Giving Feedback on Student Work: How to Make Your Comments Actually Drive Improvement

Most feedback teachers give students doesn't improve learning. That's the uncomfortable finding from decades of research on teacher feedback, summarized most clearly by John Hattie and Helen Timperley's 2007 review. The problem isn't that teachers are giving insufficient feedback — many teachers spend enormous amounts of time writing comments on student work. The problem is that most of that feedback either doesn't reach students (they look at the grade, not the comments) or isn't actionable (they read the comment and don't know what to do with it).

Rethinking feedback doesn't mean working harder. It means working differently.

Why Most Feedback Fails

Research identifies several consistent patterns in feedback that doesn't work:

Feedback about the person rather than the work. "You're a great writer" or "you can do better than this" evaluates the student, not the specific work. Students can't do anything with person-directed feedback except feel good or bad about themselves.

Feedback that points to errors without showing the path forward. Circling every comma splice in an essay tells the student that they made comma splices but not how to fix them or why the rule exists. Error identification is not the same as instruction.

Feedback given too late. When students get a graded essay back three weeks after turning it in, the learning window has closed. Feedback has to arrive while students still care about and can act on it.

Feedback that competes with a grade. When students receive a grade and written comments simultaneously, most of their attention goes to the grade. The comments become rationalization for the grade rather than a guide for improvement.

Too much feedback. Marking every error overwhelms students and doesn't tell them what to prioritize. A student who receives twenty-five comments on a single essay can't do anything useful with that information.

The Three Questions That Make Feedback Work

Hattie and Timperley's model organizes effective feedback around three questions: Where am I going? (What are the goals?) Where am I now? (How is my current work measured against those goals?) How do I get there? (What next steps will close the gap?)

Most feedback addresses only the second question — it tells students how the current work measures up — without adequately addressing the third. "Your thesis is weak" is feedback about where the student is. "Your thesis states your topic but not your argument about it. Try rewriting it to answer the question: what do I want my reader to believe by the end?" — that addresses where the student is AND how they get somewhere better.

The third question is the most important and the hardest to answer well, because it requires knowing specifically what the student needs to do next. Generic next-steps advice ("develop your ideas more fully") is nearly as useless as no advice at all. Specific, actionable next steps based on what you actually see in the work are what change outcomes.

Making Feedback Actionable

Actionable feedback names a specific skill, connects it to what the student actually did (or didn't do), and gives a concrete strategy for improvement.

Instead of: "Your analysis needs more depth."

Try: "You've identified that the author uses repetition in paragraph three, but you haven't explained why this technique is effective for this particular audience. Ask yourself: what effect does repeating this phrase have on a reader who lived through this event?"

Instead of: "Show your work."

Try: "I can see you got the right answer, but I can't tell whether you used the formula correctly or whether you worked backward from a memorized result. Write one sentence explaining your first step."

The shift is from evaluation to instruction. Feedback that teaches — that gives students a specific intellectual move to make — is feedback they can act on.

Strategic Feedback Approaches

Given the time constraints teachers face, strategic feedback — targeting specific aspects of work rather than everything — is both more feasible and often more effective.

Focus feedback on one or two skills per assignment. Before returning papers, decide what you're giving feedback on. An essay you're assessing for argument structure should have feedback focused on argument structure. Surface errors (spelling, punctuation) can wait for a draft where you're specifically addressing conventions.

Use whole-class feedback before individual feedback. When a class-wide pattern appears in student work, addressing it with the whole class is more efficient than writing the same comment thirty times. "Half of you made the same error in your explanations — let me show you the pattern before you revise."

Build in response time. Feedback that students can't respond to is incomplete. Build revision cycles into your assignments. Ask students to write a one-sentence "what I'll do differently" based on your feedback. This closes the feedback loop and helps you see whether the feedback actually reached them.

Use peer feedback intentionally. Peer feedback can be as effective as teacher feedback when students have clear criteria and training in how to give specific feedback. It also provides far more frequent feedback than any teacher can deliver alone.

The Grade vs. Feedback Problem

The single most effective change many teachers can make is separating grades from feedback in drafts and formative work. When you return a paper with only written feedback and no grade, students read the feedback. When a grade is present, most students treat the feedback as an explanation for the grade rather than a guide for the future.

This doesn't mean never giving grades. It means being intentional about when grades serve learning and when they undermine it. For drafts, formative assessments, and practice work, feedback without grades often serves learning better. Grades on final work are appropriate and necessary.

The goal of feedback is a student who takes the next step. Not a student who understands why they got a 78. The question to ask about any feedback you're about to give is: after reading this, will the student know what to do next? If the answer is no, revise the feedback.