What I've Learned from Direct Instruction

A few ideas I've incorporated into my teaching

A little over a year ago I read the book Direct Instruction: A Practitioner's Guide by Kurt Engelmann. It's a good book. I recommend reading it. I wrote a review, which I think is worth checking out, and I ended up going down a bit of a rabbit hole. I want to write a bit of an update about how I'm thinking about Direct Instruction and why I think it's worth learning about.

Kurt Engelmann is the son of Siegfried (Zig) Engelmann, the lead author of most Direct Instruction (DI) programs. Direct Instruction is a family of scripted curricula first developed in the 60s and 70s. The programs are characterized by explicit instruction, scripts for teachers, breaking skills down into small steps, frequent checks for understanding, and lots of practice.

Kurt is the current president of the National Institute for Direct Instruction, an organization that supports schools adopting DI. Kurt reached out to me by email after I wrote my post a year ago. He's a really nice guy. We exchanged a few emails, discussing the ins and outs of DI. He offered to send me two DI teacher's guides for free so I could get a closer look at the programs, and recommended the longer and more technical book Theory of Instruction by Zig Engelmann and Douglas Carnine. I mention this to emphasize that Kurt is a well-meaning educator who cares deeply about schools and what's best for children. I'll get to some of my critiques of DI in a moment, but I want to emphasize this because I sometimes hear rhetoric that Direct Instruction is driven by people who hate children and are part of a sinister plot to destroy the teaching profession or something. I don't think that's true. You can disagree with the DI approach, but I think turning curriculum choices into some proxy battle between the forces of good and evil doesn’t help anyone.

Why I’m Still Not Interested in Teaching Direct Instruction Myself

I’ve learned a lot from Direct Instruction but I still have no desire to become a Direct Instruction teacher. DI doesn't provide much variety in how students participate — it's mostly choral response, cold calling, and written work. DI doesn't do much to show students the beauty and richness of math. When I write about fuzzy math, or how math is awesome, or explorations, those things don't have much of a place in DI. I also don't think DI does a great job of building off of what students know, and helping students see that their ideas have value and are worth learning from. If you want to hear more about those, you can read my review from last year.

Also, Direct Instruction isn't just a lesson plan or a set of workbooks. It's a whole-school system. The programs are designed to assess students at the beginning of the year, place them in homogeneous groups based on their results, and then reassess and regroup multiple times per year. These groups aren't restricted to specific grade levels, so students at the same level, even if they’re in different grades, would be placed together in a single class for each subject. DI isn't something I can adopt on my own; it needs to be a whole-school project. That piece is often ignored when I hear people talk about DI.

But I’m writing this post because there are a bunch of things I’ve learned from Direct Instruction. Here are three big ones:

Choral Response

The most common way that students participate in a Direct Instruction lesson is through choral response. I don't do this nearly as often as a DI lesson, but I have incorporated it into my everyday teaching and I'm a huge fan. Here's how it works:

Beginning the first day of school, I teach students the basic routine. Let's say I'm asking, "what is 50% of 12?" I say, "think in your head, what is 50% of 12?" I pause and let students think. Then I raise my hand above my head, say, "Ready, go", as I say go I bring my hand down, and students say the answer. This gives two separate, simultaneous signals for when to respond and helps keep the responses crisp. It takes a bit of practice, but students get good at it and it becomes a seamless part of everyday teaching.

You might be thinking, "ok but questions like 50% of 12 are easy, that seems like a lot of effort for a really small thing." I disagree. Here are a bunch of benefits:

• Lots of chances for students to respond keep them engaged. The longer I go without asking students to do something, the harder it is to tell if they’re with me.

• Choral response is a decent check for understanding. It’s not perfect, but I can tell whether most of the students answered correctly. I can’t hear every student’s answer, but I get a much better sense than asking a single student.

• It takes very little prep. I can write the problems on the whiteboard, or on a piece of paper under a document camera, or prep slides if the problems are more complex.

• A quick round of choral response is a nice contrast with other modes of doing math (paper, whiteboards, etc) to create some variety in math class.

• Choral response doesn’t work well for more complex problems, but math is full of little pieces of problems that lend themselves well to choral response. I might do a round of choral response asking students whether a question’s answer is positive or negative, or practicing whether tip, tax, discount, etc are percent increase or decrease.

• When I first started using choral response it often felt unnatural or forced, but as I got better I found more and more places to use it and found more ways to adapt questions so they had a clear and concise answer suited to choral response.

Now I use choral response in most lessons. It’s quick — often 2-4 minutes of total time in a lesson. But those quick chunks of practice and checks for understanding are often a kind of glue that help the other pieces of my lesson fit together, or help me get a sense of where students are and whether they’re ready to move on.

Short Chunks of Practice

A Direct Instruction lesson does not have a single objective. Instead, in each individual lesson there are a bunch of different chunks focusing on different objectives. Rather than teaching a lesson on the topic, spending a day or two practicing, and moving on, that practice is spread out over multiple lessons, often multiple weeks. There is still a lesson where students are first introduced to a topic, but then students continue practicing in short, 5-10 minute chunks and gradually solve harder and harder problems over time.

