Homework Is Worth 5 Months In Secondary, Only 3 In Primary

This is the education commute. Both voices on this show are synthesized. The judgment isn't. I want to establish a scenario right at the top of our time today. And well, it is a highly, highly relatable one for anyone listening right now. Picture this. OK, I'm picturing it. It is Sunday night. It is exactly a quarter to 10. The heating is probably just clicked off. The house is completely quiet. And you are sitting there at the kitchen table. Oh, I know this scene. Yeah. Right in front of you is a stack of 31 exercise books.

And you've got half of them marked. It's the, uh, the universal teacher experience, but very familiar and honestly slightly painful scene to set. The Sunday night dread is a real tangible thing. It really is a heavy feeling. So there you are holding book number 17. You're looking down at it. Your red pen or green pen, whatever your school policy dictates is just hovering over the page. Yeah. And you are presented with a very stark, very immediate choice. Does book number 17 get real? targeted feedback, or does it just get a tick?

And the conflict in that exact moment is that professionally, pedagogically, you know exactly which one it should get. You know what the right answer is in a perfect world, but you also know that it is a quarter to ten, your alarm is going off at six, and well, you still have lunches to make. Exactly. The pressure of the time forces the decision. You look at the clock, you look at the remaining pile of 14 books after this one, and you just make the call. You slash a quick tick on the page, move to 18 and just push through.

It's a shared reality. We've all been there. But what I find so striking about that moment is the unseen connection. What do you mean? Well, somewhere in a staff room, 40 miles away, there is another teacher in the exact same position. They are marking the exact same subject. But they are marking it for a class that is half the age of yours. Oh, right. And as they sit there, putting their own thin ticks on the pages just to survive the night, they are quietly wondering if any of this homework is actually landing at all.

Does it even matter? Two teachers. One Sunday. sharing an unchecked assumption that the mere act of setting and ticking this work is doing something valuable and that assumption is well It's exactly what we are pulling apart today. I'm glad we are. Because we are looking directly at the data from the Education Endowment Foundation, specifically the EEF Teaching and Learning Toolkit, the current edition. And we're looking solely at the homework strand today. Which is arguably one of the most vital pieces of the entire toolkit, simply because it drives so much whole school policy, but it's often the most misunderstood.

Right. So let's get straight into it. I want to deliver the first major claim. the headline of the EEF data. Across 43 studies, the data shows that homework delivers an average of plus five months progress. Plus five months? I mean, in the realm of educational interventions, that is a substantial number. It really is. It is the kind of number that makes head teachers just mandate an hour of homework a night across the board. But hold on, before we even get to the caveats, we need to unpack what that metric actually means in the real world.

Because we throw around plus five months, like it's a completely normal unit of measurement. What does that actually look like? Yeah, it's confusing. Are we saying a 10 year old magically reads like a 10 and a half year old just because they did some worksheets at home? How do you even measure a month of progress in a human brain? That is a crucial pushback. No, a child isn't suddenly aging up. cognitively. What the EEF does is take the statistical effect size from these academic studies, which is usually measured in standard deviations.

Which is completely impenetrable for anyone who isn't a statistician. Exactly, impenetrable. So they translate it into a metric of months of progress, so it's accessible. It basically means that over a year, a student doing this intervention would, on average, perform as if they had had five extra months of standard teaching compared to a student who didn't. Okay, so it's an indicator of accelerated learning, like a student grasps fractions by March instead of July. Exactly. It's a slightly blunt instrument, but it helps us compare different strategies on a level playing field.

Well, speaking of blunt instruments, we have to look at the other side of this headline. Yes, it says plus five months, but we must explicitly note the evidence rating attached to that number. This is the important part. In the toolkit, the evidence rating is low. Visually, it is represented as two padlocks out of five. And we really need to pause on that. Low security on the headline number itself is something we have to be completely honest about. Two padlocks out of five is incredibly easy to misinterpret.

