Feedback That Students Actually Use: What Research Shows About Making Feedback Work

Feedback is, in theory, one of the most powerful tools available to teachers. John Hattie's synthesis of research found feedback among the highest-impact factors in student learning. In practice, most classroom feedback doesn't produce the effects the research promises — because most feedback doesn't meet the conditions the research identifies.

Here's what the evidence shows about feedback that actually works.

The Problem With Most Feedback

Dylan Wiliam, who has studied feedback longer and more carefully than almost anyone, has written that "feedback is the most powerful single influence on achievement" — and that most feedback in schools doesn't work. The gap between feedback's potential and its typical implementation is one of education's most significant failures.

The research identifies why most feedback falls short:

It comes too late: Feedback on a summative assessment returned two weeks after submission can't be acted on — the student has moved to new content and the motivation to improve is gone. Feedback is most powerful when students can use it immediately.

Students don't read it: Studies consistently show that students look at their grade, look at their peers' grades, and largely ignore written comments. This isn't laziness — it's a rational response to feedback that has no actionable pathway.

It's too general: "Good job" and "needs improvement" communicate evaluative judgments without information about what to improve or how. General praise may feel good but doesn't improve performance. General criticism feels bad and doesn't improve performance either.

It focuses on the person, not the work: "You're not trying hard enough" and "you're so talented" both attribute performance to fixed traits rather than modifiable work. The research on praise and feedback is consistent: process-focused feedback ("you used evidence effectively here") produces better outcomes than person-focused feedback ("you're such a good writer").

Students can't act on it: Feedback that arrives after the work is submitted has nowhere to go. Feedback that arrives during the work can be incorporated immediately.

What Effective Feedback Looks Like

Research from Hattie, Wiliam, and Sadler identifies several features that distinguish feedback that improves learning:

Feeds forward, not just back: Effective feedback doesn't just tell students how they did on past work — it tells them what to do differently on future work. "Your argument would be stronger if you addressed the counterargument in paragraph three" is more useful than "your argument is weak."

Is specific to the task and the goal: "Your thesis statement doesn't address the second part of the prompt" is more useful than "your thesis needs work." The specificity creates a clear action.

Arrives when students can use it: Formative feedback during a task is more powerful than summative feedback after it. The student who can revise in response to feedback learns more than the student who receives feedback on a final product.

Creates a response opportunity: Feedback that students are required to respond to in writing ("write two sentences explaining what you changed and why") produces more learning than feedback that sits in a margin. The act of responding to feedback is itself a learning activity.

Is calibrated to the student's current level: Advanced feedback on fundamentals that a student hasn't mastered is useless. Fundamental feedback to a student who has already mastered fundamentals and needs to develop sophisticated craft is also useless. Effective feedback meets students where they are.

Practical Feedback Strategies

In-process conferences: Brief (2-3 minute) individual conversations while students are drafting or working. "What are you trying to do here? What's getting in the way?" — these conversations produce more change than any written comment on a final draft.

Two stars and a wish (with specifics): Identify two specific things that are working and one specific thing to improve. The "wish" should be specific enough to be actionable, not general.

Margin codes instead of margin comments: Create a code system for common feedback (e.g., "EV" = needs more evidence, "ARG" = the argument isn't clear here, "TRN" = needs a transition) that students learn. Coding is faster for the teacher and requires students to decode and respond.

Feedback before grades: Give feedback on a draft before giving a grade on the final product. When grades come first, they end the conversation. When feedback comes first and grades follow only after revision, the feedback becomes instrumental.

Peer feedback with structure: Students giving feedback to each other can be powerful — but only with significant scaffolding. Specific protocols, modeled examples of useful feedback, and teacher review of the feedback improve quality substantially.

Self-assessment before teacher feedback: Ask students to identify what they think is working and what they think needs improvement before receiving teacher feedback. This activates metacognition and makes students more receptive to specific feedback.

The Grade Problem

Grades undermine feedback in a specific way. Research by Ruth Butler showed that when students receive grades and comments, they look at the grade and largely ignore the comments. When they receive comments only (no grade), they engage with the comments and improve more.

This doesn't mean never grading — grades are often required. But it does mean:

• Separating formative feedback (no grade, focused on improvement) from summative grading
• Giving feedback on drafts before grades on final products
• Requiring students to respond to feedback before seeing the grade

Making Feedback Sustainable

The main barrier to implementing high-quality feedback is time. Writing meaningful specific feedback on 150 essays takes hours. A few strategies for making it sustainable:

Focus feedback on two or three areas at a time: You don't need to address every issue in every piece of work. Identify the two or three most important things to improve and give specific feedback only on those. Students can't act on fifteen pieces of feedback anyway.

Use audio feedback for efficiency: Recorded verbal feedback can be faster to produce and more personalized than written comments. Many students find it easier to respond to than written feedback.

Use whole-class feedback when appropriate: If 80% of the class is making the same error, address it whole-class rather than writing the same comment 25 times. The efficiency gain is significant.

Return to the same areas across assignments: Consistent feedback on consistent areas allows students to see growth over time rather than facing a new set of concerns with every assignment.

LessonDraft supports the planning that makes feedback more strategic — knowing in advance what you're looking for makes giving feedback faster and more focused.

Feedback that improves learning is specific, timely, actionable, and responded to. Most of what passes for feedback in schools is none of those things. Closing that gap is one of the highest-leverage changes a teacher can make.