Teachers Need Information On Scientific Research—But It’s Only A First Step

A recent book provides valuable information about what kinds of teaching strategies are likely to be effective, but many teachers will need more explicit guidance in how and when to implement them..

The book, whose lead authors are Bryan Goodwin and Kristin Rouleau, is called The New Classroom Instruction That Works: The Best Research-Based Strategies for Increasing Student Achievement. As the title suggests, its purpose is to inform teachers about research on education and enable them to apply it in their classrooms.

The book begins by asking, “Is teaching a profession?” Yes, the multiple authors answer—or at least, it could be. Professions like medicine and engineering, they argue, have a shared body of knowledge and vocabulary, grounded in scientific evidence, that enables individuals to work together. But teaching, they imply, has too often relied on “fads and unsubstantiated theories” rather than drawing on “cognitive science and experimental studies.” And they highlight 14 instructional strategies backed by such studies.

As the book’s title suggests by using the word “new,” this isn’t the first time such an effort has been launched. The original version of the book came out in 2001 and featured as one of its authors the prominent education guru Robert J. Marzano. According to the introduction to the new book, that first version sold “more than a million copies.” And yet, judging by the data on student achievement—which were pretty dismal even before the education disruptions caused by the pandemic—the book hasn’t had a discernible impact.

That’s not to say that trying to connect education research with teacher practice is useless. This book and others like it—along with the conferences held internationally under the aegis of an organization called researchED—are extremely worthwhile. Teacher-training programs generally don’t provide future teachers with reliable information about how learning actually works, and most instructional materials and professional development also run counter to scientific findings. Making accurate information accessible to teachers is an important first step.

The book—which I’ll just call Classroom Instruction—covers several powerful strategies backed by cognitive science, including retrieval practice (trying to recall information helps it stick in long-term memory); spaced practice (spreading studying out over a period of time works better than cramming); and interleaving (giving students a mix of different kinds of problems works better than giving them problems that all focus on the same concept).

It also provides important pointers, such as warning against using student-directed pairs or groups to delve into brand-new topics—something many teachers do. Instead, the authors advise, use group and pair work to help students process information and skills that have already been explicitly taught.

But despite the authors’ best efforts to translate evidence into actionable classroom tips, their recommendations are likely to be insufficient or even misleading for many teachers.

Strategies May Not Work Without Content-Rich Curriculum

One basic problem is that for many of these strategies to work, the curriculum needs to include meaty content in areas like history and science. In most elementary schools, however, the curriculum consists almost entirely of math and reading. And while many of the book’s recommended strategies could be helpful for math, few are likely to work for reading.

Teaching kids how to decipher written words is crucial, and most teachers lack the training and materials that would enable them to do it effectively. This book, however, is too general to be of much help to them there.

Nor will it help them teach reading comprehension, which takes up most of the time spent on reading instruction. The standard approach is to have students practice comprehension “skills,” like “finding the main idea” and “making inferences,” using books or texts that are easy enough for them to read independently. The topics of the texts are random, and teachers don’t focus on getting kids to retain substantive information or vocabulary. But as cognitive scientists have long known, comprehension depends far more on knowledge—of the topic, or of academic vocabulary in general—than on abstract skill.

If a teacher uses retrieval practice to, for example, help students retain the concepts of “sequence of events” or “author’s purpose,” that won’t be enough to boost their achievement in any meaningful way. And some of the recommended strategies in Classroom Instruction will be difficult or impossible to carry out in the typical reading “block.” For example, the authors advise educators to “teach vocabulary words in multiple contexts.” But how do you do that if kids are reading books on a random variety of topics?

Apparently oblivious to the widespread overemphasis on comprehension strategy instruction, the authors actually recommend it. I’m not challenging the evidence they cite, but in the current context it’s important to point out that—according to cognitive science—strategy instruction on its own, without any effort to build kids’ academic knowledge, is likely to be harmful. That’s especially true for students from less educated families, who are less able to pick up this crucial knowledge outside school. Given that the authors say they’re particularly concerned with students in that category, the omission is puzzling.

The authors do note studies showing that children who receive certain kinds of social studies and science instruction gain more knowledge of those topics than their peers who get standard literacy instruction. No surprise there. But they don’t mention studies—at least some of which would meet their criteria for scientific design—showing that combining literacy instruction with social studies and science content boosts kids’ reading comprehension.

Another possible area of misinterpretation stems from the book’s recommendation to focus on “higher-order questions” rather than questions involving factual recall (except when it comes to retrieval practice). But teachers have long been encouraged to go straight for those “higher-order questions”—the kind that involve synthesis or analysis—and to skip over questions that ensure students literally understand a concept. The result is that teachers often assume students have understood material that has actually gone right by them.

For example, on the APM Educate podcast a college biology instructor recounted how during a period of remote teaching, she used the polling feature on Zoom to pose questions. As a warm-up, she would include a question she thought would be easy, only to discover that over half the class would get it wrong.

Teachers Need Curriculum That Incorporates Strategies

But even when the strategies recommended in Classroom Instruction are on solid ground, the authors’ summaries of studies often don’t provide enough information to enable teachers to transfer them easily to the classroom. What teachers really need is curriculum that provides rich content for students, in a coherent and logically sequenced manner, and incorporates research-backed instructional strategies in a way that’s likely to work best.

Failing that, what is probably more helpful than a compilation of 14 strategies is an overarching theory of how learning happens that enables teachers to evaluate whether a certain activity or approach makes sense in a specific context. I would recommend cognitive load theory, a framework that one education expert has called “the single most important thing for teachers to know.” There are at least two teacher-friendly books explaining it, one by Greg Ashman and one by Oliver Lovell. I’ve also written a post applying the theory to literacy instruction.

The Limitations of Scientific Evidence

Plus, while I’m all in favor of science, it’s important to remember that studies have their limits. Researchers often study one kind of intervention in isolation—like comprehension strategy instruction. The research may be interpreted to mean that intervention is all that students need in that area. But in the real world, lots of different things need to happen at the same time to ensure students will learn.

And most studies compare the intervention being evaluated to some vaguely described “business as usual.” But educators and policymakers need to know whether one specified intervention—or program, or method—is likely to work better than another specified one.

Lastly, if you just look at experimental studies, you’re limiting yourself to what researchers have chosen to study, which often translates into what they’ve been able to get funding for. There may be other factors—like content-rich elementary curriculum or explicit, carefully sequenced writing instruction—that simply haven’t yet been studied as much. We have to bear in mind that some interventions that haven’t been researched much may actually work better than others that have—and also hope that some of those under-researched interventions will get more attention in the near future.