How to Give Good Feedback And What the Research Actually Says

Most of us think we know what good feedback looks like. We’ve been giving and receiving it our whole lives. But when researchers study the feedback that happens in classrooms and homes, they find something surprising: most of what we call feedback isn’t really feedback at all.

“Great job.” “You’re so smart.” “I’m proud of you.” These feel like feedback. They’re warm, they’re encouraging, and they land well in the moment. But they don’t tell a learner anything about what they did, how they did it, or what to do next. They’re praise. And praise and feedback are not the same thing.

This distinction matters more than most people realize. Research on feedback is among the most studied areas in education, and the findings are remarkably consistent: the right kind of feedback is one of the most powerful tools available for improving learning, and the wrong kind, however well-intentioned, can actively backfire.

What the Research Actually Says

John Hattie and Helen Timperley’s landmark 2007 review of feedback research, published in Review of Educational Research, is the most cited model of feedback in education. Hattie and Timperley themselves called it “one of the most powerful influences on learning and achievement.” Their framework is deceptively simple: effective feedback answers three questions.

Where am I going? What is the goal? What does success look like?

How am I going? Where am I right now relative to the goal?

Where to next? What do I need to do to get there?

Feedback that answers all three questions is dramatically more effective than feedback that answers only one, or none. And the level at which feedback operates matters enormously. Hattie and Timperley identified four levels: feedback about the task itself, feedback about the process of doing the task, feedback about self-regulation (how learners monitor and manage their own learning), and feedback about the person. That last one — the self-level — is where most of us spend most of our time. And it’s the least effective.

Telling a child they’re smart, talented, or special doesn’t give them anything to work with. It feels good, but it doesn’t build anything. Worse, it can actively harm learning, and that’s where Carol Dweck’s research comes in.

The Praise Problem: What Dweck’s Research Showed

Carol Dweck, a Stanford psychologist who has spent decades studying achievement and motivation, conducted a series of studies with fifth graders that changed how researchers think about praise. In her experiments, students completed a set of problems and were then praised in one of two ways: some were told “You must be really smart at this,” while others were told “You must have worked really hard.”

The results were striking. Students praised for intelligence (the ones told they were smart) subsequently avoided challenging tasks, performed worse on harder problems, and were more likely to misrepresent how well they had done. Students praised for effort sought out harder challenges, maintained their performance, and reported enjoying the process more.

One phrase, one interaction, produced meaningfully different outcomes.

Dweck’s explanation is her now-famous theory of fixed and growth mindsets. Children who are praised for being smart come to see intelligence as a fixed trait — something you either have or don’t. When they struggle, they interpret it as evidence that they don’t have enough of it. Children praised for effort come to see ability as something that grows through work. When they struggle, they lean in rather than pull back.

One important nuance Dweck has been careful to add: growth mindset is not just about praising effort for its own sake. Telling a child “Great effort!” when they’re not making progress and not learning anything isn’t helpful either. The goal is to praise the strategies, the process, the persistence — and then help the child find a better approach when what they’re doing isn’t working.

For Teachers: Feedback That Moves Learning Forward

In a classroom, feedback happens constantly — in written comments on student work, in verbal responses during lessons, in the way a teacher responds when a student gets something wrong. The research is clear about what makes it work.

Focus on the task and the process, not the person

The most effective classroom feedback tells students specifically what they did, how it connects to the goal, and what to do next. It operates at the level of the work and the thinking, not the student’s identity or character.

Instead of: You’re such a great writer.

Try: Your opening sentence draws the reader in immediately. The next step is making sure your evidence in paragraph two connects directly back to your claim.

Instead of: This isn’t quite right.

Try: You’ve got the first step correct. Look at what happens when you apply the same approach to step two — what do you notice?

Verbal feedback in the moment

When a student makes an error during a lesson, the feedback response matters. The instinct is often to correct quickly and move on. But research suggests that slowing down and prompting the student to think through the error is more powerful than simply providing the correct answer.

Instead of: No, that’s not right. The answer is 42.

Try: Not quite — what did you do first? Walk me through your thinking.

This approach engages the student in the reasoning process rather than positioning the teacher as the sole source of correct answers. It also gives the teacher information about where exactly the thinking broke down.

