Thinking about thinking: the Dunning-Kruger effect

The seventh in a series of articles where Whitianga resident, Max Ross, is exploring the way we think.

The Dunning-Kruger effect is a bias within our thinking in which unskilled people overestimate their ability and skilled people underestimate it. This bias has been measured across a number of different tasks and within a number of different studies.

This bias about our abilities is perhaps due to the fact that when we are new to or unskilled in what we do, we do not know what good looks like in that specific field and when we are good at something, we believe lots of other people are good at it also.

The negative side of this is sometimes called the dual burden effect. Not only are we not good at something, but we also do not know that we are not good at it. Incompetence comes with not being aware that you are incompetent.

This is a bias that encourages you to be humble. If you think you are great at something, it is likely that you are not and conversely if you do not think you are that good, you may be better than you thought.

It is important to be aware of the Dunning- Kruger effect as it is a bias of self-assessment. People may choose careers they are not suitable for or give up career opportunities believing they are not good enough when they are.

This bias may also have immediate consequences, for instance where we decide to jump off a bridge despite not having done many jumps before or drive a powerful motorbike without being aware of the extra skill level required.

The Dunning-Kruger effect was identified in a 1999 study and paper. It has really gained in popularity, has been verified in many similar studies and makes sense. I like the lesson of this bias to be cautious and humble about ourselves.

Recent research, however, is questioning this bias. Mathematical papers released in 2016 and 2017 showed that by generating random data points, pretending they came from studies on the Dunning-Kruger effect, they could show that this bias was happening. The problem is that random data should not be able to show this relationship and so the authors looked into the statistics. They believe that this bias is a statistical artefact of the way the questions were measured and presented.

So now we may think that maybe this bias doesn’t happen and some people judge themselves better and others worse despite their prior knowledge. A bias that seems like the way the world works and has been tested and verified many times, may well be a mathematical anomaly within the way that the statistics are processed.

I am not sure what this means. I am not an expert at statistics, so maybe I am overestimating my ability. It is important to question experiments and to question what we are told, and it is important to stay up to date with the latest developments regarding what it is that we believe and think. I have the thought that incompetent people do not know they are incompetent (and maybe I am included in that). Research now shows that may or may not be true.