Scientific Findings and Classroom Practice

How much difference will it make if federal support for education research disappears?

---

Recent drastic cuts to the US Department of Education are worrying for a number of reasons. While federal funding makes up only about 10 percent of what K-12 schools spend, it’s almost entirely directed at students who are the most vulnerable—those from low-income families, for example, and those diagnosed with learning difficulties. The government did just release nearly $7 billion in funding it had previously frozen, but the future remains uncertain.

The DOE’s education research arm has also been the target of drastic cuts, and many consultants have had their contracts terminated. “It just feels like we’re going back into the dark ages,” one told The 74.

How concerned should we be? Some of that research is undoubtedly worthwhile, including the data that enables officials to choose representative samples of students for the reading and math tests the department is required to give every two years. But some may be useless—or even misleading. The fact that one kind of intervention has been studied more than another might not mean it’s better; it might just mean it’s easier to study. Research has been good at tracking students’ lack of progress. It hasn’t been so good at enabling us to do something about that.

Even when research is illuminating, it often has little impact on what happens in classrooms. For decades, the DOE’s Institute of Education Sciences has operated the What Works Clearinghouse, or WWC, intended to help connect educators to education research. It publishes “Practice Guides” that synthesize research and distill it into recommendations. That sounds useful, but a lot of the research doesn’t show significant benefits. In addition, some conclusions in the Practice Guides have been criticized as misleading, erroneous, or politically motivated.

In any event, the guides can be dense, and a lot of the recommendations are too vague to be useful for practicing teachers. A guide to teaching secondary students to write, for example, recommends that students be advised to “use different kinds of sentences” but doesn’t provide much information on how to teach them to do that. Commentators and academics may cite the guides as a gold standard, but it’s far from clear teachers read them.

Some have predicted that reduced support for the WWC will “leave district leaders unequipped to navigate the billion-dollar world of school-based products and services.” But even with the WWC, district leaders haven’t been well equipped for that challenge.

Studies usually look at specific “interventions” in isolation set up against some unspecified approach labeled “business as usual.” It would be far more useful to compare the specific curricula or programs that district leaders are trying to evaluate. Instead, that task has been left to organizations like EdReports, whose evaluations are uneven and don’t even consider evidence of how well the curricula work.

I understand the theory behind the typical approach to research. Programs combine a bunch of different practices, making it hard to determine which ones are actually causing whatever effects are observed. But in real life, teachers don’t just do one thing at a time. And the people who are choosing curricula need to know how different options stack up against each other.

Because it can take years to see the results of a curriculum, these comparative studies should ideally last for three years or more. That, however, would be enormously expensive, and the federal government is unlikely to be pouring money into such studies anytime soon.

Promoting Evidence from Cognitive Science

Rather than funding new research, a slimmed-down federal education research agency could devote more time just to publicizing the significant body of existing research in cognitive science that few teachers ever hear about.

That would include findings like the importance of building knowledge for reading comprehension and learning in general; the value of having students retrieve information stored in long-term memory that has been slightly forgotten; the usefulness of deliberate practice in mastering complex skills; and the value of explicit, interactive instruction over inquiry or discovery approaches when students are new to a topic. Most teachers not only never learn about these principles, they’re often inculcated with beliefs that contradict them.

But the WWC has issued only one Practice Guide that focused on these findings, back in 2007. And it limited the relevance of its recommendations to “subjects that demand a great deal of content learning, including social studies, science, and mathematics”—as though the findings have nothing to do with reading comprehension, when in fact, they do. In any event, like the other Practice Guides, this one has had little or no discernible effect on teacher practice or on the content of teacher-prep programs.

Knowledge of cognitive science has spread more widely in some other countries, in part because of government support—specifically in England, Australia, and New Zealand, all of which share with the US a history of education orthodoxy that conflicts with many scientific findings. Some of the support has come from independent but government-funded organizations like AERO in Australia and the Education Endowment Fund in England.

A recent story by Greg Toppo in The 74 described the enthusiasm for cognitive science among educators in England, noting that some American authors and speakers who focus on the topic are in far more demand there than in the US. (That hasn’t been my experience, although I’m not sure why. I have a busy speaking schedule in the US, and I’m about to head to New Zealand to speak at a conference—followed, a couple months later, by a packed three-week speaking tour of Australia. But I have yet to be asked to speak in England.)

The Risks of Government Support

In the US, where local control of education is deeply ingrained, it could be tricky for the federal government to embark on the kind of support for cognitive science that governments in other countries have provided. The federal government can only influence education indirectly, by dangling or withholding funds, and when they’ve done that in the past it’s sometimes resulted in unintended consequences. The federal No Child Left Behind legislation, for example, along with the federally supported Common Core standards, led to high-stakes reading tests that exacerbated an emphasis on supposed reading comprehension skills as opposed to building content knowledge—a practice that actually does leave many children behind.

