Joel Rose on What K12 Schools Should Look Like in the Fall

Welcome to the future of education. I'm thrilled with my guest today on the show, Joel Rose, the founder CEO of New Classrooms, longtime friend in the education space. We've been at this world of trying to figure out how to undo the factory model of K12 schools for about a decade together, I think, Joel. So I'm thrilled to have you on welcome and I'd love you to kick off. Just let people know who are tuning in and may not know what is New Classrooms? What is Teach to One? What's the solution that you all have pioneered in the field?

Great. Thank you, Michael. And it's really a privilege to be here with you. So New Classrooms is a nonprofit organization focused on designing innovative learning models by illustration. We have a model called Teach to One Math that's implemented in dozens of schools across the country. And the way the model works is imagine you're a seventh grader and you have reading first period in room 206, you have PE second period in the gym third period, you have math. And instead of walking into room 105, you walk into what's typically a big open space with lots of different stations. In some stations, kids work with teachers. In some stations, kids work with software. In some stations, kids work collaboratively with one another. And when you walk in, you look up and you see a big TV monitor that looks like what you may see at a train station or an airport, and you see your name and you see which station you're supposed to go to. So you might spend the first 30 minutes working on linear equations with the teacher and 15 other kids who need that skill at that point in time. The next 15 minutes, next 30 minutes, working on a peer to peer activity in linear equations. And then the last 10 minutes, you take an online assessment, linear equations, and then you're off to social studies. We then take the data and create a new schedule for you for tomorrow based on how you did today.

Gotcha. So essentially, you're creating those playlists overnight. Teacher knows what each student will be doing. Or it's really a team of teachers. Right. That are working with the students at. What does the classroom look like? I'm assuming it's not a traditional classroom, if my recollection serves me correctly.

Right. Well, what you see, whether we're doing it in a traditional physical environment or the large open space, is a number of different stations. And so what happens in those stations are these different, what we call instructional modalities, whether it's virtual learning, live learning, collaborative learning. And so when you walk in what you Would see are. Are lots of students doing different activities, some in groups, some individually, based on where they are and where they ultimately need to be perfect.

So this flies in the face of traditional math instruction, say in middle school, right, where everyone's doing it in lockstep. The teacher's saying, today we're doing quadratic equations. Everyone get your books out. We're all doing this at the same pace. Doesn't matter if you didn't understand some earlier concept, you all the program you've built literally will adapt well below grade level. If I have a gap right. As I understand it, if I'm missing something that's foundational in fifth or sixth grade math and I'm in the eighth grade, your program will start to unearth that and start to pull me back in to make sure I get the concepts. And how do you all toggle that? Because that's an enormous breadth of curriculum, obviously.

That's exactly right. So there is currently this sort of standard instructional path. You learn fifth grade stuff in fifth grade and sixth grade stuff in sixth grade and seventh grade stuff in seventh grade. And that path works for about a third of kids across the country. The challenge is that some kids, for whatever reason, are falling off that path. They walk into fifth grade or sixth grade already behind. And there is not sort of an on ramp to get them back to where they ultimately need to be. So what we do is through an upfront diagnostic and running the program itself. We provide each student with a mix of pre grade, on grade, and in some cases, post grade skills. That's the fastest way to get them back onto that main instructional path.

Gotcha. Now that, like we said, that flies in the face of how most schools operate. I imagine a lot of schools are resistant to this. There's a bunch of policy barriers in place. You've written about it in a paper called the iceberg problem. Before we get. Actually, let's dive in there first. Cause I think let's lay the foundation. And then I'm very curious in this moment of COVID and sort of, as we think about next school year, where schools could innovate. But you all wrote this paper about the iceberg problem, which essentially is this incredible barrier, or it results in a barrier that inhibits schools from doing what you've suggested, which makes all the logical sense in the world. Given that math is cubic, can you sort of dive into what that iceberg problem is?

