Coaching Letter #190
On standard work and recipe testing
Hi, how are you? I hope you’re well and that the upcoming holiday season gives you the opportunity to rest and recharge, as well as have fun and eat lots. According to Substack, I now have subscribers in 37 states and 27 countries. That was a little mind-blowing. Holy cow. Thank you for reading my work, you rock, please share it with others and please let me know when you find something interesting or useful. The Coaching Letter is only as good as its utility for others.
This Coaching Letter is about the concept of standard work, and how I and my colleagues have started to employ the idea in several of our projects. Standard work is based on the idea, taken from improvement science, that improvement is very much tied to identifying the most effective way of meeting a goal, and spreading that method across a system, thereby reducing the variation in practice—more specifically, by shifting the less effective tail of the distribution towards the upper tail, thereby moving the median. (And if you really want to be geeky, shifting from a bell curve to a J curve.) Standard work is, then, the best known way to organize and execute the work of an individual or collective. I recommend reading this chapter from Beyond Heroes on the standard work of leaders.
Many of you will recognize that this is related to work we do on identifying bright spots—again, geekily, identifying positive outliers and codifying what they do. When we want to use that idea in our projects with districts, we frequently turn to this excerpt from the book Switch, by Heath & Heath, which is the story of how the Sternins worked to combat malnutrition in Vietnam using a bright spots approach—they worked to find out what the families of the kids who are not malnourished were doing, and spread those cooking and eating practices.
We should be less dependent on superstars, and pay more attention to what they actually do: in other words, the procedures they follow. Tom Hanks, playing Jim Lovell in the movie Apollo 13, says a couple of times: “And that, gentlemen, is how we do that.” Because he is acknowledging that there is a right way to fly a spaceship, and doing the right thing the right way deserves recognition. Likewise, surgeons follow procedures—for taking out an appendix, or repairing an ACL, or transplanting a heart—and no one thinks this is odd or depriving them of autonomy, and certainly no one thinks that following a surgical procedure is easy or does not require a great deal of expertise.
There is no shortage of resources for following procedures in teaching. There are books like Teach Like a Champion, or Classroom Instruction that Works, or The Instructional Playbook, or Teaching WalkThrus, which I am certainly not opposed to, and in fact can be very useful. But there are two problems with such books. First, they have an unfortunate tendency to be atheoretical; they don’t necessarily give you the reasons why certain practices work, or give you principles to follow. And just as problematic, they are not connected at the system level: they seem to suggest that improvement happens at an individual teacher level and are not necessarily linked to a shared understanding of HQI—it’s not that they couldn’t be, it’s just that it’s not an essential feature of the books’ approaches. And maybe it’s not right to blame the books for that—it is a weakness of K-12 education as an institution that we generally go about improving instruction by trying to improve the individual teacher through coaching or evaluation rather than improving the instructional system—see Instructional Rounds for more on that, or The Strategy Playbook, or Systems for Instructional Improvement: Creating Coherence from the Classroom to the District Office.
So when I tell you that the rest of this Coaching Letter is about recipes, I want you to know that even though that seems really frivolous, I actually have really high hopes for the leverage that work on organizational recipes at all levels of the system (from standard work of leaders to very granular instructional practices) affords us. So what do I mean by a recipe? Here is the first paragraph of A recipe for a recipe, which I’ve been working on with friends from Milford and Wilton.
Some ways of doing things are more effective than others. We should, as educators, be constantly in search of the best way of accomplishing a certain goal, and a nice metaphor for this process is recipe testing. When a chef creates a recipe, they give it to other cooks—some novice, some experienced—for the cooks to try out and give feedback on how well the recipe works. And everyone understands that the process of recipe testing is to recruit everyone who is part of this community of cooks in making the recipe better. So if I give someone my boeuf bourguignon recipe to try out and they tell me that they thought there was a bit too much salt and it would be better if the mushrooms were cooked separately and added at the end, then I am grateful for the help to make the recipe better. I don’t think that the recipe tester is a poor cook for not being able to make the recipe work the way I think it should. In other words, recipe testing is about the effectiveness of the recipe, it is not about the competence of the cook. (Readers who have worked with us on continuous improvement/improvement science projects will realize that a recipe is equivalent to a change idea.)
And of course, you can extend the metaphor. Perhaps your math department is engaged in creating a cookbook: a collection of recipes from others, or that they themselves have written or adapted, that they are testing and refining in order to find the best approaches to teaching their content in their context. Perhaps there is a menu for teaching a particular social studies unit. Perhaps there are yet-untapped Martha Stewart jokes.
Another argument for the recipe/cookbook approach. As we move into more student-centered instructional approaches—epitomized by Building Thinking Classrooms but not monopolized by Building Thinking Classrooms, where the content is on grade-level (actually, and not just in theory), the thinking is shared and concurrent, the thinking is being made visible, and the classroom environment supports challenge and risk-taking—very few principals and supervisors will have taught using these approaches. (Just to be clear, I think I was a really brilliant teacher and my kids learned a ton, and I used a lot of direct instruction approaches combined with some truly excellent tasks that I wrote myself and a lot of formative assessment—and I would teach differently now.) Indeed, the number of evaluators who have taught using the methods described in Building Thinking Classrooms themselves is so close to zero that it is insignificant. So what qualifies these people to give feedback to teachers using BTC? They have no personal experience of how layered, nuanced, and complicated the challenges are. They don’t have any practical insight to share. So there is little benefit. And the costs are potentially huge. Giving “helpful” task-level feedback to teachers who are trying something out risks making them feel judged for taking a risk, and that is the exact opposite impact from the one you want.
