![]() ![]() The black line indicates their average ‘perceived ability’. The grey line indicates people’s average results on the skills test. In the plot, the two lines show the results within each group. ![]() On the horizontal axis, Dunning and Kruger have placed people into four groups (quartiles) according to their test scores. It’s a simple chart that draws the eye to the difference between two curves. That’s the ‘Dunning-Kruger effect’.ĭunning and Kruger visualized their results as shown in Figure 2. What Dunning and Kruger (thought they) found was that the people who did poorly on the skills test also tended to overestimate their ability. (Actually, Dunning and Kruger used several tests, but that’s irrelevant for my discussion.) Then they asked each person to assess their own ability. They got a bunch of people to complete a skills test. In 1999, Dunning and Kruger reported the results of a simple experiment. But instead of lurking within a relabeled variable, the Dunning-Kruger autocorrelation hides beneath a deceptive chart. Much like the example in Figure 1, the Dunning-Kruger effect amounts to autocorrelation. Now that you understand autocorrelation, let’s talk about the Dunning-Kruger effect. (The variable y comes along for the ride, providing statistical noise.) That’s how autocorrelation happens - forgetting that you’ve got the same variable on both sides of a correlation. As a result, when we correlate z with x, we are actually correlating x with itself. You see, unbeknownst to my colleague, I’ve defined the variable z to be the sum of x + y. We’ve discovered something interesting!Īctually, we’ve discovered autocorrelation. Months later, my colleague revisits my dataset and discovers that z strongly correlates with x (Figure 1, right). After a bit of manipulation, I come up with a quantity that I call z. The problem is that z happens to be the sum x + y. We define a variable called z, which is correlated strongly with x. The right panel shows how this non-correlation can be transformed into an autocorrelation. The left panel plots the random variables x and y, which are uncorrelated. So far so good.įigure 1: Generating autocorrelation. I find that these variables are completely uncorrelated, as shown in the left panel of Figure 1. Suppose I am working with two variables, x and y. But what if a variable gets mixed into both sides of an equation, where it is forgotten? In that case, autocorrelation is more difficult to spot. And that’s true for the pure form of autocorrelation. No competent scientist would correlate a variable with itself. When framed this way, the idea of autocorrelation sounds absurd. Autocorrelation is the statistical equivalent of stating that 5 = 5. If this sounds like circular reasoning, that’s because it is. For instance, if I measure the height of 10 people, I’ll find that each person’s height correlates perfectly with itself. What is autocorrelation?Īutocorrelation occurs when you correlate a variable with itself. 1 It is a statistical artifact - a stunning example of autocorrelation. The reason turns out to be embarrassingly simple: the Dunning-Kruger effect has nothing to do with human psychology. For instance, if you carefully craft random data so that it does not contain a Dunning-Kruger effect, you will still find the effect. The Dunning-Kruger effect also emerges from data in which it shouldn’t. So it would seem that everything’s on sound footing. ![]() ![]() But sure enough, every time you look for it, the Dunning-Kruger effect leaps out of the data. Of course, psychologists have been careful to make sure that the evidence replicates. If you’re very very stupid, how can you possibly realize that you’re very very stupid? Everyone ‘knows’ that idiots tend to be unaware of their own idiocy. It’s the kind of idea that is too juicy to not be true. Discovered in 1999 by psychologists Justin Kruger and David Dunning, the effect has since become famous. Have you heard of the ‘Dunning-Kruger effect’? It’s the (apparent) tendency for unskilled people to overestimate their competence. ![]()
0 Comments
Leave a Reply. |