Sure. So we've been running this program for several years, and so a student would walk in, let's say, A seventh grade, seventh grader who's got a number of gaps from fourth grade, fifth grade, sixth grade. And we want to go back and address those gaps in service of enabling that student to get back on the main path to get back to grade level. But what's been happening in a number of schools is the schools say, well, you know, we like this all tailored learning path personalization thing, but the problem is the seventh grade test is coming up and we've got to make sure that this student has been exposed to all the skills that will be on that seventh grade test. And so can you please not go more than one year below the students enrolled grades so that we can make sure we maximize exposure to grade level content. And so in some schools, we really have no choice but to put those floors in and ceilings in to maximize exposure to grade level content, even though we know and we can show that that approach is actually not what's in the best interest of each student.

Gotcha. And so those policy barriers under the assessment, these grade level assessments, in essence, that just say, are you behind where you should be, are you proficient or are you advanced against this set of standards? They're not actually much more granular than that, which I think most people don't actually realize that when we see these pictures, right, of an outcome, say from a seventh grade in a particular school, we are eighth grade may be a better example. We're not getting actually a very granular snapshot of where different students are in their learning. We're just understanding, are they below the mark, are they just above the mark, or are they well above the mark? Is that basically right?

Not only that's absolutely right. And not only that, but when we even look at this term growth, most people would think growth has sort of a starting point and an ending point. How much did you grow? My kid was five foot, now he's five foot two. When we talk about education, what we're really looking at is how do kids perform on the seventh grade test and how do they perform on the eighth grade test? And the difference between those two tests is how our sector defines growth, which I don't think is how most people would think about growth, because in many ways, the skills that a kid would need to know for the seventh grade test are just very different than what they would need to know for the eighth grade test.

So this is actually a really important point because what I would love to know is on an absolute sense, given the set of skills that I need to master in math, right, what's my Progression of mastery against those. And then we can assign a number to it, like your fourth grade level, fifth grade level, whatever, if we need to to say you grew two grade levels. But really, I want to see the growth against those skills, not just like against an arbitrary snapshot that that occurs once a year. Is that a way of thinking about it?

Absolutely. And to go even further on that, by the time a child gets past basic numeracy, There are roughly 300 mathematical skills and concepts they need in order to be ready for college. If you start that process on the first day of fifth grade and you just go, you could learn just about one a week and get through all 300 before you graduate. And one a week is an incredibly slow, slow pace. We even met the kid that can't learn a skill a week if they're getting the right skill. What is happening across our system is there is this massive inefficiency because we're either teaching kids skills they don't know and they're not yet ready to learn, or we're teaching them skills they already know. And it's that inefficiency which is causing so many students to fall behind.

Okay, so let's fast forward now. Right. Like we're now, you've been doing this for many years. You wrote this report, the iceberg problem. I'll put that back up on the screen so if those interested can check it out. It's a great report. It really presents the challenge in front of us very, very well. Now fast forward. Covid hits. Schools went to remote learning. A lot of places we know were just not delivering instruction. For the last several months, gaps presumably have gotten even wider than they would have been in ordinary years. We're going to come back to some sort of schooling in the fall. What, in your opinion, should schools be doing? Like, what should this look like as we come back in the fall?

Sure. Well, I think the first thing to know is, so in the schools that we worked in, we were fairly easily able to pivot to remote instruction. And we had a number of schools that ran teach to one remotely in some cases with all of the modalities. They did live instruction through zoom. They did collaborative learning through the zoom rooms. All kids took their exit slips. So if you think about other sectors in retail and banking and transportation, those sectors that had evolved their delivery model prior to Covid had a much easier time transitioning in many cases in K12, we've been focused on optimizing the current model. And then when this hit, there was lots of scrambling to figure out what to do about it. So what needs to happen when kids come back in the fall is to first in math, there needs to be a recognition that learning loss is actually not new, that it wasn't like kids were going to come back and show up on grade level in the fall. But for Covid. So this is an opportunity to actually think about learning loss much more holistically, to shift our thinking from one that's based on coverage. That was based on coverage. I've got to cover all the seventh grade stuff because I'm the seventh grade teacher to one that is based on mastery. Let's actually look back and understand what are kids specific learning gaps and how do we put programs together that surgically are able to get kids back to proficiency as soon as possible? That may take one year, but in some cases that may take two or even three years.