So, here’s the revolutionary approach that I’ve been thinking about and advocating for years but have sharpened considerably lately. Let’s get rid of the model of feedback that says that coaches and leaders should watch teachers teach and give them feedback and/or ask them questions about their practice. Specifically:
Let’s reverse the flow of feedback. Instead of evaluators attempting to give teachers task-level feedback, let’s have them find out from teachers what the stumbling blocks are when they try to implement a recipe, so that they can help with those. Which may or may not involve actually visiting classrooms.
Let’s be really intentional about coaching. Instead of asking coaches to figure out for themselves a) what to tell teachers about coaching itself, b) what to tell teachers about what good instruction looks like, c) how to coach, d) what kind of feedback to give, let’s create a coaching model that standardizes all of that and that teachers understand and find supportive. And let’s make the creation and testing of recipes—for teaching and coaching—a central part of that.
Let’s stop talking about autonomy in teaching like it’s a good thing. Horsefeathers. Teachers already make, according to Deborah Ball, around 1200-1500 instructional decisions a day. That is a huge cognitive burden and I don’t know why we would seek to increase that by increasing autonomy. I think we should be reducing it through co-developing recipes—which are NOT the same as checklists or scripts—which will:
help teachers teach more effectively;
reduce variation in instructional quality across classrooms;
reduce the cognitive load on teachers so that the decisions they do have to make are better and more equitable; and
save teachers some time by providing them with guidance on how to accomplish a certain goal efficiently and effectively.
Let’s capitalize on the positive outliers in any given part of an organization by turning what they do into recipes for others to test.
Let’s work on reducing variation by asking everyone implicated in the work of the recipe to test the recipe, as a collective rather than as individuals. (See this Dylan Wiliam article on teams, which is ostensibly about formative assessment but could be about any kind of recipe testing.)
Let’s spread and strengthen a shared understanding of High Quality Instruction across all levels of the organization—see Coaching Letter #186 for an argument for why that’s necessary—through the sharing and testing of recipes.
Let’s work on getting the grain size of implementation right. As is pointed out in Learning to Improve, we should be talking about integrity of implementation not fidelity of implementation. To lean into the recipe metaphor, my oven may heat unevenly compared to yours, and your kitchen may be more humid than mine, and so the recipe has to be tweaked to account for those circumstances. It also has to account for varying skill levels of the cook, and not assume that everyone knows how to chiffonade, or make a bearnaise sauce.
Recipes should change the work of teams at two levels: what they do and how they do it. From one of my favorite educational leaders and CL reader, Michael Clow, in response to CL #186:
We are riding the river of DATA, SEL, and PLC...what we continue to miss is that it is about teacher practice (the moves the teacher chooses) and that concept is not even in the river with us. If we ‘do PLCs’ without the introduction of new practices, we risk reinforcing whimsical or even bad practice (we are limited by what the team already knows). We ‘do data’ without the introduction of new practices and we reinforce frustration, helplessness, and defensiveness (again we are limited by what the team already knows).
(More on teams soon. Facilitating teams is the focus of the afternoon session of the Facilitation Workshop on January 23, 2024.)
We have started posting recipes on the Recipes page on the Partners’ Building Thinking Classrooms website. We would really love it if you would try them out and give us feedback. Remember the words of Mike Rother in Toyota Kata:
Now, when arms fold up and people say, “Let’s see if this will work,” I say, “I can save you the time. We already know it probably won’t work. Despite our best efforts to plan this, we know that within a short time there will be ‘charred and glowing pieces’ lying around. We just don’t know in advance when, where, or why it will fail.”
In Milford, some of the coaches have been testing recipes with teachers, which is really cool and super useful. I asked for some specifics about what they learned, and here’s some of what they came up with, which is also really cool and super useful:
Many of the coaches asked students for feedback on their experience with the recipe. Teachers found this very powerful
One coach explained that she offered to go into a kindergarten teacher’s room and model the recipe and the teacher responded by saying she did not need the coach to model because she was pretty sure she could follow the recipe on her own. She welcomed the coach to come in and check it out.
Another coach explained that a teacher was pleasantly surprised because a student who would never speak during a traditional turn and talk, did talk using the structure of the recipe.
A coach explained that one teacher commented on how she learned that she was making some faulty assumptions about a particular student she deemed a “high flyer.” She explained that when she runs a normal turn and talk she never really listened in on the partnership this student was in because she assumed he understood it. With the think-pair-share recipel she noticed that when it was his turn to share he did not have much to say because he wasn't sure what to do.
Coaches rock. (And more on recipe testing soon, in part because we are going to make that part of the afternoon session of the Facilitation Workshop on January 23, 2024.)
Huge thanks to Tom and Andrew for their help with this Coaching Letter—could not have written it without their help.
Again, have a wonderful break, and let me know if you need anything from me. Best, Isobel