I mean, two out of five sounds like a terrible restaurant review. If I see a two -star rating for a hotel, I'm not booking it. Yeah, I'd say away, too. So why on earth are we basing sweeping school -wide homework policies on something with a two -padlock evidence rating? Isn't that essentially a coin toss? It sounds like it, but here is where educational research differs from, say, a double -blind clinical trial for a new medicine. In a lab, you can control every variable. In a classroom, or worse, in a child's living room.

You can't control anything. Are they doing the homework in a quiet study? Are they doing it on the bus? Are their parents doing it for them? The noise and the data must be deafening. It is. So a two -pad lock rating doesn't mean this data is garbage. It means the best available evidence points in this direction, but we acknowledge there are massive contextual variables we can't control. It does not mean homework definitely works. It means it's an indicator. But it comes with real limits. Okay, so holding that headline in mind the plus five months progress across those 43 studies With a healthy dose of skepticism about the variables I want to look at the core data split because this is where the conversation and honestly the staff room arguments usually fracture Oh, absolutely when you split the data between secondary schools and primary schools.

You don't get the same number secondary schools show plus five months progress. Primary school show plus three months. A very clear numerical distinction between the age phases. And just to be incredibly clear, that is not me rounding the numbers to make a point. That is the toolkit's exact phrase. To quote it directly, studies in secondary schools show greater impact than in primary. So the gap is very real. And here is where I think we need to state the common orthodoxy out loud. The narrative that we all just assume to be true because it feels right.

Let's hear it. The orthodoxy goes like this. Homework matters more for older kids because they are simply more independent. They don't need a parent sitting with them at the kitchen table dragging them through a worksheet. Younger kids, primary age, absolutely need that scaffolding. And if they don't get it, the homework just doesn't do as much. It's a neat, logical story. It is a brilliant story. It's exactly what I thought when I was reading this. You look at the plus five for secondary and the plus three for primary.

You picture a self -sufficient 16 year old versus an easily distracted eight -year -old, and you immediately draw a straight line between the two. But I actively challenged this assumption. I went looking for the why in the text. I wanted the proof. And here's the twist. The toolkit doesn't actually say why. Wait, they report a gap that significant and just don't offer an explanation for it. Nothing. Zero. I checked directly because I didn't want to just assume the text backed up our collective hunch. I read through looking for the explanation and there was no section on independence.

Wow. There is no claim about self -regulation. There's absolutely no line in there stating that older pupils manage their time better. The toolkit reports the split as a flat fact and then the text moves straight to the next figure. So they just drop the stat and walk away. Essentially, yes. Which means I'll believe they're just more independent. the day someone shows me the study that actually says it, not just the effect size set next to it. That really exposes something fundamental about how we consume research and education.

We see an effect size, we see a gap in the data, and nature abhors a vacuum. Yeah, we have to fill it. Exactly. Because the researchers didn't explain the gap, the profession couldn't handle the empty space. So we immediately invented a mechanism to explain it. We told ourselves a story about independence. And then over time, we dressed that invented mechanism up as if it were the research itself. We start treating the independence narrative as peer -reviewed data. When in reality, nothing in the evidence actually answers the question of why the gap is there.

And I think we have to be comfortable saying we don't know the why. I would much rather concede that the mechanism is a mystery than invent a story and build school policy around it. Which is the perfect pivot. If age and independence aren't the magic levers we thought they were, what actually moves the number? Did the toolkit find anything that universally moves the needle, regardless of whether you're teaching year three or year 11? Yes, and this is far more useful than chasing a Y that isn't there.

The claim that matters most here is about how the homework is handled once it comes back into the classroom. The defining factor is feedback. Studies with feedback showed higher impact than those without. It is the single biggest variable that changes the impact completely, regardless of the age phase. So if we synthesize that down to its absolute core, the marking matters more than the setting. Yes. But let's push on that. Why? Why does feedback beat frequency? Yeah. We need to explain the cognitive mechanics here rather than just reciting the finding.

