Ep 64. Rosenshine's Principles of Instruction with Tom Sherrington
This transcript was created with speech-to-text software. It was reviewed before posting but may contain errors. Credit to Canadian Podcasting Productions.
In this episode, Anna is joined by Tom Sherrington, education consultant, author of Rosenshine’s Principles in Action, co-author of the Teaching Walkthrough series, and a former teacher and school leader with over 30 years of experience. Tom shares how Rosenshine’s Principles of Instruction became a foundational framework for his own practice and why they remain one of the most useful entry points into evidence-based teaching.
They explore where the principles came from, why they matter, and what they actually look like in real classrooms, with a focus on sequencing and modelling, effective questioning, review, and structured practice. This practical, research-informed conversation offers clear, concrete guidance for educators at all levels who want to align their teaching with how learning works.
This episode is also available in video at www.youtube.com/@chalktalk-stokke
SHORT COURSE
La Trobe Short Course: Evidence-informed Mathematics Teaching – An Introduction https://shortcourses.latrobe.edu.au/evidence-informed-mathematics-teaching
PRINCIPLES OF INSTRUCTION RESOURCE
Principles of Instruction: Research-based Strategies That All Teachers Should Know by Barak Rosenshine: https://www.aft.org/sites/default/files/Rosenshine.pdf
TOM SHERRINGTON WEBSITE https://teacherhead.com/
TIMESTAMPS
[00:00:22] Introduction and listing the 10 principles
[00:04:19] Who was Barak Rosenshine?
[00:04:45] From 17 evidence-based instructional procedures to 10 principles
[00:08:18] What research backs Rosenshine’s principles?
[00:14:54] UK teachers’ familiarity with Rosehnshine’s principles
[00:14:08] Pace of adopting Rosenshine’s principles in other countries
[00:18:07] The more practice the better: Rosenshine’s on practice
[00:23:06] What does aiming for an 80% success rate mean?
[00:27:09] Importance of checking for understanding
[00:28:39] Checking for understanding and good questioning
[00:32:55] Understanding the forms of guided practice
[00:34:33] Teaching between desks
[00:38:54] Principle 1 & 10 – The reviewing stages
[00:43:51] When to review before teaching your lesson
[00:45:11] Presenting new materials and small practice steps
[00:49:47] Principle 4 – Providing models
[00:51:34] Principle 8 – Provide scaffolds for difficult tasks
[00:54:59] Do these principles apply to any subject?
[00:57:40] Can we get 80% success in classrooms with a wide range of learning levels?
[00:00:00] Anna Stokke: Welcome to Chalk & Talk, a podcast about education and math. I'm Anna Stokke, a math professor and your host. Welcome back to another episode of Chalk & Talk.
Today I'm joined by Tom Sherrington, an education consultant, author of Rosenshine's Principles in Action, co-author of the Teaching Walkthrough series, and a former teacher and school leader with over 30 years of experience. When people ask me where to start with evidence-based teaching, the first thing I usually point them to is Barak Rosenshine's Principles of Instruction. They're clear, practical, and grounded in research, and they come up a lot on this podcast.
In this episode, Tom and I talk about where Rosenshine's Principles came from, why they matter, and what they actually look like in classrooms, with particular attention to sequencing concepts and modeling, effective questioning, reviewing material, and the stages of practice. This is a conversation that's useful for educators at all levels who want concrete, evidence-informed guidance that they can use in classrooms. Now before we get started, I have an exciting announcement.
I'll be co-delivering a four-session online short course on evidence-based math teaching through La Trobe University's School of Education starting in April 2026. The course is open to teachers anywhere in the world. I'll include a link in the show notes for registration.
Now on with the show. I am thrilled to be joined by Tom Sherrington today. He is an education consultant and author.
He has 30 years of experience as a physics and math teacher, or math teacher, and school leader. He writes the popular blog, teacherhead.com. His books include The Learning Rainforest and the bestseller Rosenshine's Principles in Action, which we're going to talk a lot about today. He also co-authored the teaching walkthrough series with Oliver Caviglioli.
And he works with schools around the world to provide training and consultancy support on improving the quality of teaching through the walkthrough materials. He co-hosts the podcast, Mind the Gap, which I had the pleasure of being on not too long ago. Welcome, Tom. Welcome to the podcast.
[00:03:45] Tom Sherrington: Oh, thank you for welcoming me. I'm really honoured to be on your show.
[00:03:48] Anna Stokke: Thank you. We're going to talk a lot about Rosenshine's Principles of Instruction today. Barak Rosenshine published a wonderful paper that we've talked about a lot on this podcast called Principles of Instruction, Research-Based Strategies That All Teachers Should Know.
And you've actually written a book on Rosenshine's Principles, so you're the expert here. Before we get into the principles themselves, who was Barak Rosenshine? What was his background?
[00:04:19] Tom Sherrington: So as far as I understand it. He started off being a history teacher in the 60s, which he did for a few years, but then he went into higher education and then he studied in education as a phenomenon, really.
In the 70s, 80s, 90s, he was a researcher working. I think he's from Chicago, but I mean, I could be wrong about that. He had a career as a researcher studying what teachers do in classrooms.
And he came to prominence in the UK through this very condensed version of his work, which we'll talk about. But I know various people knew he had a higher profile before, but I don't think he's as well known as he deserves to be probably. And he died a few years ago, so we can't really talk to him about it directly.
He has left a legacy of papers and videos online, but that's essentially who he was. He was a teacher and then he became a researcher, and then he became someone who promoted these ideas about instructional teaching.
[00:05:16] Anna Stokke: He was an American teacher who became a researcher, and he left a great legacy because I think when people ask me, what's something that you can point me towards about evidence-based instruction, that's usually the first thing I'll say, is Rosenshine's principles because they're so clear and they just make a lot of sense.
My understanding is actually he had come up with 17 evidence-based instructional procedures and that those informed his 10 principles. Is that correct?
[00:05:47] Tom Sherrington: Yeah. And that was an evolution from earlier work.
One of my favorite things, since I wrote that booklet about Rosenshine's principles. People send me stuff. One of my favorite things is a paper he wrote in 1986, which is called Instructional Functions.
And that's got six things. And over time, kind of added a few. And then by the time we get to 2010, when he wrote the paper, which is the one we know, he was forced to summarize it.
