal entries that I wrote from 2013 on. Those early years – when I had just started my business – were rough. The entries are fun to read now, but I was clearly going through a tough time.

As the years went on, however, the entries started to feel more positive. I had tried a lot of things and failed at most of them. But a little ray of success started to peek through the clouds and began to shine brighter as the entry dates moved forwards. I had learned from my failures, gained experience, and finally understood how to be humble.

Despite my confidence levels continually proving to be wrong, the experience I was gaining gave me a clearer look at what it would take to move forward. Less rose-coloured glasses; more cautious optimism.

My journals show an experience that looked something like this:

Dunning-Kruger Effect - My Business Journey

Reckless confidence and little experience in late 2013/early 2014 married to produce a rather impulsive decision – quit my job and work on the business full-time! My confidence peaked in 2015 and started to slide, as I realized just what I had gotten myself into. 2016, thankfully, saw me working harder to improve, as I began to grasp the extent of what I needed to learn. And 2017 was the year I feel like I actually got better, and could begin to see what the journey ahead might look like.

Making decisions

I’ve found that having a sense of where I am on the curve helps me to make decisions.

For example, should I give more weight to my own opinion, as opposed to the opinion of another? Do I have an accurate-enough sense of the scope of [field], or is it likely that I’m deluded? Are there outsized returns available in learning the basics well, or am I in the land of diminishing returns?

Here’s how I would position myself in a few more domains:

Dunning-Kruger Effect - Marketing

Dunning-Kruger Effect - Development

Dunning-Kruger Effect - Finance

Dunning-Kruger Effect - Leadership

Keep in mind that you can go as broad or as narrow as you like. Instead of marketing as a whole, for example, you could plot yourself on a content marketing, Facebook ad, or SEO chart.

I’ve found this useful because it visualizes where I may want to focus my energy, and where my judgment may be lacking the most. If you’re a marketer, for example, this method feels to me almost like another way of visualizing the skills required to be a T-shaped marketer:

T-Shaped Marketer
Credit: Brian Balfour. Also see Buffer.

Once you’ve identified where you need to focus, these frameworks also help with asking the right questions.

Asking questions

Learning is all about asking questions. Better questions lead to better answers, and a good question always begins with an understanding of how much knowledge you have about the topic.

The Dunning-Kruger effect has allowed me to visualize where my experience and confidence levels might sit in a particular domain, and the type of questions I should be asking.

Here’s a few I’ve found useful:

  • Where am I strong, and where am I weak?
  • Am I entitled to an opinion here?
  • How much weight should I put in [other person’s] opinion?
  • Is this domain like another that I have experience in?

I understood the Dunning-Kruger effect, at a basic level, long before I came across the research. Why? Because many times over, I had held what I thought was a strong position only to have it destroyed by another’s expertise.

I learned that in everything, the rabbit hole goes down much further than a cursory glance would suggest. That I should approach a new subject, a political issue, or an unusual opinion with openness, and a willingness to find out where I’m wrong.

Because I’m always wrong–we all are, at least a little bit–and the goal is just to be right (or less wrong). With a rough idea of your expertise, confidence in your knowledge, and a willingness to learn, it seems to me that decision-making can only get easier.