Stop me if you’ve heard this one before.
Student: “Mr. Stepan, I’m done.”
Me: “Are you sure? Have you checked over our _________ (insert method of assessment here), to make sure you haven’t missed anything?”
Student: (Quickly looks at work.) “Yep.”
Me: “But I can tell, just from glancing at it, that you’ve forgotten ________, and you’ve left out _______, and what about ______?”
As a teacher this frequently occurring narrative would frustrate me. Why wasn’t this student able to go through a list of “look fors” and edit his/her work? How might I support the development of self-assessment skills?
Over the past few weeks, through discussions with two colleagues, Patt Olivieri and Jamie Louis, I’m starting to think that self-assessment (in its current state) is impossible. The majority of our conversations have revolved around some of David Dunning and Justin Kruger’s work (“The Dunning Kruger Effect”). The gist of their research is that, “the skills that engender competence in a particular domain are often the same skills necessary to evaluate competence in that domain.”
Which is maybe a good explanation for the “Yep.” student. What if it’s not the self-assessment that need tweaking? What if it’s cognitively impossible for a struggling student to assess the quality of his/her work? Does that mean that we shouldn’t use self-assessment? Does that mean that the teacher needs to control all assessments?
I don’t think so.
What if we combined peer-assessment and self-assessment to create a new assessment hybrid. Peer assessments would be done more frequently, then students would use peer feedback to self-assess. Maybe we could call it “peer-informed self-assessment”. Trends in peer feedback could be reflected upon and used as tools for metacognition.
According to Dunning and Kruger’s research, improving metacognitive skills also improves accuracy of self-appraisal. So maybe that’s the way to proceed. Self-assessments powered by peer feedback.
Something to try in my SWST schools.