There's a journal called the International Academy of Education Journal, it's very weird, long, thin booklets. And he published the principles in there, but he was asked to do them as 10. And so, he morphs the 17 into 10.
And then that was republished in the American Educator magazine in spring 2012, which is an incredible issue because it's got other things in there by Kirshner, Sweller and Clark and Dan Willingham. So, it's like an epic issue. And that was the thing that got popularized in the UK, that those 10 principles.
I don't know how he felt about it, but some people prefer the 17 because they're more precise, but it's a lot of things to remember and go through. So, it gets a bit overwhelming.
[00:07:01] Anna Stokke: I think 10 is a good number. The 17, there's a bit more detail, right?
[00:07:07] Tom Sherrington: Yeah, they're very specific things. They're literally, as precise as begin a lesson with the review of prior learning or something. And you think, OK, that sounds sensible.
And I always say to people that even though they seem sensible on the surface, there's a lot in them. One of them is provide many examples. And you think, OK, so which examples of which type? There's so much to unpack in every single one of them.
And even things like check the responses of all students and that little cheeky all in there. He doesn't say check the responses of a few students, he says of all students. And you think, well, OK, so even though they seem kind of common sense on a surface level, each one of them takes a lot of thought to think, OK, well, how do I actually do that properly? I think it's a very rich source because everything in there is not as obvious as it seems.
And so, teachers have to think quite hard about doing them well.
[00:08:07] Anna Stokke: So, they're research based. Did he conduct that research himself, or did he come up with them based on research done by other scholars?
[00:08:18] Tom Sherrington: He says, in the paper that is popularized, it's a little bit obscure, to be honest, because you have to know that he did research.
One reading of it is that it was imagined he hadn't known it himself because he reports it in the third person as research was done or has been done. He's not saying we did this. But when you know about his research, he says there are three types of research.
One of them is cognitive science study. So, when he's citing various people who have done studies into memory and learning, that's not his research. He's also talking about some intervention studies.
Things like reading, reading studies, experiments with types of scaffolds. That's another type of research. And then the third type, which he's contributed to, but he also cites others, is observational, watching teachers who get good outcomes and comparing them with teachers who get worse outcomes and codifying kind of what they do and what the differences are.
And he cites various other people who have contributed to that research. And it's an interesting debate to have because some people, I don't know why, but I suppose that some people contest the idea that Rosenshine's principles are very, strictly speaking, evidence informed in the sense that each one of the principles has been subjected to a specific study. It's a bit more subtle than that.
It's like a combination of sources coming together. And then he's proposing them as mainly features of practice that teachers demonstrate when they're getting good outcomes, which also are supported by the evidence from cognitive science. And he says very clearly, there's no conflict between those things.
So, we should have confidence. And I always say that in my training sessions, that imagine if cognitive scientists were saying, the research about memory and learning suggests teachers should do these things. But weirdly, when we go into classes where the outcomes are good, they're doing totally different things.
That would be confounding. But that isn't what happens. And Rosenshine makes a big point of saying there's no conflict between them.
So, when you observe teachers getting good outcomes, the things they do are supported by the evidence from cognitive science. I think that's the spirit in which he says they’re evidence informed. He does cite various studies for each of them.
But I do think there is a kind of interpretation and assimilation of ideas packed in there.
[00:10:40] Anna Stokke: It's kind of like there's three things coming together. There's cognitive science research, there's experimental research and there's observational research.
And they all kind of land on the same things. Like, for instance, practice. Nobody should really be disputing that practice is important, is basically verified by those three types of research.
[00:11:02] Tom Sherrington: What I like about that is, in practice sounds obvious, but he puts quite a lot of parameters around that. The nature of the practice, high success rate, practicing success and only moving from guided practice to independent practice when this has been a success rate. So, yeah, no one should contest it.
But it's not as obvious as just, oh, yeah, go practice. It's like, what exactly are you practicing? He's very precise about that. And I think that's important to always say those things that it's the precision of it, which I think makes them useful.
[00:11:03] Anna Stokke: I agree with you. And that's one of the things I definitely want to talk about today is practice and the type of practice and how you know when someone's had enough practice and that sort of thing. We're going to come to that right away.
I wanted to ask, because you mentioned that in the UK, people use Rosenshine's principles or they know of Rosenshine's principles. Would you say that most teachers in the UK know about Rosenshine's principles now?
[00:11:56] Tom Sherrington: I don't know if it's most. I think if you had to sort of score the things which are most known, Rosenshine's principle would be one of the highest.
But that doesn't mean like in any group of teachers, say more than half of the teachers will know exactly, even if they've heard of the name, because it gets bandied around. They won't really know what it is, because a lot of teachers are very subject focused. And so, their reference points are all about subject curriculum.
And so, something generic or like Rosenshine isn't something they ever encounter or really think about. At some schools, I go there to do training and I'm expecting, we know, and actually they don't, like how many of you read Rosenshine's principles? No one. Oh, wow.
Okay. No one's read it. Okay.
Interesting. Because you're sort of thinking, where have you been? But then the reason, where have they been is not on Twitter or something. It's quite common for teachers in a whole group of people not to have given any attention whatsoever.
But I don't think there are any things more well-known than Rosenshine, put it that way, that shows you kind of how any one particular reference is never that well-known, even something like teach like a champion, Dudlamov. Most teachers, I would say, have heard of growth mindset, but have never read Karol Dweck directly or the debate around it. You just know about that it's a thing, but you haven't heard of this.
You know what I mean? It's quite hard to gauge. And even if I asked someone to what, if I tested anyone, I know I don't actually do this, but you sort of sense that the detail behind the label, Rosenshine's principles, isn't that well-known.
[00:13:37] Anna Stokke: Do you work with teachers in the US and Canada sometimes?
[00:13:41] Tom Sherrington: I have. I mean, I've only just a small degree. So yeah, I've been to a research head in Canada and US and I've been to, I did some work in near Winnipeg, Manitoba earlier this year, which was fantastic.
[00:13:56] Anna Stokke: I'm just curious if you've noticed that maybe the UK has been moving forward a lot more quickly with these types of evidence-based instructional principles.
[00:14: 08] Tom Sherrington: Oh yeah, I think so for sure. And Australia is quite fast moving as well, I would say. And to the point now where in the UK, there's a move, which is you're pushing against an open door. So, people want to know, and they're asking for more, and we're not trying to fight the battle of why we should be talking about cognitive science.