If a kid does a math sheet, hands it in, and gets a tick, why is that less effective than getting a specific comment? It comes down to how the brain processes task completion. Think about the analogy of practicing a golf swing in the dark. OK, I'm with you. You can stand out there on the driving range for two hours, putting in the time, sweating, hitting 100 balls. That's high frequency. But because it's pitch black, you have absolutely no idea where the ball is going. Are you hooking it, slicing it?

You don't know. The effort is there, but the learning isn't because there's no feedback loop. You're just entrenching your own mistakes. Precisely. Turning on the lights is the feedback. When a student hands in work and just gets a tick, their brain registers the task as complete. The cognitive loop closes, and the brain essentially discards the material. The tick signals the end of the transaction. But when they get a targeted comment, say, you used the right formula here, but check your order of operations in step three, the loop stays open.

The brain is forced into a secondary cognitive step. They have to re -engage with the material to resolve the comment. That makes total sense. The feedback isn't just a nicety. It's the mechanism that forces the brain to revisit the learning. Now before we look at what this looks like in a classroom, there are two modifiers in the smaller print of the toolkit that we need to address because they also move the number significantly. What are they? First, homework using digital technology shows plus six months progress.

Second, homework with peer collaboration also shows plus six months. Plus six months for both. So they're beating out the average plus five and definitely beating the primary plus three. But let's look at the reality of that digital modifier. Plus, six months for digital technology sounds great in a sterile research environment. Right. But how is a teacher supposed to rely on that when half the kids forget their platform logins, the school's Wi -Fi is constantly dropping out, or a kid is trying to write an essay on a cracked smartphone on the bus?

It's a huge practical barrier. But if we tie it back to the mechanism we just discussed, the feedback loop, the digital bump makes sense. It's not that tech is magic. It's that digital platforms usually offer immediate automated feedback. Oh, that makes sense. Yeah, a student types in an answer, and the system instantly tells them if they are right or wrong, often with a hint. The feedback loop is instantaneous, whereas a teacher might take a week to return a book. That's a great distinction. The tech is just the vehicle for faster feedback.

Now, what about the peer collaboration finding? Plus six months is huge. It is, but there is a major caveat attached to it. The toolkit explicitly flags the peer collaboration finding as having limited evidence, meaning there are fewer underlying studies supporting it, a much smaller data pool. Caveat as content? We have to treat that as fine print. We are much less sure about the peer collaboration number, even though it looks high. Group work for homework is practically begging for one kid to do all the work while the other three play video games anyway.

Exactly. So Bringing it back to what we do know works, feedback. But we have to address the cost of this. The toolkit rates the cost of homework as very low. But let's be fiercely realistic about what that actually means for you listening to this. It mainly costs time. The one resource nobody has. Exactly. The time you spend setting it, the time you spend marking it, and the time the school spends training staff on a new homework strategy. It is not free. It is just cheap in terms of a financial budget.

It demands a massive amount of your personal time. Which brings us right back to our Sunday night scenario. If the time cost for the teacher is that high, we need to guarantee that time is being spent on the lever that actually moves the dial. And the data is screaming that the lever is feedback. So let's walk through exactly three settings. Let's show what this principle looks like in action, because it is one thing to sit here and say feedback matters, and it is another thing entirely to see it functioning in a real room with real constraints.

Let's do it. Room one, primary. What does high impact feedback look like here? It looks like a spelling list returned to an eight -year -old with one specific comment about what to try next. Not a stamp that says good effort. Not a stamp and certainly not 20 ticks in a smiley face. The 20 ticks in the smiley face just confirm compliance. They confirm the homework was done. The one specific comment like, I noticed you're struggling with the eye rule. Try writing these three words again. That is feedback on how to improve.

