'Fewer Problems, Greater Detail': EEF's Line Instead Of A Figure

This is the education commute. Both voices on this show are synthesized. The judgment isn't. Right. So if you are a teacher listening to this, I am currently looking at a piece of paper that represents what feels like the Holy Grail for math departments across the country. The Friday problem solving sheet. Exactly. Because we all tend to believe that, you know, doing more problems makes people smarter. It's a very common belief. Yeah. But what if the sheer volume of work we are giving them is the exact thing stopping them from actually building transferable strategy?

That is the big question. I'm in the subject lead's chair today, and I've got my exact routine right here. It's 12 problem -solving questions, one per pair, laminated, and I have been running it every Friday since September. Okay, so we are talking about 12 distinct problem -solving questions handed out in a single lesson. I mean, give or take, depending on the class, but yes, that is the target, and I will vigorously defend it. It is a solid, reliable routine. The pupils know exactly what is expected.

When they walk in the room, they sit down. And they get to work. It is definitely a routine. And I can completely see why it feels efficient. But 12 varied problem -solving questions at a single 40 -minute session is a massive volume of material to push through. Well, the volume is the whole point. Is it? Yeah. The logic behind it is incredibly straightforward. More problems, more practice, more prepared for whatever comes up in the real assessment. I hear that. As a subject lead, you know. I am accountable for their final grades.

Yeah. When they sit down for that exam, they're going to face a high volume of varied questions. Sure. If we have spent every Friday since September grinding through dozens of different problem types, algebra mixed with geometry mixed with ratio, they are mathematically fitter. Mathematically fitter. Yeah. They have seen it all. I can point to that laminate sheet and say, look, they did the work. It's a very reasonable sounding theory. It really is. It has an intuitive logic to it, and it feels safe because it mirrors the format of the assessment.

But I'm stepping into the researcher's chair for this one, and I want to look closely at the guidance report from the Education Endowment Foundation. For anyone unfamiliar, the Education Endowment Foundation is an independent body that synthesizes global educational research to tell us what actually improves learning. And I have read their guidance report's problem -solving recommendation. Right. And I'm guessing it doesn't align with my sheet. Not exactly, no. I've also been looking at the specific blog that picks a fight with exactly your Friday sheet. I don't see how anyone could pick a fight with a full hour of actual problem solving.

I mean, I am literally giving them exactly what they need to do in the assessment. We are practicing the exact thing they need to get good at. The blog is literally titled The Problem with Problem Solving in Maths. Wow. Okay. Yeah. And its core argument is that spending an hour completing a sequence of different problems, which is exactly what your 12 questions are, is the wrong unit of practice. Hold on. How can a problem be the wrong unit of practice? Well, think about what they are actually doing.

If the assessment is made of problems, practicing those exact problems seems like the only logical unit of practice to me. And that is the exact assumption the guidance challenges. We assume that if a pupil completes a math problem, they have practiced the skill of problem solving. Right, because they have. But the actual recommendation favors fewer problems examined in significantly greater detail. The blog argues that the completion of a sequence of different problems is actually hiding the mechanics of what makes a good problem solver. Hiding the mechanics.

Yes. When they rush through 12 different questions, they aren't practicing strategy. They are just testing whether they can already calculate the answers. So you want me to drop the volume? Effectively, yes. Are you telling me I should be handing out three questions instead of 12? That is the direction we are heading in. Because if I hand out three questions... My pupils will finish them in 10 minutes and spend the rest of the lesson staring at the wall. And we need to be very careful with how we apply this.

It is not about doing fewer problems just for the sake of doing less work. Right, because that would be a disaster. Exactly. It is fewer because each individual question has to do significantly more heavy lifting. If you just hand out three questions and have the pupils complete those three in the exact same silent rapid fire way they were completing the 12, you haven't improved their practice at all. No, I've just given them a shorter worksheet. You have just reduced the volume and created dead time in your classroom.

Yeah. The recommendation actually asks for four named components and absolutely none of them is more questions. All right, let's unpack these components because my current sheet is incredibly straightforward and it doesn't ask them to do anything other than solve. Right. It is just, you know, answer next question, answer next question. Let's start with the first of the four named components then. One, teach pupils to compare different approaches to the same problem rather than moving on to the next one. Okay, but if they got the right answer, why would I ask them to stop and compare how they got it?

Because the answer isn't the whole point. they could be using that time to secure another right answer on a completely different problem type on the sheet because Getting the right answer on one specific ratio problem does not guarantee they have built a transferable strategy for the next ratio problem they encounter. When pupils compare different approaches to the same problem, their focus shifts entirely away from the final number. They start looking at the mathematical decisions. Think about two pupils looking at the same ratio question on your laminated sheet, pointing at different parts of the paper.

Why did this sequence of calculations work faster? Where did this specific start? I see. By comparing, they elevate their thinking from just executing a rote calculation to actually evaluating a strategy. I can visualize that but it demands a complete shift in momentum. which I assume leads directly into your next point. It does. Component two, get pupils interrogating what they already know before they start rather than diving straight in. Look, if you tell a 14 -year -old to interrogate their knowledge before they start, they are going to freeze.

