Immediate vs Post-Event Feedback
Last week, we discussed the Roediger and Karpicke experiment, which showed how important assessment is, not just as a means of measuring learning, but also and above all as a means of improving learning.
To do this, the evaluation must (if possible) provide feedback. On this topic, Catlin Tucker in her book Complete guide to blended learning says something that I find very interesting: “teachers traditionally provide feedback on finished products, so there is no incentive for students to do anything with that feedback. I maintain that feedback on finished products is a waste of teacher time because it is not actionable.”
As a young teacher, I sometimes struggled to understand this when I tried to persuade my students to correct the mistakes on the papers I returned to them. Obviously, they had little desire to spend time improving what Catlin Tucker called ‘finished products’. I’d try to persuade them that they’d be able to make progress in this way and get hypothetical future good marks for an eventual production that would come who knows when.
To no avail.
At the time, I didn’t understand this simple fact: if you put a mark on a given piece of work, any comment on it fails to achieve its objective of encouraging progress. Others have put it this way: archaeological feedback, which we call this because it relates to work that has been completed a long time ago, is ineffective for the reasons set out above. Real-time or immediate feedback (i.e. before the work is finished) is more effective than autopsy feedback (feedback on finished work).
This is why it is a good idea to comment on the copy before marking it, or not to mark it at all.