Do We Need to Teach Comprehension Strategies Explicitly?

The answer might depend on whether the strategy involves abilities that are "biologically secondary."

---

The typical method of teaching reading comprehension puts “skills and strategy” instruction in the foreground. Often there’s a “skill of the week,” like making inferences, finding the main idea, or comparing and contrasting. The teacher briefly models the skill, and then students go off to practice it, using “leveled” texts on random topics. The theory is that if kids master a skill like finding the main idea, they’ll be able to apply it to any text.

Most literacy experts I’m familiar with don’t defend this approach. And cognitive scientists have noted that comprehension skills aren’t generic or transferable. The evidence indicates that whether you can apply a skill like finding the main idea depends more on whether you have relevant background knowledge than on how many hours you’ve spent practicing the skill.

Still, most literacy experts argue that a limited amount of explicit strategy instruction is valuable. For example, in a 2023 post titled “Don’t Discard Comprehension Strategies,” Australian literacy researcher Jennifer Buckingham argued that comprehension skills and strategies “are transferable to the extent that students need to know how to use them appropriately for different purposes.” While the ability to make an inference depends on multiple factors including background knowledge, Buckingham argued, it’s still important to explicitly teach students “the concept of inferencing and how to apply it.”

Strategy instruction shouldn’t be overdone, Buckingham added, nor should it be the focus of instruction. But it should be done explicitly, in a “somewhat decontextualized way, at least at first,” with lots of examples and practice. Eventually, a strategy that has been taught in isolation should be “applied and integrated with other strategies in studying larger units of work.”

I and others have argued that strategy instruction works best when it’s embedded in a knowledge-building curriculum that puts content in the foreground. That’s “a terrific idea,” Buckingham concluded, especially for older elementary students. But she added that the available evidence doesn’t establish that it works better than other approaches for younger students. Therefore, she advised, “It’s still OK to teach comprehension skills and strategies explicitly and quickly and build knowledge, initially side-by-side and then increasingly together.”

A Voice Against Any Strategy Instruction

On the other side—and perhaps as an outlier—is Margaret McKeown, a literacy researcher and the co-creator of an approach to comprehension called “Questioning the Author,” or QtA, which is described in detail in a book of the same name. McKeown recently tweeted that there’s no need to teach strategies at all. “Words like inference and summary can be taught/referred to,” she wrote, “but no need to TEACH them as routines to perform.” (She makes the same point at greater length on Season 2 of The Knowledge Matters Podcast, called “Know Better, Do Better: Comprehension” and hosted by David and Meredith Liben, which is currently being rolled out and is full of valuable information.)

Rather than having kids practice strategies, QtA asks them to answer open-ended questions about the text. An “initiating query” might be “What is the author talking about here?” A follow-up question might be “Does the author explain this clearly?”

In a 2009 study, of which McKeown was a co-author, the QtA approach was found to be more effective than a strategy-focused approach on most measures of comprehension, using a sample of upper elementary students, and it led to richer class discussions.

McKeown posits that strategy-focused questions divert students’ cognitive capacities in a way that can actually interfere with comprehension. When asked to make an inference, for example, students need to first recall what an inference is and think about how to make one, rather than just devoting their mental energies to making the inference.

I find myself agreeing in part and disagreeing in part with both Buckingham and McKeown. And in truth, I’m not sure there’s much difference between them, since Buckingham is arguing only for brief explicit strategy instruction—and apparently only with younger children. She even cites McKeown’s 2009 study as showing content-embedded comprehension instruction is “almost certainly the best approach for middle-to-upper primary.”

Biologically Primary and Secondary Skills

It seems to me that most of the skills and strategies schools spend so much time on don’t need to be taught explicitly—but at least one does, and it’s an important one.

