Listen to this post as a podcast:
This post contains Amazon Affiliate links. When you make a purchase through these links, Cult of Pedagogy gets a small percentage of the sale at no extra cost to you.
When I taught seventh grade language arts, one of my favorite things to teach was S.E. Hinton’s book The Outsiders. Every year, we began the unit with a discussion about the cliques that formed in students’ lives, how these groups interacted, the unwritten rules that governed their behavior, and what happened when groups clashed or people formed relationships across group lines. After we did some reflecting, writing, and talking, we were ready to start the book.
The reading went fine, more or less. Some chapters we did in class (I would read to them, then they would read silently), and others at home. Some students became as absorbed in the novel as I’d hoped they would; others, not so much. Predictably, some fell behind in the book like they did with all assigned reading.
I checked students’ progress with occasional quizzes, we did some work on plot and characters, setting and theme, and then, after a unit test over the whole book—containing mostly questions that asked students to identify characters, setting, and key plot points—we spent nearly three class periods watching Francis Ford Coppola’s movie version of the book. And I got to drool over Matt Dillon in the movie’s opening scene again and again and again. From start to finish, the whole unit took about three weeks.
In retrospect, I’m not sure why it was my favorite. Seriously. I mean, even though I loved the book, my students’ response to it was mostly lukewarm. Maybe it was just the idea of teaching it that I loved. Maybe it was the connections I was able to make to the stuff students dealt with on a day-to-day basis. I don’t know. I taught that book a few times, and even though I looked forward to it every time, I always finished the unit a little unsatisfied.
And it’s only now, years later, that I’m starting to understand that dissatisfaction: I can’t say with any confidence that my students actually learned something from that unit. Upon deeper reflection, I’m not confident that any of my students learned anything of lasting value from at least half the lessons I taught.
This is a hard pill to swallow, because I wasn’t half bad as a teacher. I had decent relationships with my students and I believe most of them had good experiences in my classroom, but real, durable learning? I can’t say how much of that actually happened.
I don’t love admitting that.
But at least now I understand why my teaching didn’t produce a lot of actual learning: I never set clear, measurable learning targets.
Things would have been so different if I’d known about backward design.
When teachers talk to each other about the stuff they’re teaching, they often say things like this:
“What novels do you do in 8th grade?”
“Oh that’ll be perfect. I can use this when I teach the American Revolution!”
“I don’t think I can fit that in; we’re doing moon phases next month.”
We tend to talk about our teaching plans in terms of the broad topics we cover. This shorthand is practical; we’re not going to drill down into specific skill and knowledge objectives while waiting our turn at the bagel table. But when I think about the lessons I gave my students, the ones I observed in my colleagues’ classrooms, and the work I’ve seen my own children do, I think this shorthand might be a pretty fair representation of what many of us are still doing: churning out lessons that keep students busy with our content without ever getting clear about what we want them to learn.
Instead of starting with a topic, we’d do better if we start with an end goal, and that’s where backward design comes in.
For many years, teachers have been planning lessons and units of instruction like this:
Step 1: Identify a topic or chunk of content that needs to be covered.
Step 2: Plan a sequence of lessons to teach that content.
Step 3: Create an assessment to measure the learning that should have taken place in those lessons.
Notice that in this approach, the assessment is created after the lessons are planned. Sometimes it isn’t created until most of those lessons have already taken place. The assessment is kind of an afterthought, a check to see if students were paying attention to the stuff we taught them.
For most of my teaching career, this is how I planned. It’s presumably how most of my colleagues planned. I believe it’s still how many teachers plan.
So what’s wrong with it? Well, when we plan this way, we’re more likely to include content and activities that have questionable value. When teaching the American Revolution, for example, if our goal is just to “teach about the American Revolution,” we can throw in anything that has any relation to that topic: a coloring page of the Boston Tea Party, a colonial flag craft project, or a worksheet where students unscramble words like minuteman, independence, and Hancock.
This random approach creates two problems.
The first and most important problem is a lack of durable, transferable learning. One reason so many of us don’t remember much of what we learned in school is that we learned it through this haphazard, topic-driven approach. These random activities are taking up precious time that could be spent on much more valuable stuff.
The other is poor student engagement. Our students know when they’re being asked to do something pointless. If they don’t see the relevance of what they’re learning or a direct line between the content of your course and a desirable outcome, they’ll tune it out. Sure, many students will do what you ask anyway, because they want good grades and the benefits that come from them. But they’re not learning. If you don’t believe me, ask them.
In their book Understanding by Design, which was originally published in 1998, Grant Wiggins and Jay McTighe introduced us to backward design, an approach to instructional planning that starts with the end goal, then works backward from there. The “full” version of Wiggins and McTighe’s original approach is pretty complex and can be time-consuming to implement. If you’re ready for that, I recommend digging into their book. For now, though, I’m just going to share the most basic version of backward design.