Written feedback on student work

Written comments are where many teachers invest significant time and energy, and where the investment often goes unread. Most of us have done it ourselves: flipped straight to the grade and skimmed past the comments. When grades and written feedback appear together, the grade tends to win. Research confirms what teachers observe every day: many students engage far more with feedback when the grade isn’t competing for their attention.

A few principles that help:

•  Be specific and forward-looking. Comments like “good” or “needs work” tell students nothing. The most effective written feedback names exactly what is working and gives a clear, actionable next step.

•  Less is more. Comprehensive feedback on every error overwhelms rather than guides. Identify the one or two highest-leverage things to address and focus there.

•  Consider separating feedback from grades: returning work with comments before revealing the grade. When students see the feedback first, they’re more likely to actually read it — which is the whole point.

Growth mindset language in the classroom

The language teachers use every day sends students messages about what learning is and what they are. Small shifts in phrasing can reinforce a growth mindset rather than a fixed one.

Instead of: You’re a natural at this.

Try: You’ve worked really hard on this skill and it shows.

Instead of: Some people are just better at math.

Try: This is hard. That means your brain is doing important work right now.

Instead of: You’re almost there — keep trying!

Try: You’re almost there. What’s a different strategy you could try for this part?

The last example reflects Dweck’s nuance: effort alone isn’t enough. The goal is to help students see that struggling is an invitation to think differently, not just try harder.

For Parents: Feedback at Home

The feedback parents give matters too, perhaps more than most parents realize. Children absorb the messages they receive at home about what learning is, what success means, and what to do when something is hard. A few principles worth carrying into everyday conversations.

Separate praise from feedback

Warmth and encouragement are important. But they work best when they’re paired with specific information rather than offered as a substitute for it. “I’m proud of you” is meaningful. “I’m proud of how you kept trying even when it got frustrating” is more useful, because it tells the child exactly what to repeat.

Ask questions instead of evaluating

When a child brings home a piece of work, the instinct is often to evaluate it: “That’s great!” or “Could you make this part a little neater?” A more powerful approach is to ask the child to reflect on it first.

Try: Tell me about this. What part are you most proud of?

Or: What was the hardest part of this for you?

Or: If you were going to do this again, what would you do differently?

These questions shift the child into an evaluative role — the self-regulation level that Hattie and Timperley identify as one of the most powerful for deep learning. They also give parents far more useful information about how the child is experiencing their own work.

Respond to struggle with curiosity, not comfort

When a child is struggling, the natural parental instinct is to reassure: “Don’t worry, you’ll get it.” But empty reassurance doesn’t give the child anything to work with. A more useful response engages with the struggle itself.

Instead of: Don’t worry, you’re great at reading.

Try: This one is tricky. What’s confusing you? Let’s look at it together.

The message this sends is that struggle is normal, that understanding the confusion is the first step, and that figuring it out together is possible. That’s a different message from “you’ll be fine,” and it’s a more honest and more helpful one.

Watch your language around intelligence and ability

Dweck’s research applies directly to everyday parenting conversations. The off-hand comments parents make about intelligence — their own and their child’s — shape how children think about ability.

Instead of: She’s always been my math kid.

Try: She’s put in a lot of work on math this year.

Instead of: I was never good at reading either.

Try: Reading can be hard. Here’s what helped me get better at it.

These shifts aren’t about policing language. They’re about noticing the quiet messages we send about whether ability is fixed or grown, and choosing the message we actually want to send.

The Bottom Line

Good feedback is specific, forward-looking, and focused on the work rather than the person. It tells a learner where they are, where they’re going, and what to do next. It treats struggle as information rather than failure. And it builds the belief that effort and strategy — not fixed talent — are what make learning possible.

That’s as true in a third-grade classroom as it is at a kitchen table on a Tuesday night. The research doesn’t distinguish between the two. Neither should we.

Your Turn

For teachers

What’s your current approach to written feedback on student work? And has there been a moment in your classroom when feedback — yours or a student’s — landed in a way you didn’t expect?

For parents

Is there a phrase you find yourself using with your child that you want to rethink? Or a conversation about struggle that went better than you expected?

Subscribe so you don’t miss next Sunday’s article.