The distinction I’m making isn’t between “skills” and “strategies,” which—as I’ve argued before—I find to be slippery. Instead, I would distinguish between comprehension skills and strategies that are what some cognitive scientists call “biologically primary” and others that are “biologically secondary.” (This is a categorization that was originally put forward by cognitive psychologist David Geary—although as far as I know, neither he nor anyone else has applied it to comprehension skills.)

Biologically primary skills or abilities are those that human beings have evolved to do naturally, like speaking one’s native language or walking. We don’t need to teach kids to do those things—and maybe we actually can’t. Making inferences, as I’ve argued before falls into this category (and David Geary mentions it as an example of a biologically primary skill).

That doesn’t mean kids will necessarily transfer their natural ability to make inferences to written text. Many will need to be guided to make those inferences, especially if the text is complex or leaves out a lot of information that would help readers connect the ideas. But do they need repeated instruction and practice in making inferences in the abstract? McKeown says no, although they should be taught the meaning of the word “inference.” And I’m with McKeown on this one.

Biologically secondary abilities, on the other hand, are things like reading, writing, and math—the kinds of things kids go to school to learn. Human beings have only been doing those things for about 5,000 years, and for most of that time only a small elite engaged in them. So we haven’t had time to evolve to do them naturally.

The ability to summarize a text—which is basically the same thing as finding the main idea—is something many kids don’t just pick up naturally. Multiple studies have found that summarizing can boost comprehension and learning, so it’s important that kids learn how to do it. Here I would agree with Buckingham: summarizing needs to be taught explicitly.

How to Teach Summarizing Effectively

At the same time, as I’ve argued before, I don’t think the way we’ve been trying to teach summarizing is effective. Rather than being told to just put in the “important stuff,” which is what usually happens, students need repeated modeling, practice, and feedback to learn how to determine what is important.

Initially, that instruction should be embedded in content that’s familiar—maybe a widely celebrated holiday (Halloween might be a good one), or curriculum content that’s already been covered. That ensures students aren’t trying to juggle too much new information simultaneously, overloading their cognitive capacities. Ideally, this instruction will be done using an explicit method of writing instruction like The Writing Revolution.

But once students have grasped the concept of a summary and the steps involved in constructing one, it’s time to embed summarizing activities in content they’re currently learning. I’ve given a fairly detailed example of what that might look like here, and many more examples—of summarizing and other strategies—are provided in the book called The Writing Revolution, which I co-authored.

So yes, I think we do need to explicitly teach kids how to summarize—but I don’t think we should view it primarily as a comprehension skill or strategy, or even a writing strategy. Essentially, summarizing is a learning strategy, and it should be used across the curriculum, throughout a student’s school career. Summarizing and other writing activities, including constructing sentences, reflect what students have been able to learn—serving as a comprehension check—and also deepen and reinforce their learning.

Does all of this mean that something like summarizing is a transferable skill? I would say no—at least, not in the sense that riding a bike is, or decoding words. Background knowledge is still crucial. The fact that students can create a good summary of a text on a topic they know a lot about doesn’t mean they’ll automatically be able to write one about a topic that’s unfamiliar. Still, if they’ve been explicitly taught how to create a summary, they’ll be in a much better position to tackle summarizing anything.

Of course, making inferences and summarizing are only two items on the usual list of skills and strategies that are routinely taught. I won’t try to catalogue them all, but trying to view them in terms of biologically and primary abilities could be a useful guide to how to approach them.

“Visualizing,” “predicting,” and “activating background knowledge” all strike me as abilities kids pick up naturally, like inferencing, but don’t always realize they need to apply to understand text. I don’t think these things need to be taught explicitly, in the abstract, but students will probably need to be guided repeatedly to do those things when they’re appropriate to a particular text.

It may be that summarizing, or finding the main idea, is unique among comprehension skills in that it requires highly explicit instruction. But from what I’ve seen in classrooms and heard from teachers, I’d say that many of the other skills we try to teach through comprehension instruction will develop if teachers use a content-rich curriculum and adapt a method like The Writing Revolution to its content.