That's super helpful. So that's a mindset shift up front. Tactically, what is that going to. And by the way, folks that are watching live right now, I see that there's a number of them. If you have questions or comments, pour them in and I'll try to do my best to ask Joel as well. But tactically, what is that going to look like? That's a big mindset shift. Teachers have been trained in delivering the content. So what does that look like to execute on what you just said?

So we're actually going to be putting out a report in the next three to four weeks that actually gives them sort of a guide for how to do that. But some of the highlights are it begins with having a skill map, and there are a number of them out there. We have one, there are others out there that can help teachers understand how different skills relate to one another and what the predecessors are. The second step is using diagnostic tools to understand exactly where students are relative to the skills on that map. And then the third thing is making a series of design choices of how do they want to implement the program. For example, do you want to replace your core program or do you want to implement a supplemental program? Do you want to use just teacher led? Do you want to introduce other modalities? Do you want to group kids? Viability? Do you want to heterogeneously group students? And so we provide them with a whole series of design choices and considerations in order to make those decisions more real to them.

Gotcha. So say a little bit more about the diagnostics, like how should that occur? Are there particular diagnostics that you would recommend more for math, I'm assuming you don't want to speak outside of math, but if you do, I think people would love to hear your thoughts on that. And, yeah, what does that diagnostic work look like and when should it occur?

So in math, one of the challenges with diagnostics is, as I mentioned, there's roughly 300 skills and a student's going to sit down for a test that's maybe going to have 30 or 40 items. So how do you possibly test 300 skills with 30 or 40 items? It's very tricky. Most diagnostics today are either focused on grade level, which isn't going to be especially helpful given the fact that we're talking about learning loss, or if they're adaptive, they can provide data really at the strand level, but not necessarily at the skill level. So teachers are going to have to sort of use sometimes those assessments, combined with their own assessments and other approaches and other data points, to really try to get a sense more probabilistically of what skills each students know and don't know. I will say that our organization is currently working on an approach to diagnostics that would provide information at the skill level that's still in development mode, but we're hoping to have that available for this coming fall.

So that's pretty cool. You're actually developing your own instrument. Are you using other people's items that already exist and then basically putting it in your own engine, or how does that work?

That's what we're exploring right now. We'll probably know a little bit more in the next few weeks about whether we can actually announce that that is something that will be available for free.

Gotcha. So. So one other question, which is, I've heard some people opine, you know, jumping in on day one or two or three of school and going right into diagnostics might be the wrong idea because, like, these people have, you know, students have been out of school for the last many months of the previous year, they've had a summer break. There's going to be a lot of trauma in certain students. In some cases, like, there needs to be a deep reservoir of work for first in place of building the culture and sort of getting to know students again, if you will, before you do the diagnostics. In your opinion, when is the right time to start to take those checks and start to figure out where students are and what skills they do or don't have?

Well, I think you're absolutely right. Jumping right into diagnostics, given sort of the social, emotional implications of school closures, is not the right move. In part because learning itself is social, emotional, sort of in some ways, reconnect with their peers, reconnect with their teachers, feel safe. So there are others that are probably more. Have more expertise in terms of whether it should be two weeks, three weeks, six weeks. I don't know that I have a great sense for what the answer is to that. But what I do know is diagnostics are not going to be worth their salt if they're sort of implemented at a time before students are really beginning to feel safe and connected in their environment.

Gotcha. Okay, so that's super helpful. One other question on the diagnostics that you all are developing, would you offer that as a bundle with the Teach to one model, or would it be separate so that schools maybe say, we just can't wrap our heads around implementing a whole new model or budget, because budgets are going to be a big concern. We're going to do that separate. How does that work?

Yeah, we're looking at doing that as a separate solution. So we're trying to see if there's a way where every student can have their own academic roadmap effectively. And then if they want to work with us to help sort of execute a plan for that, they could do that. But if they had other ways of sort of understanding how to help each student progress, they could do that as well.