And that we're not having to sort of talk endlessly about inquiry-based learning. People understand that the memory and retrieval practice and various other things are important ideas, and I think that's, we've come a long way. Whereas in the US you still have some debates, which I've been involved in at certain events, which I haven't had in the UK for a long, long time.
Where people sort of even contest the idea of whether or not children should know facts and stuff. And think, well, okay, we're really at that point. It is interesting the journey different places are on.
[00:14:58] Anna Stokke: We've got to start moving more towards that evidence-based instruction soon. That's my hope. But absolutely, you will definitely hear that in Canada.
There'll be people who say learning facts doesn't matter. You can just Google it. That's still happening here.
And I would guess that not many people have heard of Rosenshine's principles here.
[00:15:21] Tom Sherrington (15:21 – 16:16)
It really is interesting that it's always something that's so slightly confounding that most of the people that we refer to in the UK, are, you know, Dan Winningham, people like Dudlamov, whether it's Rosenshine, all sorts of people, Fiorella and May or Shimomura, and these sorts of people who have been very influential, I think, are from the US or North America generally. And they're not, and E.D. Hirsch, people who've been celebrated for breaking through in ideas about not knowing things and the research.
And even like movements like Deans for Impact and these sort of places where talking about how teachers learn, big influence, but they probably are at scale less influential in the US than they are in the UK, because we're more condensed. We're more homogenous in our system. So, the ideas get disseminated more intensively, I suppose.
[00:16:16] Anna Stokke: Oh, you think that's what it is?
[00:16:18] Tom Sherrington: I think there's a lot to do with it. I think the explosion of social media in a sort of 2012, 13, 14, which connected practitioners to policymakers, to researchers in one space with the same curriculum, the same national system. I think that was very powerful as a way of mobilizing information about learning.
And in the US, that's just nowhere close to being that dynamic because it's just so big.
[00:16:49] Anna Stokke: We should talk about Rosenshine's principles. And I like the way you do this in the book, and I do recommend for people listening that you get the book because it's a nice read. It's a quick read.
I think it'd be really helpful for helping your teaching. You actually condense the 10 procedures into four strands. So, you have sequencing concepts and modeling, and then you have questioning and then reviewing material and then stages of practice.
So, I thought we'd go through some of those strands. We'll see what we can get through. I just would like to talk about practice.
We've talked a lot about things like modeling and explicit instruction, the actual how you break things down, that sort of thing. We might come back to that, but let's start with stages of practice.
So, we have three principles that you put in that category. Principle five, guide student practice. Principle seven, obtain a high success rate.
And principle nine, independent practice. Rosenshine states that the best way to become an expert is through practice. Thousands of hours of practice.
And the more the practice, the better the performance. And I would completely agree with that. So, what else does he say about practice?
[00:18:07] Tom Sherrington: Well, he basically says this thing all the way through his principles, which is,
he says more effective teachers typically do this compared to less effective teachers. This isn't a kind of magazine style write-up.
And he says more successful teachers spend more time guiding students' practice of new material than less effective. And guiding the practice is key because he has a very big thing about success rate and to get a success rate around 80%. He cites a study, and to be honest with you, this is one of the things in the principles I'm slightly skeptical about the actual numbers.
So, he cites a study where, for example a fourth-grade maths class. It was found that 82% of the students' answers were correct in the most successful teachers, but the least successful teachers had a success rate of only 73%, I thought 82 to 73, that feels like margin of error in the measurement. It feels a bit too neat to me.
It's not that different, but basically the point is he's saying high success rate is a factor of classrooms where the outcomes end up being high and you might think that sounds like a sort of self-fulfilling thing, but it's actually to do with the construction of the lesson. And the more you're practicing getting things wrong, the more habits you're forming around bad methods in math, for example, or inappropriate language or incorrect language in English or whatever. It's a concept of high success rate with bridges between guided practice and independent practice.
And in order to get the success rate high, which has many benefits, it builds confidence, but it also has challenge. It's not 95%. Notice he says it has to be challenged.
You have to be pushing towards something slightly ahead of you. And that links with the small steps idea. So, while I felt like about the principles, the intersect, so guided practice around small steps.
We don't guide practice by saying like, here's five steps you need to go do and then you think, here's a step. Let's see if you can do that. Okay.
Here's the next step. Let's all make sure you can do that. So, you guide, you interact with the class all the time.
Instruct, check for understanding, guides and practice, instructs them more, check for understanding, guides and practice. It's very incremental step by step, small steps, and you vary the steps according to the success rates, according to the confidence or background knowledge of the students. And the more sophisticated the students are, the more you can get them to practice at once.
And you have to adapt and your guiding thing is the success rate. If the success rate is low, well, you may be getting them to practice too much at once. So, you break it down to small steps, maybe instruct a bit more to get the success rate up again.
And success rate to me in this context is a nominal idea. It's very hard to measure in any meaningful sense. It means most of them are getting most of it right, most of the time.
And you're looking for that to be as high as you can make it. And I think that's essentially the spirit of it. But he then goes on to say more.
He says, one of the key things to have in mind is the independent practice to come, that you're not going to let them do things on their own, which you haven't guided them to succeed in the first place. And I was talking about this with my wife recently, you know, he's a deputy principal of a secondary school. And this weird phenomenon, the teacher going through the server easy ones, then says, right, now you try these hard ones, but you haven't modeled the hard questions, which you've now given them to do as the extension from your simple guided, you've got to model the harder questions, get them to be successful at the harder questions with guidance and now say, right now, try some of these harder questions by yourself because you've shown you can do it with some help.
Now, see if you can do it without the help. And that's why I think Rosenshine is genius. It's not that it is well established in people's thinking.
A lot of teachers think, oh, well, it's better if they've done it by themselves and struggled it out. And he's saying, no, it isn't. It's better if you've guided them to be successful and now see if they can be successful independently doing the same thing.
That's the key part of this whole thing about practice. And he says it explicitly where students are practicing things independently, which they have not previously succeeded at with guidance. The outcomes are worse.
It's about building fluency and confidence with things you've already succeeded at. And I think that's, it's not like earth shattering, but it's quite significantly different to what some teachers actually do in their lessons.
[00:22:46] Anna Stokke: Let's just talk a little bit more about the 80% success rate, because that could mean a lot of things.