Here’s an example. I teach integer addition and subtraction. It’s a hard skill. Some kids pick it up pretty quickly, some students struggle and need a bunch of practice to get it down. A typical approach would introduce the topic, do some focused practice for two weeks or so, and move on. The Direct Instruction approach would spread this skill out, probably over multiple months. Each day builds on the last, getting gradually harder.

A lot of teachers have found that the common sequence of teaching a hard skill, spending a full class period practicing it, and moving on doesn’t work very well. I agree. Some kids get it, and some kids don’t. Spreading practice out like this builds in a ton of time to check for understanding, help kids who are feeling stuck or confused about something, and build confidence over time. The other benefit is for the kids who “just get it.” Some of those kids who get it in a conventional teaching sequence still don’t retain what they’ve learned because they don’t have enough distributed practice. This approach spaces practice out over time, improving retention. The second benefit is that extended practice can feel really boring for kids who figure something out quickly. Breaking practice into quick chunks each day and then moving on to something new helps to create a sense of pace and purpose.

I really feel like this move away from a single objective each class has transformed my teaching. I don’t need as much time to introduce new topics because the practice is spaced out across future lessons, which frees up time for more small chunks of practice. I can also prioritize and put extra practice time toward key skills that students will use often in future math classes or in the world outside of school. If all my students are confused I can punt and come back the next day and I haven’t wasted a full class, just that 5-10 minute chunk.

Practice is so important in math class, but the typical way practice is structured with a focus on a single skill, then moving on and rarely returning, doesn’t make any sense. It would be like a basketball team that spent a week of practice just working on layups, then a week of practice just working on defense, then a week of practice working on free throws. No basketball team does that. Instead, each practice works on a few different elements. No coach would introduce five new things in a single practice — there’s still probably just one new thing introduced at a time. But you don’t expect players to master that new thing right away, and you keep practicing all the old stuff mixed in with the new. This approach uses the same logic. To put a number on it, a typical Direct Instruction lesson spends about 80-90% of the time practicing and reviewing ideas that have been introduced previously. I don’t think I’m there, but it’s an interesting goal to shoot for.

Preparing for a New Topic

This is similar to my last point, but my last point was about how practice is structured after introducing a new topic. This is about what happens before introducing a new topic.

First, it’s helpful to look at a visual of what a Direct Instruction program looks like:

Here the horizontal axis is the lesson number, showing 120 lessons in total. Each row is a topic, and the bars show the lessons where a given topic is included.

I don’t structure my teaching like this. I still have units, where I’m in a percents unit, or in an equations unit. In the last section I described how I stretch out the practice beyond what you would typically see in a math curriculum after teaching a topic. But on the other side I also spend a bunch of time reviewing and practicing prerequisite skills before I get to a unit.

The best example is equations. I teach two-step equations, and students learn one-step equations in the previous grade. But if I don’t say a word about equations before we get to that unit I’ll discover that some students are pretty rusty with one-step equations. Then I have to squeeze in a bunch of quick reteaching, it will feel rushed, and two-step equations will be a mess.

Instead, I space out that teaching over time. I introduce one-step equations gradually, including the tape diagrams we use as our primary representation. We practice in lots of small chunks for about two months before that unit begins. It doesn’t take tons of time, but it gives me the flexibility to adjust as necessary, help students feel confident with one-step equations, and set them up for success with two-step equations.

I do this same thing with every unit. We practice evaluating percents before our unit on percent increase/decrease. We practice unit rates before our unit on proportional reasoning. We practice working with angles before our unit on geometry. I could list plenty more. The more I do this, the more little skills I find to teach before we start a unit, and the smoother that unit goes.

Here’s what a conventional scope and sequence looks like:

Here’s what I’m doing now:

To be fair it’s not quite that simple. Some units require more prep or practice, some less. There’s plenty of practice in the unit itself. But this helps to get across the rough idea. Teaching completely without units, like the Direct Instruction example above, feels overwhelming for me. My approach captures some of the benefits while allowing me to use the basic scope and sequence I’m used to.

Finally, this ties back to my original point about choral response. Teaching in this way means I have lots of small chunks of class working on a specific skill. Choral response is a great fit for many of those. I don’t use it exclusively — I also use a lot of mini whiteboards, paper and pencil, and DeltaMath. But choral response is a great, low-prep tool to get some quick practice, especially for some of the prerequisite skills I want to work on with students before a unit begins.

Closing

There’s a lot I do as a teacher that’s very different from Direct Instruction. If you wander into my classroom in a month when we’re back in school you might see a bit of choral response, or a quick round of practice of a specific skill in the style I described above. You might also see us doing a Which One Doesn’t Belong, or mixing kool-aid to learn about proportions, or playing a math game.

Direct Instruction is laser-focused on helping students become proficient with the core skills of the math curriculum. There’s a lot of value in that, and Direct Instruction does it well. Teaching those skills is one big part of my job. But to me, there’s more to math class than making sure students can solve two-step equations. All that other stuff is why I’m not interested in becoming a pure Direct Instruction teacher. Using some DI tools has helped my students feel more confident and successful with their math skills. It’s not the only thing I do, and that’s great. My opinion is that far too many teachers don’t know anything about Direct Instruction, and they would have a lot to learn if they were open to it.