They might at first, yeah. When they have 12 questions to get through, they know they have to dive straight in to get to the end of the sheet. My current routine relies on that forward momentum. And that forward momentum is precisely the issue. Why? When they dive straight in, they're usually just guessing at a calculation based on the surface features of the problem rather than genuinely analyzing it. Okay, so they're just grabbing at numbers. Exactly. Interrogating what they already know requires a deliberate pause. It means looking at the geometry question and asking, what mathematical structures are actually present here?

What prior knowledge is required to even begin? It is the active selection of a strategy before any calculation is executed. I understand the theory, but component one is comparing approaches to the same problem, and component two is interrogating existing knowledge before diving in. Yes. Both of those require stopping. My routine is built entirely around answer next question, answer next question. Which is the exact failure mode, the blog name. Failure mode. Yes. Your Friday routine is focused on doing more, but it results in doing significantly less of the specific talking and analyzing that actually builds transferable strategy.

Okay. I am starting to see how my sheet optimizes for speed over depth. Let's look at the rest of this then. What else is the recommendation asking for? Component three, worked examples. So pupils can analyze how a strategy was used, not just whether an answer is right. We absolutely use worked examples when we teach a brand new concept on a Monday or Tuesday. Most teachers do. But on the Friday problem solving sheet, it is just the pupils and the blank questions. are you suggesting the practice itself at the end of the week should still include worked examples yes and the reason comes down to cognitive burden okay unpack that for me when a pupil looks at a completely blank problem during practice they are forced to juggle two incredibly difficult things at once right they have to figure out the overarching strategy of the problem and they have to perform the actual calculation yeah which is hard that overwhelms their working memory When they look at a worked example instead, they are not under the cognitive burden of having to perform the calculation themselves.

So because the calculation is already done for them on the paper, 100 % of their brainpower can be directed toward the strategy. Exactly. Their entire focus is directed toward analyzing how the strategy was deployed. That makes sense. They are tracing the steps, unpacking the logic behind each mathematical move, and learning the architecture of the problem. If they only ever see their own final correct answers, they never see the strategic decisions that led to success. That does shift the Friday focus from purely generating answers to understanding the pathway.

What is the fourth component? Component four, get them monitoring, reflecting on, and communicating their own problem solving. Saying out loud what they tried and why, not just submitting a number. Saying out loud what they tried and why is a completely different lesson from mine. I figured. My Friday routine is silent. Pairs. 12 questions. 40 minutes. The silence is how I know they are working hard. The submission of the number is how I know they have finished the task. The silence in your room might indicate compliance, and the submitted number might indicate completion, but neither of those things guarantees that they are building problem -solving skills.

Self -explanation, communicating out loud what they tried and why, forces cognitive clarity. How does speaking out loud do something? that thinking silently doesn't yeah i mean thought is thought Right. Because when a pupil works silently, their brain can easily skip over gaps in their own logic. But when they have to verbally justify their strategy to someone else, they have to linearize their thought process into language. They hit a speed bump the exact moment they realize they don't actually understand why they did a specific calculation. They catch their own mistakes.

Yes. The act of speaking the strategy out loud exposes their own misconceptions and crystallizes their understanding in a way that silently writing down a number simply does not. Let me make sure I have the four named components clear. Go for it. Teach pupils to compare different approaches to the same problem. Yes. Get them interrogating what they already know before they start. Correct. Use worked examples so they analyze how a strategy was used. And get them monitoring, reflecting on, and communicating their own problem solving out loud.

You have it exactly. And when you look at that list, you realize that absolutely none of those four components are satisfied by handing out a laminated sheet of 12 questions and asking for 40 minutes of silence. I feel like I need to defend the sheet a bit more here. Fair enough. Because the pupils do genuinely work hard on these Fridays. They are seeing geometry. They are seeing algebra. They are seeing ratio, all mixed together week after week. That kind of retrieval has to be worth something.

And I want to be entirely fair to what your sheet is actually doing. Repeated exposure to varied problem types isn't worthless. And I'm not telling you to bin it out, right? Okay, good. There is immense value in pupils having to retrieve different mathematical concepts rapidly. It builds a very specific type of fluency. I can hear the butt coming. But if it is the whole diet. Every single Friday, with nothing swapped out for talking comparison, you're optimizing for the thing the evidence says matters least and skipping the thing it says matters most.

It is an opportunity cost. Optimizing for the thing that matters least. That is a hard pill to swallow for a subject lead who has spent months building this routine. I know it is. If I'm going to drop the 12 question sheet, I need to know what a realistic replacement looks like. What does Friday actually look like instead in the exact same 40 -minute slot? It looks like three questions, not 12. We are really going down to just three. Just three. But the magic is in how you use them.

You take the same problem type and run it two different ways by two different pairs. So pair A solves question one using whatever method they choose. And pair B solves question one using their chosen method. Exactly. And then the critical part of the lesson happens. You give them five minutes where those pairs explain their approach to each other out loud before either one's marked right or wrong. Before. is marked right or wrong. So the focus is entirely taken off the final calculation and put firmly on the communication of the strategy.