Do we mean that you can move the individual student on to independent practice when the individual student is generally getting things right? 80% of the time? Is that what that means?
[00:23:06] Tom Sherrington: I've been asked this many ways, and I think I have read the section over and over to like, am I sure I'm interpreting this correctly? It definitely does not mean 80% of the students succeed. I've been asked that. If 20% of them aren't quite getting it, do you move on anyway? No.
All of the students need to be getting 80% success rate. Some of them might arrive at that point sooner than others.
And that's just a reality. So, you have to be ready for lessons to support students to move to independent practice when they're ready. And that means you have to have things like tiering in question types or structured tasks in other subjects where you need to be able to do this.
This is what I'm focusing on as you're learning, but when you've done this, move to that. So, the guided practice, the independent practice phase can form a natural extension. So that's what it means.
Every student needs to be getting that 80% success rate, which means a lot of them need a lot more help than others to get to it, but that's the goal.
And obviously the more homogenous your class is in terms of their alignment with learning rate and the more you can keep them together, the better, which is like a mastery approach. It's helpful because you're keeping people with you and you can move together. But there are times where obviously where, and you see that subtle forms of different levels of guidance, independent work in subjects beyond maths, in maths, I think it's kind of one of the easier ones to do because you've got, as long as you understand curriculum design and can sequence questions in a ladder of difficulty that's sensible, I see sometimes maths problem sets, which is really poor.
You know, you've got three easy ones, three middle ones, and three hard ones. Look kind of everyone's doing all of them. And it's so obvious that some students need like 10, 15, 20 easy ones to get that more fluid, and then other students are really ready to fight right to the end, do the hard questions in the same room, and that is where they're at.
You know, they're all algebra. They're all whatever you have to practice at the point where it's going to make the biggest impact on you. And that's the success rate is in that now.
That's why you end up having sets, ability grouping in England quite commonly, because you can no longer sustain a practice regime that's manageable practically. And that's what happens because if you've got divergence, and a bit in success rate, that's too big. But I think that's the point of it.
He actually says one of the issues with high success rate is about misconception.I quite like the section. I'll just read it to you.
“When we learn a new material, we construct a gist of this material and long-term memory. However, many students make errors in the process of constructing this mental summary. And these errors can occur when the information is new and the student did not have enough background knowledge.
These constructions were not errors so much as attempts to be logical.” So, they're trying to make sense of things. And so, they form a kind of a way of working with the material and that can get stuck.
The danger is not just when you're practicing material without enough secured knowledge. It's not just getting wrong and being demotivated. It's actually, you're forming sort of intellectual coping mechanisms or schema for the material, which are wrong.
So, you're not learning cheats and stuff, which only work for certain examples, and then they become unstuck, that type of thing. You don't want that. You want to practice doing the long method and doing it fully and properly.
So, they fully understand it rather than thinking they can cut corners because it seems to work. Providing guided practice after teaching small amounts of checking can help limit the development. That breaking down of short feedback loops, that's the language I would use is short feedback loops.
Explain, check, practice, do one like I just done, check it's right. What did we get? Does everyone understand it? Here's another one. Try this one.
And okay. It's really to and fro, and rather than less effective teachers typically go like, explain, explain, explain. Right off you go.
Practice, practice, practice, practice.
[00:27:05] Anna Stokke: Yeah. You want to be checking for understanding all the time.
[00:27:09] Tom Sherrington: And from all students as well. This is what I love about it. It's subtle, but it's absolutely in there.
Rosenshine is all, all, all, all, all. You have to have this radar, this sense for where the success is. I was talking to teachers today in the school, which is very successful school recently, some great accolade.
Literally ranked in the newspaper as being the best school in England, which is kind of unbelievable really, but it's based on various things, but the teaching was very good. But their issue is who are the five students in your class who are the lowest achievers? Do you even know who they are? Are you looking to see how successful they're being? Because you've got all this wave of positive success coming at you constantly. It's quite easy for those five to be sort of drowned out.
And that's what they were saying literally is their issue because that radar for, “oh yeah, I need to keep an eye on Jason because he finds this difficult and I'm going to check his success rate”, it's much more individual and teachers need to be looking and checking and looking and checking, which is hard, but necessary.
[00:28:18] Anna Stokke: And so that comes under the questioning category, what you call the questioning category, there's principle three, which is ask questions and principle six, which is check for student understanding. So, how frequently should you be asking questions and what does good questioning look like?
[00:28:39] Tom Sherrington: How frequently is a hard thing to quantify, but basically, he decides to study in the paper about a maths class and he gives one example where he says there's a maths, particular maths lesson where typically they were spending 25 minutes out of a 40 minute session checking for understanding and questioning, versus a teacher in the same place getting, doing 10 minutes and the 25 minutes out of 40 was much more successful because a teacher, and this is important part about this, the purpose of the questioning, which intersects with the guided practice, because they're not like separate whole phases that they're interspersed.
The questioning serves two functions. One of them is the students have a chance to think and elaborate. So, in formulating answers, if the culture is such that anyone should answer any of the questions, they all have to process the information mentally and then start making models and so on.
Benefits them elaborating the idea, but also it reveals to the teacher where the errors are and which bits to reteach and what bits of understanding are. So, it's important to have that double sense that I've come across teachers where one or other of those isn't fully present in their thinking. So, for example, did you make all the students think of the answer? And sometimes no, the routine is to select a student and ask them, select another student and ask them.
Of course, now in the UK, I don't know how popular these are in Canada, but in the UK, show me boards, whiteboards are very popular, almost universal to most schools at some point. The point of them is it makes all the students produce answers and also shows them to the teacher. But sometimes if it's done poorly, some students thinking isn't stimulated because they're not expected to answer every question.
But also, quite commonly teachers that just take the answers and look at them and don't really use it literally responsibly. Now I saw a brilliant lesson today where the maths teacher was checking understanding about some indices, laws of indices, and then saying, okay, a lot of you got that one wrong. So, here's another one like it.
We're going to come to this one. So, he wrote it on the board and then he said, but let's, let's go through really step-by-step laws of indices. He was super precise.
He was making them say things like 10, subtract negative three. And it's like, there was no minus of a minus equals a plus. No way.
It's like you subtract negative three. And so, we add positive three, that type of thing. Brilliant.
It was like, they're all practicing that all practiced it, show me boards and then high success rate. And he said, okay, now we've got that one. Now do this one.
