Whilst marking recently, I came across an excellent point that was written in a learner’s assignment. The essay task required learners to evaluate two assessment methods. The learner in question explored peer and self-assessment, with one of the points made being that the validity of self-assessment was questionable due to the Dunning-Kruger effect.
The point was excellent and inspired me to summarise my own views on this. It was not so long ago that I had come across the Dunning-Kruger effect myself, courtesy of other bloggers (@learningspy and @jamestheo). The notion is a relatively recent one and in essence asserts that incompetent people tend to be unable to recognise their lack of skill, the extent of their inadequacy and fail to recognise genuine skill in others (further reading). You only have to ask a group of individuals whether they feel that they are an above average driver to see this in action – most people are likely to say they are, thus defeating the ‘average’ rule.
I think about my own practise as a teacher. The more I know, the more I realise I don’t know and the more I feel I need to hone my skills further. Ask me what I thought about my teaching a few years ago and I’d have told you I was the best thing since sliced bread. In hindsight, a lot of what I did was superficial, fun and games. I was unskilled, a novice and due to a lack of knowledge about learning couldn’t see that I was in fact probably having minimal impact. Ask me now and I would probably be consider myself a pretty good teacher, but with a long way to go (perhaps even now I’m subject to the DK effect?!).
For those that follow my blog, you will have seen that I’ve previously raised my concerns with self-assessment, these concerns are exacerbated further in light of the Dunning-Kruger effect. Despite my reservations, for teachers across the world, self-assessment is seen as the ‘gold standard’ of formative assessment (Hattie suggesting that self-reported grades has an ES of 1.44*; Wiliam and Thompson). So how can self-assessment be used in a way that minimises the Dunning-Kruger effect?
- If you use self-assessment, use other methods to draw out understanding (e.g. teacher questioning and peer assessment).
- If you use self-assessment, try to have measurable success criteria so that it is more objective.
- If you use self-assessment, do so when you feel that your learners have enough knowledge and are capable enough to judge themselves accurately.
That’s my two pennies worth on it.
*For critique of Hattie’s findings on self-reported grades, see this link