Gotcha. Super helpful. Pivot a little bit and I'm going to push you outside. You know, you focus a ton on math. You know, in my recent article for ascd, I linked to your skill map that you all have for teachers that are trying to understand how these interlink with each other and so forth. And as you think about other areas, other subjects and so forth, there's been a lot of questions about the wisdom, shall we say, of teaching to grade level versus personalization. I think most people are more comfortable with personalizing in math and middle school. There's some exceptions, but they're much more comfortable with this notion in other subjects. How would you think about personalizing in those places and filling in gaps and diagnostics? Do you have any early takes on those other areas?

Well, Michael, I've spoken to a number of teachers in other subject areas, and one thing I've gathered through this school closure period is a number of them have said to me, obviously, this remote learning is a poor substitute for school, but there are a number of tools and a number of things that I am using that I actually do want to bring back to my. To my classroom. A number of them, for example, are making videos using Google Classroom, these other tools. And I think what then teachers come to realize is so long as they are using these tools, maybe there are other ways of offering differentiation in ways they really can't do in a traditional model. It just can't be true that live teacher led instruction for every child, every hour of the day by any teacher is the best way to do anything. There have to be ways of integrating other modalities, other approaches, other tools to enable greater levels of differentiation. That may not mean a daily exit slip, that may not mean some of the aspects of teach to one that we have, but it does mean that we can move closer to realizing the promise of differentiation and of more personalization in ways that make sense within those existing disciplines.

Gotcha. So we've talked a lot now about what you would do and how you would structure next year for schools that are returning. What do you expect to happen? Like if I put you in the role of forecaster, do you have a sense of how many schools will start to move to these approaches?

Who won't?

Is there a way to predict that in advance? If parents are tuning in, like, what can they do to push schools? Just sort of first predict, I guess, and then second, are there things that educators or parents inside of a district perhaps can do to tilt the scales a little bit?

Yeah, look, I think the inertia to get back to the way things were is going to be strong. People are going to want a return to normalcy. And normalcy means six seventh graders with seventh grade textbook, a seventh grade teacher doing seventh grade stuff, and that's understandable. But I do think there will be an increased number of school leaders, of district leaders, of teachers who will say, let's actually take this moment to revisit those core assumptions. And that'll be for a number of reasons. One, learning loss is not this hidden thing that no one ever talked about. It is now out in the open and firmly acknowledged. Two, the implications of state grade level assessments and accountability is highly uncertain at best in the near term, given how much is changing. Three, we're probably going to have some, some sort of staggered schedules or these mechanisms where kids are going to be in school, out of school. So how do you sort of keep kids engaged and moving in the right direction? Given some of those complications, some level of innovation is going to be essential. It's not going to be normal level what it was. And so I think you sort of put these together and you see, I could certainly see 5%, 10%, 15% of schools going, you know what, let's actually see, this is the moment to innovate. This is the moment to transition away from this factory model classroom. And those are the ones I think we need to do everything we can to support.

That's super helpful. And actually, since you went there, I'm curious on the policy side, because as you pointed out, you know, assessments for the past year have been suspended. Like, we're just not, you know, we're not going to know how students did for the school year in many cases, unless there are benchmark or diagnostic assessments that give us some picture across the country. What would you encourage policymakers to do going forward? To put a. I think measures in place, you know, going forward, but maybe a more, you know, what should it look like? Right. If we're going to really transform the system, what should the systems of assessments in the future look like? What's a better way of doing it?

That's a great question. So I think one place I would look is Texas. So what Texas did recently is they launched an initiative called Math Innovation Zones, where they said, we still have our regular accountability system over here, but we're going to create a zone where we have a set of providers that can offer very different kinds of learning experiences for kids. We'll look for a set of volunteer schools that really want to be on the leading edge of innovation. We'll let them innovate for three years, give them a different accountability system, and then look at the results at the end of three years. And I think those sorts of innovation zones, where they really provide the space for new models, is very much part of what policymakers should be thinking about in terms of the system overall. I think philosophically, what needs to be clarified and understood is how important it is for the incentives that are embedded within these policies to align with the instructional incentives. That's truly what's best for each student. And right now, at least in the case of middle grade math, the instructional incentives, which should be to meet every kid where they are and get them to where they need to be, are misaligned with what policy is signaling, which is always focus on grade level, no matter what.