And it was like, there's another one to do. The whole point is I think we're better now. So now see if you can have a go at this one.
So, it was all that thing of questioning, checking for understanding linked immediately to a practice opportunity, responding to the fact that previously quite a number of them have got it wrong. So, he's seeing error, reteaching, rehearsing again, and then practice again, it's like, that's how it's done. And actually, literally a teacher who is very hot on Rosenshine's principle says, it was like, he's as informed as practice and he could see it very evident.
I think that's the key thing that when you're asking a question, part of you is thinking, okay, I'll ask this question and if they're not sure, or a number of them are wrong, we'll do this. And it's just totally ready. You know, that's what's going to happen rather than go, oh no.
However, you're wrong. What do I do next? Some teachers feel like that's what's happened to them. They're not ready for the reteach.
They're just, all they're prepared for is moving on. We do this, we get them right. We do these, we get them right.
And if the students are not getting it right, where are my examples? Where are my resources? They're not totally tooled up in the moment to deal with it. And then sometimes I say, oh gosh, I'll have to come back to that later. Cause they're not sort of anticipating the very real, that likelihood of significant error and it captures them out.
That's the role of checking for understanding. It's supposed to be loops of, okay, that's telling me how well you're doing. Let me now do this.
[00:32:52] Anna Stokke: And what you're describing is also guided practice, right?
[00:32:55] Tom Sherrington: I think it is. I mean, guided practice has two forms.
One of them is circulating. I found a friend, a colleague of mine, John Thompson, who's written this lovely book about excellent teachers, which is essentially anecdotal sort of like, well, it's like a type of study, which is very expressive interviews with the teacher, interviews with the students that they teach, observing them and so on. And there's a great math teacher in this book called Chris McGrain, who's a Scottish math teacher.
And he describes what he calls teaching between the desks. He stands between the desks because he’s just supervising constantly. He doesn't stand at the front and hope everyone's doing okay.
He's in amongst checking, checking, checking, checking, checking. He says he feels he knows how everyone's doing because he's watching it happen. It's very dynamic in that regard.
And so, I feel like that type of guided practice, literally supervising and checking as you go around is one very powerful thing. And the other type is students doing questions and then self-assessing. And then we're talking about which ones did we get right or wrong or more question by question, interacting and finding out that's in math, obviously.
But I felt like guided practice generally is interactive. That's not the off you go, get on with it type mode, and that's the independent practice. Guided practice is interactive.
It's sort of important to get that right, I think.
[00:34:15] Anna Stokke: He called it standing between the desks. Is that what he called it?
[00:34:18] Tom Sherrington: Teaching between the desks.
[00:34:19] Anna Stokke: Teaching between the desks. I call that quality control.
When I have weekly labs with my students, I always say, okay, I'm going to do quality control now. And I go through and check everybody's work to make sure it's of high quality.
[00:34:33] Tom Sherrington: Sometimes it's funny how different things are expressed.
So, this paper called, and anyone can listen and can search it. Instructional functions. There's this beautiful text.
It's kind of tight. It's fabulous. It's like a photocopy of a typed document that Rosenshine wrote as a prep for a talk, 1986.
And he has his, at that point, step three, initial student practice. He describes it in this way. High frequency of questions and overt student practice from the teacher and the materials.
Prompts are provided during initial learning, all students have a chance to respond and receive feedback. The teacher checks for understanding by evaluating student responses, continue practice until students are firm success rate of 80% or higher. And then later in independent practice, so that students are firm and automatic and he calls this seat work.
This is when you're just getting on with like you're banging out loads of algebra problems, looking for overlearning and automaticity. So, it's nothing new.
Now you're ready for the hard stuff. The independent practice is of the stuff you've already done. Getting a 95% or higher, he says.
So I feel like that, that's an important idea that sometimes when teachers think everything's always harder, harder, harder, it doesn't have to be, sometimes it needs to be like consolidatory and repetitive so you have this overlearning and fluency building and sometimes I think we get that wrong, that we set them off to go into sort of bad practice because they don't know how to do the things they're practicing.
[00:36:04] Anna Stokke: Yeah. I think it's often the number one thing we get wrong in math.
Students are given stuff to do that they haven't been taught how to do. That's the first thing. And then the second thing is they just don't get enough practice.
And it's so important. Sometimes it's repetitive and it has to be like, that is literally how you get good at math.
[00:36:28] Tom Sherrington: Yeah. And this isn't from Rosenshine directly, but it's like, it's true of any subject that you start mixing up the mode of presentation and then trying to see if you can apply the same then to other problems, synoptic problems, and where you can identify the problem type rather than being told the problem type, recognizing you need to use Pythagoras rather than trigonometry to find the length of a side or whatever, but that comes from practicing all the other things first, so you can't suddenly be deep ended into being confident with those things, but I mean, this is slightly off the Rosenshine, but one of my bits of this sort of insight that I've gained over the years is from a fantastic math person called Mark McCourt.
He has this fantastic insight about how learning, I mean, we're talking about math, but sometimes I feel like math is just a good metaphor for all the other subjects, but when you're climbing up a mountain, the higher you go and you look back down, you make sense of the journey you've been on so far more clearly than when you were on it at the time, so you might be in a bit of a gully and like thinking, why are we here? Well, where are we going? What's all this for? And you see from above, you go, oh, I get it. That's where we were.
Now it all makes sense. And sometimes you have to support students through that period of uncertainty, trusting the fact that in your sense of the way the curriculum progresses, they'll reach a point where it'll all land, it'll all make much more sense later, but not just yet. I think the tolerance of some ambiguity around nailing total understanding, it can sometimes be a false promise.
They're not going to fully understand it yet. Later, you'll see how things like Pythagoras and trig overlap and work together and are kind of the same thing more or less, but to begin with, they're just topics, which is to teach discreetly. And then later they'll link up and then you can link it to.
And that's true of history, like literature, science, everything.
[00:38:24] Anna Stokke: Yeah. I mean, you often don't see the big picture or get the understanding piece until later.
When you're good at something, you can put all the pieces together and you know how things work, then it becomes easier to, to understand it.
We should talk about review. So, reviewing material is another one. That's another one of the categories.
So, principle one is daily review. And then principle 10 is weekly and monthly reviews. So, it's like a full circle.
It makes sense that principle one would be daily review, right?