In any event, I’m not sure it’s a good idea for any presidential administration, and especially a conservative one, to throw its weight behind cognitive science. In an ideal world, endorsing a body of solid evidence relating to education wouldn’t be seen as political, but we don’t live in that world. Education issues have almost always been politicized, and that’s particularly true in the current polarized environment. To some educators, just advocating for knowledge-building marks you as a Republican, never mind your politics.

If Education Secretary Linda McMahon were to come out tomorrow in favor of, say, retrieval practice, I imagine that many American teachers—a group that skews to the left of the general public—would only see it as confirmation of their suspicions that teaching approaches designed to enable students to retain information are inherently right-wing. Even systematic instruction in phonics has long been identified with the right.

Politicization has already happened in other countries. There’s been politically tinged pushback to government efforts to promote science-informed teaching in Australia and in England, where those initiatives were enthusiastically undertaken by the Conservative government. Now that Labour is back in power, it remains to be seen how long the changes will last.

How to Get the Message to Teachers

If the federal government shouldn’t get involved, what’s the best way to acquaint millions of American teachers with instructional practices that have solid evidence behind them?

One possibility is for state governments to take the lead. I’m not aware of any that have promoted cognitive science per se, but by now most have championed the kind of phonics instruction that’s supported by evidence from that field. State officials would just need to expand their focus from early reading instruction to learning in general.

But that’s a big “just.” Many teachers are already feeling overwhelmed by the “science of reading,” trying to absorb a bunch of complex concepts that contradict their training and then figure out how to implement them in the classroom. Telling them they now need to do the same thing with the “science of learning” may not go over well. Besides, as Jim Hewitt and Nichi Sachdeva have observed on their Substack, The Science of Learning, education “functions more as a belief-based profession than a science-based one.”

What could work is this:

First, accessible, engaging introductions to the basics of cognitive science from a reliable source. Teachers don’t need a graduate-level course, but they do need to have some sense of why what they’re doing conflicts with the evidence. That’s what Emily Hanford’s Sold a Story podcast provided for phonics instruction. It would be great to have on-ramps like that for cognitive science as well. And now for a shameless plug: The recently released podcast series that I co-hosted with Dylan Wiliam and Doug Lemov—season three of the Knowledge Matters Podcast, “Literacy and the Science of Learning”—might be a good place to start.

Second, curriculum materials that are grounded in scientific evidence, so teachers don’t have to figure out how to translate evidence into classroom practice themselves. Instead, they can devote their limited time and energy to figuring out how best to deliver an evidence-based curriculum to their students.

That, of course, would require curricula to be developed by people who have a pretty deep understanding of cognitive science. I’m not a curriculum expert, but I know of a few curricula out there that seem to align well with the evidence, including Doug Lemov’s middle-grades ELA curriculum, Reading Reconsidered. On the social studies side, the guys behind the Four Question Method are working on an excellent and science-informed eighth grade U.S. history curriculum.

At the very least, a curriculum should have rich content rather than a focus on abstract comprehension or critical thinking skills. All 10 of the ELA curricula listed on the website of the Knowledge Matters Campaign meet that criterion. (I serve on the board of the nonprofit that is the Campaign’s parent organization.) Instructional principles grounded in cognitive science implicitly assume that you’re teaching either a transferable skill (in which case you can use deliberate practice) or substantive content like history or science (in which case you can use approaches like retrieval practice and elaboration). They don’t work as applied to essentially non-transferable skills like “making inferences.”

Last, I suspect that a focus on writing instruction—embedded in a content-rich curriculum—is the most effective way to enable teachers to align their pedagogy with cognitive science, across subject areas. Teachers in the US may not be clamoring to learn cognitive science, but they are desperate for a way to help their students become better writers.

For my book Beyond the Science of Reading and for the last episode of the recent podcast series I co-hosted, I focused on a high-poverty district in Louisiana called Monroe City. Like most teachers, those in Monroe weren’t familiar with concepts like deliberate practice, retrieval practice, and elaboration. They just wanted to enable their students to express themselves coherently in writing.

But they discovered that when taught in a manageable way and embedded in curriculum content, writing can provide all the benefits of science-informed instruction while also familiarizing students with the complex syntax of written language, boosting their reading comprehension. That’s likely why, as one teacher told me, “We realized that teaching students to write clearly was actually teaching them to think clearly.”

Seeing that happen has led some teachers in Monroe to investigate cognitive science for an explanation. But they didn’t need to know about the science in order for their students to benefit from it. (The freely available ELA curriculum used in Monroe, Louisiana Guidebooks, pairs rich content with writing activities created by some of the Monroe teachers, based on The Writing Revolution method.)

If we want a fairer and more effective education system in the US, we might not want to lead with calls for the “science of learning”—especially if they come from the federal government. It’s likely to work better for some educators to say to others: Hey, need help teaching your kids to write? I found something that works. And then see where that leads.