Yeah, makes a ton of sense. Helpful to know about Texas because that was off my radar. It's actually, it occurs to me, a much better use of an innovation zone, which is to change the incentives as opposed to prescribe what the innovation ought to look like, which is often the way these innovation zones historically have worked. That seems to make a lot more sense what you just described, I guess, last question for me is where Else are the bright spots that you've seen around the country that you would say to people, check out this school, check out this district that are doing good things?

I mean, I think they're all over. I certainly districts like Miami Dade seem to be a little bit ahead of the curve when it comes to distributing technology and sort of preparing teachers to teach remotely. I think states like North Dakota are actually trying to become real leaders when it comes to innovative learning models and mathematics. I think there's some interesting things coming out of Georgia where they're piloting summative assessments that are based the MAP NWA is actually making them that have both growth measures and will pull the grade level items from those measures to sort of also come up with the summative assessment. Nebraska is looking at that as well. Wow. And so I think those are the things that I would. I would really keep my. Keep my eyes on something like that.

Just curious, will they be able to. Because the limited time. Right. And the number of items that a student says they'll be making an approximation of based on the results, we think that this is what a student's able to do and what they've been able to grow. Or is there a different conception for how that test works that will actually get the coverage, if you will, to understand the gaps, but also, you know, how it lines up against grade level standards?

I think the conception is that a student will take the MAP three times a year, right. In the fall, the winter, spring, and there'll be grade level items in each of those. So you can use the aggregate information from those particular those three assessments and from that information draw a conclusion around both. What does grade level proficiency look like? And then use the full breadth of all the assessment items in order to measure growth?

Yeah, it seems to me that's. I mean, it's super smart because we obviously don't want to dumb down standards for kids in the long run. We want them shooting toward mastery of these 300 mathematical skills that you outlined after numeracy. But we should understand the growth of each child along that trajectory as a much better way to understand transparently what's actually happening for each child and incentivizing what's right for each child as well. I guess my last question for you is what else should we be thinking about? What else is on your mind that maybe we haven't tackled or things that we ought to be thinking about in this moment of extreme pain, obviously across the country, but also, I think some really interesting opportunities for innovation?

Well, I think this moment has really laid bare what happens when we don't invest in innovations. One of the reasons the factory model classroom has persisted for over a century, and Michael, you and I have talked about this, is not because anyone loves textbooks or loves the way we do things. It's because we haven't sort of had the supply of new approaches to replace it. Most textbook companies are going to keep selling textbooks. Most venture capitalists want nothing to do with the K12 sector because it's fragmented and irrational and doesn't move fairly quickly. And so we don't have. Unlike in health care. I mean, you think about what's happening now with COVID there's all these energies from organizations from both the public and private sector working on vaccines. We do this in defense, we do this in health care, we do this in energy. There's a whole bunch of money that goes now towards clean energy solutions. Like the supply of innovation is what is so essential to actually making innovation happen. The demand matters a lot, too. But even if the demand is there, we need the supply. And so what's so essential is what we need to learn from COVID especially at the federal level, is how do we invest in the kinds of innovations that can enable this transition to ultimately happen?

Yeah, a lot more R and D on the front end would go a long way toward, I suspect, spurring this sort of creativity. And it's notable how little of that occurs right now and the only source of it is in the federal government. So, Joel, super appreciate the work that you continue to do across the country, what you've unearthed, and thanks for the conversation here to shed light on how schools can use this moment, if you will, 5, 10, 15%, as you said, hopefully to move us into the forward and pass this factory model system and really personalize for each child. Super appreciate it.

Great, thank you, Michael. Thanks for having me.

Yeah, absolutely. And we'll be back next week, of course. But thanks so much to Joel Rose. And check out again new classrooms. I'll post it there. You'll want to see what they're up to. And keep an eye on those diagnostic assessments as well. They're promising. I think that could be, frankly, a game changer for the sector. So thanks so much.