[00:38:54] Tom Sherrington: It does. It's interesting because when I organized this into my four strands, the reason I was doing it is because I was just reading the paper as the PDF, it was just being tossed around all over the internet, fantastic free PDF to carry a copy around with me in my bag everywhere. And I was doing a talk on it actually at research ed and I was thinking, I was sort of getting my thoughts together.
I wrote this blog and I just thought, well, 10 things is way too many. Like when you, if you've got like 40 minutes and I realized I was kept on jumping, daily review is the number one and weekly and monthly reviews is number 10, I was thinking, well, to me, you have to talk to them at the same time. But I just thought, actually it's a weird place to start because if you're preparing to teach second step, present new material in small steps is where you really begin and then models and scaffolds.
Which models? So, I felt like the kind of the stuff you can prepare, but if you go into the classroom, other small steps that you're thinking about, how do I explain this? And how big are the steps are in relation to where my students are? And I think that's really fascinating work to do with teachers. I've done this a fair bit now is get them to break down, explain something into steps. And sometimes teachers don't realize how much their knowledge as the expert has masked their need to break it down.
Like they think, gosh, you know that they might not know that. So, I have to explain that and explain that. And they're using words, which you think, gosh, that's another word I've used, which I have to explain.
Yes. Or, you know, all of that, but they don't know all of that. You can't just start talking about acute and acute angles or something all of a sudden.
Well, what are those? Oh gosh. And how should we teach that? And when do we begin? I can understand in terms of a lesson that start with data review, but in terms of thinking about a whole unit of learning, presenting material in small steps is where I started.
And then the model, what do I teach them? Which examples do I introduce? And then lots of subjects. That's really important. Isn't it? Like, where do I start? Which examples do I give to make this abstract idea concrete and sequential and build a schema based on what they know? So, it's so subtle, but you then have to flex it a bit.
And Rosenshine says that, you know, more effective teachers are better at identifying the steps being small enough and getting the students to practice after each one. That's where I start. The review only comes in then.
This is a fascinating piece of work that has just been circulated by Carl Hendrick in this area, which is around where review works and where it doesn't. And where it works is where the students have already been successful. And they're literally reminding themselves of that success.
Where review does not work is where they weren't successful in the lesson before, and now you're doing a quick recap quiz, which they can't do because they didn't do it before. So, you have to get back to the success rate. If the success rate is high and the review is a consolidation of that success, if the success rate is low, the review is confounding.
And sometimes there are just way too many questions in a daily review, but obviously you've got to avoid the illusion of short-term task completion. All of you managed to do this. This used to break my mind.
I should teach maths, for example, and I would get a hundred percent success rate of students, two to the power of four, two multiplied by two, multiplied by two, multiplied by two, that's 16. Yeah, yeah, yeah. We're all getting two to the power of four.
Next day, two to the power of four. You've got all these eights appearing. And so, you realize even sometimes the success rate can be only because of recency, you've just taught them it.
It's buzzing in their sort of working memory pretty much, but they haven't really stored it, and so the review can sometimes reveal, burst your bubble of what Mark McCall calls recency, which I don't know if he coins the words, but he talks about recency being a major factor in that, that you have to reduce that, so testing later in a lesson or reviewing tomorrow to see if yesterday's learning has stuck. That's very true in languages as well. Like when you've got lots of small knowledge items, are they still with you tomorrow? Do we go back? We go back.
So, review has that factor because you've got to sort of make it part of the normal learning flow that we touched back with where we were, so it's so important to do that.
[00:43:19] Anna Stokke: That happens to me all the time. You teach something and they get it and you're sure that you're done.
Like they all get it. And then the next day it's like, you didn't teach it. I think that is a really important reason to review.
And then also just when you're teaching math, because math, as I always say, is very cumulative. So, if you're teaching something new, it's probably going to rely on something that they learned before.
It's good to review those things before you start that particular lesson, right?
[00:43:51] Tom Sherrington: Oh, for sure. And I think sometimes in a way these become sort of routinized, if I can use that phrase in English school, certainly, is that there's sometimes there's too much of a disconnect in my opinion, between the review and the lesson. Sometimes it's very common to see the review being of other topics than the one we're about to follow.
And I think preparing for today's lesson, like activating prior knowledge for learning that's about to come. A weekly, monthly review is supposed to be more of that kind of checking back in. And Rosenshine does make that distinction.
So daily review is more driven to, I like that phrase, activating prior knowledge. It's as simple as saying like, we did all these yesterday, guys. Here's another one.
Just warm yourselves up. Should be straightforward. And often that works nicely. You just go brilliant, nice and familiar. We did it yesterday. Do it again.
And we're ready to build. But if that's unsuccessful, then of course you've got to deal with that at the time.
[00:44:44] Anna Stokke: We talked about three of the four categories. And then the last one, which is actually your first one, because you said the thing that you always start with is present new material and small steps with student practice after each step. And I think your reason for that is because if you're the one designing the lesson, you want to think about how you break that topic, that concept down into steps and what steps are involved, right?
[00:45:11] Tom Sherrington: Yeah, exactly. So, I mean, it's something which I, in my Rosenshine's booklet, the Pentel's booklet, I use as an example is, and I do try to use many other examples, but there's a particular math example in it, which I tell the story of a student I encountered in a college, in England, a college would mean something like 16 to 18, but retaking the compulsory maths exam.
So, the GCSE in maths is graded one to nine. And unless you get a grade four, you have to take it again. So, if you've got a grade one, two, or three, you have to take it again.
So you've got all these classes of students in colleges where they have to be taking maths, having failed it before. As he's a student of 16 and this boy who was doing fractions and he'd written as a third, the question was a quarter plus two thirds, and he'd written three sevenths. So, when I say to him, yeah, I can see what you've done there, but I will tell you straight away, I know that's wrong because three sevenths is smaller than two thirds.
And so I know it's wrong. So, let's explore like why it's wrong and what you're doing. And then I was asking him, you know, what's a quarter like in a chocolate bar or something, and he was saying, what do you mean? And I said, you know, well, there's a one over four.
What does that mean? And he's just, what do you mean? And I was like, okay. When he sees one over four visually, he doesn't understand that to mean like 25% or a quarter as a sum. He's literally not seeing that as representing a quarter.
It's like, wow. That whole time you've been in school, when it's like two over three, he's not conceptually going ping two thirds. You know, I get a feel of how big that is.