Here are the steps:
Step 1: Identify what students should know and be able to do by the end of the learning cycle.
Step 2: Create an assessment to measure that learning.
Step 3: Plan a sequence of lessons that will prepare students to successfully complete the assessment.
The difference in order is significant: Plan the assessment first, then plan only lessons that will contribute to student success on that assessment.
I was first introduced to this concept in my sixth year of teaching, and the genius of it completely blew me away. I used it when planning my next unit and experienced the biggest spike in student success I’d ever seen. On top of that, I was actually excited about teaching the lessons I had planned. For the first time, it felt like none of my class was wasted; everything actually mattered. There was something a lot more satisfying about doing things this way.
Let’s take a look at an example to illustrate the difference between a unit planned the traditional, topic-driven way, and the same unit planned with backward design.
At some point in their school careers, students study the phases of the moon. One pretty typical way to teach this is as follows:
In many classrooms, teachers also have students track the appearance of the moon over the course of a month, so that might be added as well.
Following this plan, a teacher would feel pretty satisfied that they “covered” the topic of moon phases. But if we look at the Next Generation Science Standards (NGSS), the standard relating to moon phases says that grade 6-8 students will:
“Develop and use a model of the Earth-sun-moon system to describe the cyclic patterns of lunar phases, eclipses of the sun and moon, and seasons” (MS-ESS1-1).
Note the language here: Students are meant to develop a model, then use it to describe these patterns. But in the plan above, students merely copied a model, and they didn’t use it to describe anything; even if the model required some written captioning to explain what was going on, because the model was a copy, it can’t be safely said that students were really the ones describing the system.
Then there’s the test. If we assume that a large portion of a student’s grade is based on the test, then students are not being measured on their achievement of that standard. The standard does not require students to memorize the phases of the moon. Nor does it ask them to “demonstrate knowledge” of how the whole system works. The standard wants students to develop a model and use it to describe the system.
It would be easy to blow off this distinction, to say Bah, same difference. The test asks students a lot of questions that would show an understanding of these concepts, so we’re covered.
But not really. Asking a person to develop a model is a much higher-order task than asking them to copy a model. Describing systems and patterns is way more challenging than selecting the correct description.
Developing models and explaining things is the work of real scientists: They notice phenomena, study it, then figure out how to represent those phenomena in order to make it clear to other people. To say, “Look at this! It’s interesting and it explains why things are the way they are!”
And that’s exactly what the NGSS authors had in mind: “Any education that focuses predominantly on the detailed products of scientific labor—the facts of science—without developing an understanding of how those facts were established or that ignores the many important applications of science in the world misrepresents science and marginalizes the importance of engineering.” (National Research Council, 2012, p. 43).
In other words, a superior education will teach students to think and practice like scientists. If we don’t plan learning experiences that make that possible, we’re giving them a sub-par education.
So if we re-do this unit plan with backward design, we’ll need to start by developing an assessment that would measure success with that standard. That means the assessment would not be a test where students merely label the moon phases, but a student-developed model of the moon phases along with some kind of presentation where students use that model to explain lunar phases, eclipses, and seasons.
When designing this final assessment, it’s essential that the teacher crafts a rubric that clearly outlines specific, high standards for both the model and presentation. The rubric should list criteria for the accuracy and functionality of the model, plus the quality of the presentation itself. Here’s an example of what that might look like:
With a good rubric in place, we then work backwards to determine what lessons students need to do excellent work on the final assessment.
With this “after” version, every lesson is designed to prepare students to give excellent presentations at the end. The whole time, they are using the lunar cycle vocabulary, correcting each other’s misconceptions, and just like scientists, thinking about how to explain concepts to other people.
Okay, so we’ve looked very closely at one small unit for a middle school science class. Now, take this same process and apply it to the things you teach.
Like I did, you probably also have some favorite lessons and activities. Some of these might turn out to be not just fun to teach, but also solid in terms of equipping students with knowledge and skills that will last.
If it turns out that those favorite lessons don’t really align with any standards, you might be able to revise them so they do. Or you might keep them for other reasons—not every minute of class time has to be spent on standards-based instruction. Some activities have value because they help us get to know each other better, they help students develop social-emotional skills, or they simply offer a bit of fun. But if a lesson doesn’t do any of these things, if it’s disguised as learning but is doing little more than keeping students busy, it’s time for it to go.
Using a process like backward design helps us get better at making these decisions. By making this approach part of our regular practice, we’ll be able to look back on a day, a week, or a year of teaching and say with a lot more certainty that when they were under our care, our students learned.
National Research Council (2012). A framework for K-12 science education: Practices, crosscutting concepts, and core ideas. The National Academies Press. https://doi.org/10.17226/13165.