It forces the cognitive loop open. Okay. Let's move down the corridor to room two, secondary. This is where time is often the tightest. A secondary teacher might see 150 kids a day. What does it look like for them? It sounds like a subject teacher intentionally building five minutes into the very start of their next lesson. Not to teach new content, but specifically to hand back the homework with a verbal steer. Yes. It is an actual spoken loop. in the physical classroom. It's not just yelling, check your grades online over the bell as 30 kids shove their way out the door.

No, the default is so often to let the digital system deliver the grade passively, but building in those five minutes for a verbal steer saying, look at question four, a lot of you missed the second step, talk to the person next to you about why, that is where the feedback actually happens in secondary. And finally, room three, a specialist setting, because we can't leave them out of the conversation. Here, the homework might look entirely different. It could be a task strip or a photolog of a student practicing a life skill at home.

Absolutely crucial here is that it serves as a homeschool communication tool that gets an actual response from the teacher. The format doesn't matter. The feedback loop still exists, even where the homework looks nothing like a traditional exercise book. The teacher sees the photolog and sends a note back to the parent and child saying, great job sequencing those steps. Let's try doing it without the visual prompt tomorrow. Right. And it is so important to note that walking through these three rooms isn't about citing specific research that proves something unique.

about primary, secondary, or specialist environments, the toolkit doesn't get that granular. It is simply about translating the principle of feedback into reality. It is about what the data sounds like across different contexts. Which brings us full circle back to our Sunday night resolution. It is a quarter to 10 again. You are sitting at the same kitchen table. The house is still quiet. The same 31 books are there. And book number 17 is back in front of you. But this time, knowing what the data actually rewards, knowing that the tick just closes the cognitive loop and discards the learning, the tick isn't the choice.

The tick isn't the choice. Instead, you change the strategy. You decide which three books tonight get one real targeted sentence each. rather than giving 31 thin ticks across the whole class. Right, focus it. You scan the rest for completion, sure, but you focus your energy. It is the exact same amount of time spent. You are not staying up until midnight. You are still going to bed at 10. But it is fundamentally different homework in a student's hands on Monday morning. Fewer full comments chosen on purpose and rotated through the class over the term instead of 31 meaningless ticks every single week.

Because the feedback is what drives the progress, not the frequency of the marking. So we have some mandatory challenges for you to take into your week. We want to move this from theory into your actual school environment. And I want you to literally do this at break. Ask whoever's next to the kettle. Does our homework policy actually say anything about feedback, or does it just say how often homework should be set? Because if we are honest most policies are entirely focused on frequency. They say year eight must have 45 minutes of geography a week.

They rarely dictate the quality of the loop. Yep. And here's the test to follow that up. Think about this closely whether you're a classroom teacher or a senior leader. If a parent asked tomorrow why homework in one subject gets a comment back and another just gets a tick. Could you point to a whole school answer or would every department give you a different one. That is the ultimate stress test of a policy. If every department gives a different answer, you don't have a whole school policy, you just have a bunch of silos.

So what are the immediate takeaways from all of this? Let's distill it down. First, an immediate action for the very next time you are setting work. Next on Workset, pick three pieces to get one specific comment each. Do not spread a thin comment or a quick tick across all of them just to feel like you've marked it all. Choose three because feedback beats frequency every single time. And our top three review of the data itself from the EEF. Number one, the split is real. The data shows plus five months in secondary and plus three months in primary.

Number two, the why isn't proven. Do not repeat the mechanism of older kids are more independent as if it is a proven fact. It is the story we invented to fill a gap in the data. And number three, Feedback, not frequency, is what the data actually rewards. That is the lever you pull if you want the time you spend to actually matter. Everything we talked about is linked in the show notes. The toolkit page, the techno appendix, all of it. We want you to read them, to look at the padlocks yourselves, and to challenge the assumptions in your own staff rooms, because we're the trailer, not the film.

Both voices on this show are synthesized. The judgment isn't. Safe trip in. See you at the gates.