Doesn't have that schema. He doesn't have that knowledge. He's just procedurally manipulating numbers and this weird line between them.
And that's just mind blowing. And then go, how can you miss that? But he had, I mean, loads of reasons why, but, and he really struggled with it. So, I had to go back to saying, and I talked to the teacher about it and they say, yeah, this is a trouble with this.
Sometimes that's just fractions is conceptually a lot of them find it hard. If it was decimals and percentages, they're much better for some reason because of money and things. So, getting into, honestly, all the multiples of quarters and two eighths being the same as a quarter, all of that was weak as anything.
You know, there's no point adding a quarter and two thirds, trying to work out lowest comma multiple of four and three and everything until we've really solidified our understanding of the meaning of it. And so just because he's 16 doesn't mean that's where he's at. So, you have to teach that to him because you can't just say, well, you should know this, I know, but I don't.
That is what it takes. So, in order to teach that, well, you have to deconstruct it. You have to sort of work backwards and say, what are our basic assumptions? And check them, the size of fractions, the meaning of the denominator and the numerator in what they are.
And then the classic one with fractions is something like you multiply the denominators together. So, four multiplied by three is 12. So, it's a bad example to start with because it works.
If you do multiply them, what you're actually doing is finding the lowest common multiple. So, you have to work with ones where halves and quarters, where the lowest common multiple is four, not eight, that type of thing. So, you have to choose your examples so that you confound the shortcutting and build the whole thing about variation theory and where you start.
So, all of that, I think that's really important because that informs everything else in Rosenshine's principles. It's very curriculum centric. It's all about knowing, and I would then exemplify that with other examples in a training day about where, how do you communicate world war II? How do you communicate solid liquids and gases? How do you communicate volcanoes and earthquakes and plate tectonics? You know, what do we know? Where do we start? What's next? How does it build? Unless you have a sense of that trajectory as a teacher and what the next steps are, it just becomes stuff and the students don't form schema coherently.
And if you don't know it as a teacher, you can't really deal with the questions and anticipate misconceptions. So, I think that's, for me, where I love Rosenshine because it totally knits together with curriculum analysis, although it's quite generic. It makes you think about your curriculum very strongly so that you can do all the other principles.
[00:49:34] Anna Stokke: Principle four is provide models.
So that includes modeling the steps, thinking aloud and worked examples. We could talk about that if you want, or I've talked about it a fair bit on the podcast.
[00:49:47] Tom Sherrington: I think there's a subtle difference in language here. I think I've sort of, to be honest, taken some license with this, which is because I'm very keen for all teachers to find resonance when they do training and often they do training together. So modeling is showing students what to do. It's the act of demonstrating.
Whereas provide models, this can sometimes mean sort of exemplars of a different, like in art, for example, there's a slight difference like in writing. Like you might say, here's an exemplar of a piece of writing and here's another one, which is slightly different. Can we see the comparison between them? That writing is more effective than that, isn't it? Let's explain why, but you're not modeling in the sense of producing the writing.
And I feel like you need to allow both of those to be part of the thinking. And that also then links to the scaffolding, because I think modeling and scaffolding sometimes hard to separate. So, to some extent, modeling is a form of scaffolding.
You're helping students access ideas through supported thinking, which later they have to do without. In a way, when you're thinking about your modeling or what I'm going to show and what I'm going to tell, and how am I going to demonstrate it? Which examples? You're also thinking what scaffolds might help to sort of pin this to, like little rules, little heuristics, opening phrases, or more concrete scaffolds like sentence openers or half completed problems, that type of thing. So, your modeling performance, if you like, your explanatory input is connected to the scaffolds you're then going to lead the students with afterwards.
And so, you have to kind of have to design them in tandem.
[00:51:28] Anna Stokke: Okay. Because that's the last one. Principle eight, provide scaffolds for difficult tasks.
[00:51:34] Tom Sherrington: This is interesting, actually, because another bit of Carl Hendrik bombshell reveal
the idea of scaffolding to me has always been just well established as a sort of general idea in teaching. But evidently, it was first brought to our attention by some specific people who did some experiments with like bricks, building blocks and stuff. And it turns out the study has been not replicated in recent much bigger study.
So, Carl Hendrik was saying, this is a bit of a bombshell. The origin of the concept of scaffolding was based on a very bad study. I didn't know about that study though.
So, the fact that it happened or not happened is almost like my world isn't shattered. But it is interesting about the detail of that. But I think the idea of scaffolding has transcended its origin, in my opinion.
And it's become a kind of idea that you're supporting people to piece ideas together. It's a bit like a catalyst or something, like enzyme catalyst. You have molecules, things come together with a place to sit together on, and then that thing disappears and they're joined.
It's like, that's what you're doing with scaffolding. You're providing a way for ideas to coalesce for a student, and then that thing can disappear and they're stuck. So, anything which helps students to form ideas, which they then no longer need after is a scaffold.
It's quite a broad idea. And in math, one of the most obvious ones is what Rosenshine calls worked out examples. This always makes me laugh because in England, some of those things which doesn't translate, everyone talks about worked examples, but he calls them worked out examples as a form of scaffolding.
So that you actually show students what it looks like to be successful, and then they might do partial completion and that type of thing. And also narrating your thinking. So, I like this because it links to the idea of a metacognition, which is another sort of big ticket labeled thing, which people talk about bandied around without knowing what it means.
And Rosenshine just doesn't even go there. He just says, thinking allowed by the teacher. I just think, yes, that's what we mean.
We mean, why did I do this? What step should I take next? What am I told in the question? What do I know over here? What equations can I use? Right. So, let's see what would work first, trial and error or logical system, whatever it is, then narrating their thinking process to model that. And that type of cognitive apprenticeship idea is really important that we're modeling to students how we as an expert think, so they try to emulate our thinking.
And that comes under the section of scaffolds. Scaffolds don't have to be tangible resources. They can be thinking processes, heuristics.
I saw a great lesson today where the teacher was running through it. It was like, so remember when we're facing a problem like this, first we do this, then we do that. And they all knew like, do this, check this, then do that.
I was like, duh, duh, duh. And then the kind of, whatever you call it, the recipe, the standard approach was being modeled as a scaffold. Of course, and I think that's a very powerful part of lots of subjects where we have to make the implicit explicit.
