REDRAFTED

Feedback That Moves Learning Forward

Praise feels kind and corrections feel thorough. Here’s what the research says about the feedback that actually changes a pupil’s next attempt.

By Ronnie · 20 November 2026 · 8 minute read

The Question

Ask any teacher what makes the biggest difference to learning, and feedback will be near the top of the list. Ask what good feedback actually looks like, and you’ll get thirty different answers.

Too often, it still looks like a pile of books, a red pen and the same few words going on page after page: “Great effort, well done!” The books go back out, and the same mistakes turn up in the next piece of work.

I was fortunate to train in a school that had begun to look at smarter feedback. Don’t get me wrong, there were still plenty of workload related tasks that could be refined, BUT things were being trialled. At first we were told to “take your books and mark them, every page”. Red pen on pages for the sake of it became a box ticking exercise, taking time and resources that could be used better elsewhere. So the school trialled cards with numbered codes linking to certain task types or things to improve, but naturally these were still very generic. I was then introduced to CRIB sheets (Corrections, Recall, Improvements, Betterment), a one stop shop for going back over a key block of content. I found these more powerful and specific for the pupil, BUT they weren’t fast enough for the pupil to improve in the now. Not only that, misconceptions were left to stay exactly that, feeding other ideas incorrectly.

So my question for this Evidence post: what is the difference between praise, correction and feedback a pupil can act on, and which one moves learning forward?

The Thinking

Praise tells a pupil how we feel about them or their work: “Superb”, “You’re a natural”. Correction tells them what was wrong: a cross, the right answer written in, “remember your units”. Actionable feedback tells them what to do next, clearly enough that they can do it, ideally straight away.

In 2020, Benedikt Wisniewski, Klaus Zierer and John Hattie brought together over 400 studies of feedback in education. Feedback that rewarded or punished had a small effect. Corrective feedback did a lot better. Feedback rich in information, telling pupils about the task, how they went about it and what to do to improve, had roughly twice the effect of correction. Results do vary hugely from study to study.

In 2021, Avi Koenka and colleagues looked at a question every school argues about: grades or comments? Compared with no feedback, grades helped achievement a little but lowered motivation. Compared with comments, grades came off worse on both. But anyone who has handed back a test will recognise it: the pupil finds the number, compares it with their friend’s, and the comment underneath might as well not be there.

The Education Endowment Foundation’s guidance report on teacher feedback, also from 2021, is the most practical of the three. It says feedback should focus on moving learning forward, and that we need to plan for how pupils will receive and use it. The report also says the principles matter more than the method, written or verbal, which is why live marking can do the job.

That’s what the research says. Here’s what I take from it. The 2020 review’s three groups line up with praise, correction and actionable feedback. Praise isn’t the enemy. Pupils need to know their effort is noticed (it’s part of the relationships I wrote about in Week 2). Correction matters too. But only actionable feedback tells a pupil what to do differently next time, and it only works if they get the chance to use it.

The Redraft

So here’s the one thing I’d ask any teacher to try: give feedback in the lesson, while pupils still have the pen in their hand, and use your assessment data to know where to look first.

Before the lesson: question level analysis. After any and all assessments, whether it’s a full mock, an end of unit test or a ten question quiz, I do a QLA in some shape or form. Instead of one total per pupil, I look at every single question. A total tells me WHO is struggling. QLA tells me WHAT they are struggling with, and exactly where each pupil’s areas for development lie.

Early in my career I thought it was acceptable to say “the data shows that in paper one you are struggling with atomic structure as a unit”. But within that unit there are multiple specification points, and most pupils are struggling with one or two key areas. Knowing which ones lets me be far more tactical. I can reteach the gap the whole class shares, plan the comments I’ll need, and target my circulation at the pupils who need me first. The feedback each pupil gets is then about their own next step.

Smart tools have really helped this process. You can now get AI to read a test and break it down into a QLA for you, and my triple scientists have benefitted from this massively. If you choose to use AI, use no names (add them afterwards into your offline or school stored documents) and check the QLA in detail before using it, because the teacher must always remain in full control. AQA’s MERIT Reporting, accessed through Exampro, gives you a live QLA built from AQA’s own tests. Being a science teacher, I have on occasion leant on my technicians (a good science department cannot function without them!), but when doing it yourself, you have to manage your time.

During the lesson: routine circulation and live marking. Live marking became my focus after those early years. Working with colleagues who had already mastered it, I simply went back to basics: observe, note and trial. It’s now a normal part of how I teach. The Do Now, I Do, We Do, You Do, Review structure gives ample opportunity to circulate and live mark, especially during the Do Now and the You Do, while the We Do gives a wide view for in the moment whole class feedback. On a planned route (the well worn path from last week’s post), starting with the pupils my QLA flagged, I check the work against the success criteria, give one comment the pupil can act on there and then, and come back to check they have.

I once had a pupil with the most outstanding book you have ever seen, and I often left them to their own devices. One day I circulated to them and noticed that, despite it looking amazing, the application of their learning was minimal. They were simply looking busy, quickly adding text when we went through self assessment. Live marking is the fastest way I know to pick up on that, and on individual misconceptions and common trends I can act on in the moment.

Here’s the difference on one science answer. Question: “Explain why the rate of reaction increases when the temperature increases.” Pupil: “The particles move faster so there are more collisions.”

  • Praise: “Brilliant work, well done!”
  • Correction: “Not enough detail. 2 out of 4.”
  • Actionable: “Good start: faster particles. Now add the two ideas that get the rest of the marks. Collisions are more frequent, and more of them have at least the activation energy, so more are successful. Write that now and I’ll be back in two minutes.”

One thing I always say to new teachers: “Don’t just tick and flick for the sake of it. All that comes from that is time from you and nothing for the pupil. Be precise and consistent in your feedback.”

The Reflection

What the research can’t tell us yet is how one comment lands with different pupils. The same words can motivate one pupil and flatten another, and most studies measure the next task rather than what sticks weeks later.

Early in my career I was heavy on praise and light on carefully calculated feedback to move learning on. Some of that came from inexperience and gaps in my curriculum knowledge, and some from simply not being far along pedagogically. I now know that while praise is positive, feedback is truly a gift: a gift that moves pupils on.

So here’s my current draft: feedback only counts when the pupil acts on it, so give it while they still can. I’ve no doubt I’ll redraft it.

I’d love to hear from you. How do you use feedback in your own lessons, and what have you found to be most effective?

Sources

Every lesson is a draft. Share what you’d redraft in the comments.

Join the conversation

What did this make you think about in your own classroom? Comments are read before they appear here, so it may take a few days for yours to show.

Your email is only used if I need to reply to you.

Keep reading