The power of feedback: what almost 200 studies found

The gist. Feedback can carry close to double the effect of an ordinary year's teaching, but only when it names a clear goal, shows progress toward it, and points to the next step.

What the study found

Hattie and Timperley pulled together 12 meta-analyses covering 196 studies and almost 7,000 effect sizes on feedback in classrooms. The average effect was 0.79, close to double the typical effect of a year's teaching (0.40), placing feedback among the top 5 to 10 highest-scoring influences on achievement out of more than 100 studied, alongside direct instruction and reciprocal teaching.

That average hides huge variation. A wider synthesis of 74 meta-analyses found feedback on the task and how to do it scoring near the top, while praise and rewards scored near the bottom, ranging from 0.12 for teacher praise up to 1.24 for feedback to learners with special educational needs.

What it might mean for your classroom

The strongest feedback in this evidence answered three questions for the learner: where am I going, how am I doing, and where to next. That pattern sits alongside other formative assessment strategies already common in strong classrooms.

• Make the goal specific. Answer 'where am I going?' with a goal a learner could describe back to you, not a vague one.
• Write on progress, not just a grade. A written note on how the work is going beats a mark alone; try one written next step instead of a tick.
• Aim comments at the work. Target the task or the process rather than the learner, as many marking strategies already try to do, since praise about the person carries little useful information.

Treat it as an experiment, not a recipe: this is a synthesis of what dozens of studies found about feedback in general, not a test of any single classroom routine. Swap one grade-only comment for one written next step, then watch whether learners can tell you what they will do differently next time.

How solid is this?

Promising. This is a widely convergent synthesis: a dozen independent meta-analyses point the same way, which is about as convergent as education research gets. But it draws together existing meta-analyses rather than running one new, pre-registered systematic review with its own documented search method, the headline averages hide wide variation, from 0.12 for praise to 1.24 for special educational needs, and the underlying studies are international rather than verified as UK-representative. Treat the direction, that feedback on the task and next steps helps more than praise alone, as well established. Treat any single number as an illustration of scale, not a promise for your class.

The source

Hattie, J., & Timperley, H. (2007). 'The Power of Feedback.' Review of Educational Research, 77(1), 81-112.

Read the original study