And sometimes teachers don't realize how explicit they have to make things for it to translate to students.
[00:54:51] Anna Stokke: So, I think we got through all 10 principles.
[00:54:53] Tom Sherrington: Yeah, good stuff.
[00:54:54] Anna Stokke: Rosenshine's principles, they apply to any subject area, right? Teaching any subject area.
[00:54:59] Tom Sherrington: There are a few caveats. When you explore his, you know, writings.
So, in certain places he's written that it works more for where there's a kind of well established body of connected knowledge. So, you therefore have a knowledge, declarative procedural knowledge thing, which has to break down. So, it's much less applicable to what in the sense of his direct contact with it, things like creative subjects, where you trying to model a creative process rather than any specific checkable on outcomes or such.
But I still would say, and I always challenge people. If I just use the four strands, is there a subject where you don't sequence things in steps and model how to do things? Is there a subject where you don't question a check for understanding? Is there a subject where you don't review prior learning and move on from there? Is there a subject where you don't guide the practice and then move to independent practice? And I'd say, no, like that's almost absolutely universal. The nature of it is highly variable.
And it's an important thing to really be clear about. And even in some of Rosenshine's own writings, for example, he talks about processing, processing information from knowledge within knowledge structures. And so, he gives up, for example, he talks about extensive reading, explaining material to somebody else, writing questions and answering questions, developing knowledge maps, writing daily summaries, applying ideas to new situations, giving a new example of something, comparing and trusting different examples and so on.
It's like studying for a test. There's so many different ways of exploring your schema. So, in lots of his writings, he's explicitly giving various, various different ways that you might do this. It's never five questions a day at the start of the lesson. So those sorts of routines that people are formulating and saying, oh, we follow Rosenshine. We always start a lesson with a quiz.
I just think, well, he doesn't say that. He says, review the learning. But the way you review the learning can be varied, in fact, should be varied.
And it's kind of a misreading to be very linear about that in the name of Rosenshine's principles, because he's quite good at showing, stressing variety of approaches.
[00:57:18] Anna Stokke: We started by talking about practice and 80% trying to get all students to the point where they're getting 80% correct. A lot of people listening are going to think that's not possible.
And if you have wide gaps in the classroom, it's not possible. So, what do you think? Is it possible?
[00:57:40] Tom Sherrington: I think it is, but it's not about where you start from. It's the old joke, the classic joke of stopping someone in the street or asking for directions and saying, how do I get to the museum? And they say, well, I wouldn't start from here.
You're in the wrong place. And obviously that's how that can feel. So, for example, if you have a class, which is a wide range of attainment already, and then you're asked to teach them to mastery, well, that's going to seem like impossible, isn't it? Because you think, well, they're all at different places.
And that's the truth of it. So, the only way to get 80% success rate there for everyone is to have a very stratified curriculum of practice tasks and teach topics so that everyone's got a similar kind of place on where we are. But within that, the question levels might be quite varied, and you have to tailor or provide extensive scaffolds for the students to be successful in the same English scenario for teaching English.
Like you might have some students writing like blank page, get going and others with kind of like structured guidance to like first paragraph, second paragraph, word lists, and all sorts of support to reach the same sort of place. And in maths, you might be having also partially completed examples and that kind of thing. So, there are lots of different approaches, but I do think it's harder in some contexts than others.
But I'd also say, well, what's the alternative? The alternative is to say, well, tough, some of you are going to get less than 80% success rate. And that's just the deal. I'm doing my best, but what else can I do? And I think that's basically saying, I'm not expecting you to do very well in my course because your background knowledge isn't good enough.
That's brutal, isn't it? I feel the fundamental responsibility when you're allocated your students in a real teaching scenario is to say, well, these are the guys I've got. I've got to take them from where they are. You don't have a choice.
And that's why I think in a lot of subjects, anticipating that in the planning is essential. You have to plan a sort of like strapless curriculum hierarchy of progression routes through the curriculum so that we're all in the same territory, but levels of sophistication and depth are variable because they have to be not through because of design, but because of necessity. Otherwise, some students won't practice and having go-to resources for going backwards as well as forwards.
I went to a lesson on a school visit with my daughter recently. She's not a teacher, but she came with me. We were watching a teacher explaining negative numbers and she came out of the lesson and the first thing she said to me was, oh boy, he just needed a number line.
And it's like, it's totally in her interest that this class did not have a number line. And it's like, to her as a non-math teacher, it was so obvious that how can you explain this without showing where the numbers are in relation to each other? And that type of thing, like I know I'm always going to need that. Like it's going to be there presently.
You see some math classes where it's just there above the board all the time because we're constantly referring to it. And in other words, in a lesson where other students have got that and they can get all the practice, you've got that to go to and keep visually reminding people and always having the sense of right back to the core kind of conceptual framework, the building blocks of this concept. It's never a bad idea to go right back down and reconnect with the fundamentals.
And so that benefits everyone, but it definitely benefits the students who need it. Absolutely need it. And that type of approach of, I think some teachers could do more on like, keep it concrete, keep it visual, keep it graphic, keep it connected to stuff people will know.
Otherwise, it's so abstract. You're just going to lose the lower third of your class constantly and that type of thing. So, having that in mind, these confident students all the time, what's this going to feel like for them whilst pushing everyone ahead? I mean, it's easier to say this stuff, isn't it? I'm not saying it's easy, but it is the goal.
So as long as you're striving towards it all the time and knowing it's happening, that's the other thing. How do you know what success rate is? Unless you're really checking constantly. So, you have to have this constant feedback loop to yourself of success rate and then doing your best to increase it.
That's what you can do. You have to just keep trying.
[01:01:51] Anna Stokke: I think that's a great place to end it. Great answer. And it was an honor and a pleasure to talk to you today. It was a great conversation and awesome to talk about Rosenshine's principles.
So, thank you. Thank you so much for listening. If you enjoy this podcast, please consider showing your support by leaving a five-star rating on Spotify or Apple Podcasts.
Don't forget to subscribe on your favorite podcast app or on YouTube so you never miss an episode. You can stay connected with me on Instagram, Facebook, TikTok, X, Blue Sky, or LinkedIn. All links are in the show notes.
And check out my website annastokke.com for more information. This podcast is funded by a grant from La Trobe University and from the Trottier Family Foundation through a grant to the University of Winnipeg to fund the Chalk & Talk podcast.