Precisely. They have to articulate exactly what they tried and why. They have to actively compare their approach to the other pair's approach. They are analyzing how the strategy was used and think about the worked examples component. If pair A got stuck on the algebra, pair B's verbal explanation acts as a real -time worked example. Oh, wow. Yeah. The actual learning happens in that five minutes of talk, not in the isolated silent calculation. That is a massive change from 12, silent, done. It is. My current setup requires very little facilitation once it is running.

I just walk the room. The three questions and talk model requires a completely different type of classroom culture. From both the pupils and the teacher. Absolutely. It moves the entire focus from rapid completion to deep analysis. And if I'm going to make a massive shift like that, giving up 75 % of my question volume, I need to know the payoff. I understand that. When I look at the research for something like peer tutoring, the guidance gives us a clear, quantified number. We call it an effect size, a mathematical way of showing how many months of progress an intervention has.

Right. Feedback has a hard number attached. How much faster does this new problem -solving routine improve their assessment results? What is the headline figure? And here's the honest bit, and I know it's going to frustrate you. Okay. I can't promise you it improves test scores faster than your current sheet does, not with a number attached. Wait, you are telling me you can't promise a number, this recommendation has no headline figure at all? No. The specific recommendation doesn't carry a headline figure the way peer tutoring or feedback do.

You are asking a subject lead to completely change a working, reliable routine on the strengths of the guidance thinks this is better rather than here is the exact effect size. I know. I can count how many questions my people's completed on the 12 question sheet. Yeah. It is quantifiable. That is exactly what's being asked. And I know that is a tough sell. It really is. But the evidence for why we should do this is solid. Worked examples and self -explanation have a long proven research history behind them generally.

Right. When people self -explain, their structural understanding of the mathematics deepens. They stop treating every problem as a brand new event. But this specific recommendation doesn't get its own quantified number. And I would rather tell you the honest truth than invent a number just to win the argument. I respect the honest concession. It is far better to tell the truth than invent an effect size. Exactly. But it does leave me in a very tricky position with my department. The 12 question sheet is safe. This new approach relies entirely on the quality of that five minute explanation between the pairs.

If they talk about the weekend instead of the math, the whole routine collapses. It absolutely relies on the quality of the talk, which is why interrogating knowledge and comparing approaches has to be. actively taught, not just expected, measuring the immediate impact of that deepened understanding is far more complex than tallying up correct answers on a Friday afternoon. I am willing to look at the reality of this because I want my pupils to be robust mathematicians, not just calculators. Which is the goal. But I am not going to throw away a working routine overnight based on a theoretical argument without a headline figure.

I will give you a trial, not a conversion. A trial is exactly how we should approach a structural change like this. We will split the cohort, half my classes keep the 12 -question sheet, and half do your three -in -talk version. Okay. We will compare what happens over half a term. We keep the time allocation identical, 40 minutes every Friday. We just completely change the unit of practice for half the pupils. That is a genuinely fair test, and it's fairer than either of us just asserting our side.

Right. Over half a term, you will see the difference in how the pupils tackle complex problems. You will see whether the repeated exposure of the 12 questions holds up against the deep analysis of the three questions. It will definitely highlight the difference in classroom dynamics. It actually makes me think about what our current practice looks like to an outside observer right now. It is worth reflecting on. Let's pose a challenge to the listener based on exactly that. If you are listening to this at break, ask whoever is next to the kettle.

In your last problem solving lesson, did pupils ever explain a strategy out loud to each other or did they only ever hand in an answer? It is a highly revealing question because if the only thing being handed in is a final number, the strategy remains completely invisible to both the teacher and the pupil. Exactly. And I will add a second challenge for you to consider. If a colleague asked to see evidence that your pupils can compare two approaches to the same problem, not just complete 12 different ones, could you show them a lesson where that actually happened?

If the honest answer is no. That is this recommendation's whole point summarized in one sentence. Yeah. We are assuming they are building problem -solving strategies through sheer volume, but we aren't actually creating the classroom conditions for them to analyze, compare, or communicate those strategies. Exactly. When we hand out 12 questions and demand silence, we are optimizing for doing rather than optimizing for understanding. It is a massive amount to consider for the Friday routine, especially for a subject lead. Let's pull this all together into the main takeaways.

Four things off the drive. One, no headline figure for this recommendation. The evidence for talk and worked examples is strong, but this specific recommendation isn't quantified. Two, the guidance's actual finding is fewer problems examined in far greater depth, not more problems completed faster. The wrong unit of practice is the sequence of different problems. Three, comparing strategies, interrogating existing knowledge, worked examples, and self -explanation are the four named components, not practice more. We want them communicating out loud what they tried and why, and analyzing how a strategy was used.

Four, a silent answer and move -on sheet isn't wrong, but as the whole diet, it optimizes for the part the evidence rate's lowest. Repeated exposure has its place, but not at the expense of transferable strategy. Everything's linked in the show notes. 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.