What if feedback only counted when it changed the learner?

Abstract

As you know, most teachers give written feedback as if it is a required deliverable, like a checkout assistant handing every customer the printed receipt, even though few use them. The recent fad for setting return times for feedback is also like this: guaranteeing a service with no attention to whether it has any useful effect. E-assessment is if anything even more focussed on "delivery" without the slightest regard for actual impact. What if we judged our feedback strictly by the observable effect it had on the recipient learners?

I have followed a lot of advice on feedback e.g. balancing positive and negative, stimulating discussion of it with both the tutor and peers; yet without much sign of impact. Any evidence of learners actually learning from it has been absent. Recently however I've come across two different cases where there has been something approaching success. I discuss the issue, the many signs of "no effect", and these glimmers of hope.

The two "glimmers" i.e. methods that address these are:
a) Prompting learners to process the feedback, which I'll describe in the talk;
b) 2-dim feedback, which I'll speak about in the afternoon.

Draft design principle: (the workshop was couched to be about "best practice guidelines"):

Web site logical path: [www.psy.gla.ac.uk] [~steve] [talks] [this page]
[Top of this page]