The science of feedback
Where you aim feedback matters more than how much you give
Hattie, J., & Timperley, H. (2007). The Power of Feedback. Review of Educational Research, 77(1), 81-112.
Hattie and Timperley's synthesis reframed feedback as answering three questions - Where am I going? How am I doing? Where to next? - delivered at one of four levels. The level you choose decides whether feedback builds skill or just noise. And counter-intuitively, a little well-aimed feedback beats a heap of it.
Data table
| Item | relative effectiveness for learning |
|---|---|
| Process ("the strategy that worked") | 88 |
| Self-regulation ("how to check yourself") | 82 |
| Task ("right / wrong, here is the fix") | 70 |
| Self / praise ("you are so talented") | 18 |
The three questions#
Effective feedback answers feed-up (where am I going?), feed-back (how am I doing?), and feed-forward (where to next?). Most coaching does only the middle one. The forward-looking "here is your next move" carries some of the largest effects in the review - and is the easiest to skip.
Data table
| Item | effect on learning (Cohen's d) |
|---|---|
| Some, well-chosen feedback | 0.39 |
| Lots of feedback | 0.32 |
The four levels#
Feedback can land on the task (correct or not), the process (the strategy behind it), self-regulation (your ability to monitor yourself), or the self ("good girl", "you're a natural"). Process and self-regulation feedback generalizes to new situations. Self-level praise is the most common and the least useful - and shades into the harm Kluger and DeNisi measured.
What it means for Speech Away#
Our report is built as feed-up, feed-back, feed-forward in that order: a clear success criterion (the Structure Sandwich), one earned strength and one fix, then an explicit next action. And we cap it at two to three points - because the research says the extra ones get discarded anyway.