I’ve recently been reading a new edition of a major AP stats book. Let me stipulate that the book is good: it covers the material thoughtfully. It’s well-written. It has good problems. And when I read it, I get seduced: I start to think that my course ought to be just like it. For example, I start to think that when I introduce distributions of continuous data, students ought to attend to shape, center, spread, and outliers, and that they can remember this by thinking, “don’t forget your SOCS.”
But the truth is, I don’t even want to “introduce distributions.” I want to introduce situations and then notice when distributions show up. But as I read the book, my resolve starts to soften. Should I be so relentlessly new-age-constructivist-progressive-touchy-feely in my attitude towards curriculum, or should I just get real and teach distributions? This spinelessness gets worse as we get into parts of the course I haven’t thought of as carefully: gee, if I’m not sure how to teach this, maybe I should go with the book. And more insidiously, I’m not sure whether to teach this; but it’s in the book, so maybe it’s important.
Textbooks are massive and authoritative. They are literally massive, contributing to student back pain; and they create considerable (figurative) instructional inertia. They are authoritative in the best sense in that the authors have worked really hard to get it right. Mistakes do creep in, but these books are often very carefully reviewed for technical details, coverage, and overall direction. But they also bring with them the authority that comes from looking impressive and being costly. If we spend that much (the reasoning goes) and ask our students to haul them around, we ought to use them.
Another problem with massive and authoritative is that you get the impression that they’re comprehensive. They do cover the content, often so thoroughly that there’s no way you can get though a book—so you have to cut, and then fill in the connective tissue that holds the concepts together. But the “coverage dragon” is hard to slay. Authors want all of their ideas in the book; and we’re so independent in this country, we don’t want some publisher deciding what’s not going to be in our text.
A textbook may excel at content coverage, but it generally doesn’t do too well at the big overarching ideas, the really important points of the course (which are the ones I’m having the most trouble deciding how to assess). For example, how can a textbook help kids put on “data goggles” and look for data in their everyday lives? How can it encourage them to use multiple representations? No book can actually teach, but it can tell you, through its organization, what is important. And you know that if there were a chapter on “data goggles,” we would skip over it quickly on the way to the real math.
You could spread such things throughout the book. For example, this text wants students to use a four-step plan for statistical investigations. They describe it. A four-step icon appears in the margin on many pages. But in the sea of visual widgetry, a kid still won’t know how important it is. And of course the book can’t reliably assess it in chapter tests. In this case, the demands of the AP exam will help, but it’s really up to the teacher, not the book, to give importance to something as “non-content” as a four-step plan.
Looking for the right tool (and why we don’t have it)
What should the role of the textbook be, and what should it look like so that it can fulfill that role? My favorite answer this morning is that a math textbook should be a resource: a place to go for explanations you might have forgotten, and a repository for good problems. I’m happy to let go of linear exposition, because I suspect that no one really learns from reading that. And I’m really happy to let go of four-color, marginally-relevant margin illustrations. A black-and-white paperback the size of The Girl with the Dragon Tattoo—or any of the old SMSG texts—ought to do it. An online resource could have hyperlinks and dynamic, interactive content, increasingly a plus.
It would be so much easier to have books like that. So why do big textbooks keep selling? Of course, as a namby-pamby progressive I reflexively blame big corporations for as much as possible, but it’s also true that there is a value added: money brings resources to bear. The work and organization are better than what you will find in a typical teacher’s filing cabinet full of handouts and lesson plans. (For the atypical teacher, those handouts may be too idiosyncratic for most of us mortals to use.) Having said that, I do blame capitalism for (a) the demise of many small publishers, which contributed to the collapse of the market for supplemental materials; and (b) the huge expenditures for marketing textbooks, which mean the have to be expensive. Publishers are not going to pay an army of reps to go to your school to sell you inexpensive books.
But other barriers are cultural. We don’t give most teachers enough time to plan; most of us can’t design a course in our preps, and have to rely on a textbook for planning survival. Courses therefore emphasize what the textbooks do. And at a deep level, a big book is a symbol of learning and a well-equipped school. An edgy private school could conceivably opt for a slim paperback or an online, open-source math textbook, but LA Unified? No way; not in 2010, at least.
For me, for this year, I’m just hoping to keep focus on what I think is important, as elusive as that may be. This is the best-case scenario, after all: I have time to plan, to create a new course. I know the content. I adore the kids. I may need outside work to put bread on the table—Prinz Esterhazy hasn’t been looking to be a patron for a while, dang!—but that’s a small price to pay.
Meanwhile, I did have to decide on a textbook back in May. I have 18 students on the roster and we had something like 15 usable books left over from last year. So it only cost the department 5 books to have something and a couple extra. It’s good, for a book, and it is a paperback; I’m just not sure how I will use it.
A Best-Case Scenario 