Monday, August 25, 2014

To Teach Comprehension Strategies or Not to Teach Them

I don’t hear anything about comprehension strategies anymore. Was that idea just another fad or are should we still teach those?

Your question raises an interesting point about American reading instruction. We tend to chase fads. Instead of building on past reforms and improvements we instead ride the pendulum back and forth.

Back in the 1970s and 1980s, there was lots of interest in teaching students how to think effectively about the ideas in texts. There was lots of research on how to engage prior knowledge, summarize information, ask questions, monitor understanding, and so on—and lots of interest in bringing these strategies into classrooms.

Strategies engage readers in thinking intentionally—rather than just reading a text and hoping something sticks, the reader enters the enterprise aware the text is like a mountain to be scaled or a problem to be solved. In such situations, you take actions that help you to reach the goal.

Thus, readers may preview texts ahead of time to increase anticipation and to ensure that relevant prior knowledge will be at the ready. Readers may set purposes too—like turning headings into questions to be answered. As they read, they may stop occasionally and sum up the information provided to that point—rereading if there are apparent gaps.

In the strategy world, readers need to be “meta-cognitively” aware. That means, for instance, that they should notice when they are not understanding something and to do something about it (such as rereading the pages that you you phased out on, looking up a word in the dictionary, or asking someone for help).

The whole language movement has been pilloried for nudging phonics out of the primary classroom, but—something not often noted—it booted comprehension strategy teaching, too. Strategy teaching tends to be direct instruction—the teacher explains what the strategy is, how to use it, and why it’s important. Then the teacher may demonstrate the use of a strategy and engage kids in a heavily scaffolded version in which the teacher does much of the work (“This would be a good place to ask a question about what we have read. If you ask and answer questions you’ll remember more of the information later.”). Over time, the teacher would fade the support with kids doing it more and more on their own.

Strategies came back a bit during the 2000s, probably as a result of the National Reading Panel’s review of more than 200 studies showing that we could effectively teach students to comprehend better by teaching such strategies.

As your question reveals, now strategies are on the retreat, yet again. The reason this time is almost surely due to the fact that the Common Core State Standards don’t include any comprehension strategies. They don’t prohibit the teaching of comprehension strategies, but they don’t require them either.

I’ve long been a proponent of the explicit teaching of comprehension strategies, and yet, there is a part of me that says their omission is not that big a loss.

The reason for my skepticism about strategies? I’m well aware of the fact that many students—perhaps the vast majority of students—don't actually use these strategies when they read. They use them when teachers guide the process, but they don’t do so on their own. I don’t believe, for instance, that “good readers” make predictions before they read a text, even though I have no doubt that good readers could be induced to make such hypotheses under controlled conditions.

The problem is that comprehension strategies are only useful for helping readers to make sense of text that they can’t understand automatically. Many texts are easy for me to read; they are comfortably within my language and knowledge range. This morning I read USA Today and didn’t feel the need to look up a single word or to stop and summarize any of the information.

But if you asked me to read a chapter on theoretical physics—and you were going to evaluate my understanding somehow—that would be a different story altogether. Now I’d have to suit up for heavy combat, which would mean doing various things that I don’t do in my daily reading (like taking notes or turning headers into questions).

What I’m saying is that in the past we taught strategies—overtaught strategies???—but we then asked students to apply them to relatively easy texts (texts at the students' instructional levels). Now, the new standards are asking us to ignore strategies while assigning harder texts.

Talk about the left hand not knowing what the right hand is doing.

I would encourage you to continue to teach comprehension strategies as a scaffold for dealing with challenging text. The point would be to make it possible for kids to make sense of truly challenging texts; the use of strategies could be enough to allow some kids to scaffold their own reading successfully--meaning they might be able to read frustration level texts as if they were written at their instructional level.


Kent Layton said...

As a 5th grade teacher, my MEd program prepared me to engage readers by using different types of 'comprehension strategies'. Today, we'd probably refer to them as creative instructional strategies for making the reading of narrative and expository texts fun and exciting ~ and clearly engaging. Comprehension strategies like Manzo's ReQuest and Guided Reading Procedure and Stauffer's Directed Reading-Thinking Activity are classic strategies that are easy for teachers to plan and carry out over a range of grade levels and texts. While these strategies may not directly teach a comprehension strategy that students can mentally call up each time they are reading text, they require students to do all important metacognitive tasks that good readers use naturally to comprehend and answer questions. More importantly, they allow the teacher to vary the way text is approached for reading and thus, stale routines where essentially the same steps and procedures are minimized on a weekly basis.

Ed said...

I'd suggest that many good readers don't intentionally use comprehension strategies, but do so automatically, just as with decoding/phonics strategies. For example, I think we do often make predictions about text - even easy text - before we read, but we may not explicitly activate a strategy to do so per se. Rather, that's just a very quick, automatic cognitive process that occurs.

Just a with phonics, the idea of explicitly teaching comprehension strategies may not be helpful for readers that are already effective, at least some basic strategies. Many readers naturally pick up letter-sound correspondences, as well as comprehension strategies. However, that doesn't mean they aren't effective with kids who struggle in that area.

In short, I'd suggest that we not base our evaluation of the efficacy of strategy instruction on what already-effective readers do without realizing, but on studies that investigate how such strategy instruction can make an impact.

I suppose we're not too far off on this topic, but thought I'd share anyway!

Tim Shanahan said...


Thanks for this. No question these different approaches put a varied face on your teaching and the novelty of that might interest kids (at least for a while). However, the issue is less about how to present a lesson and it is more about what is it that we need to teach children to do. The point of teaching comprehension strategies is to give students effective ways of keeping focused on the meaning of text.

Tim Shanahan said...


And those are the kinds of studies (studies showing the effectiveness of the strategies) that support the teaching of strategies. Nevertheless, there is no evidence that better readers either routinely use such strategies or that they need to (I suspect most of us are successful with this because of our language processing skills rather than because of our use of formal strategies).

AB said...

So if good readers don't use the strategies at all at times or use them unintentionally at times where were the strategies learned? Do good readers have them because they are good readers or because they have learned it some how? I would like to think that anytime you expose someone to something it creates a file in their brain and the more times exposure happens it allows the file to move into long term memory. I can say I am not an expert so any thoughts would be greatly appreciated. Thanks!

Tim Shanahan said...

Comprehension strategies can come either from individual invention or from social connection (someone else tells you or teaches you). Strategies are only useful if you are struggling to understand the text or if you need some kind of heightened performance on it. Otherwise, usual language processing behavior is sufficient for making sense of a text.

Ed said...

Thanks for your reply Dr. Shanahan. You've certainly challenged me to go back to my library and search for support I've been assuming was present in terms of teaching comprehension strategies.

Ed said...

Dr. Shanahan,

Some further thoughts on the discussion. I still haven't dug through research yet, but have some preliminary thoughts on research and reading comprehension.

Your main response to my original comment was that there is no evidence that effective readers use strategies or need to.

I'd first respond by saying that we need to clarify what we mean by "reading strategies." I'd argue that the use of "strategies" is actually the definition of comprehension. Comprehension (in the reading arena) involves mentally processing language and deriving mental representations based on text. Those processes are arguably "strategies." For example, summarizing a passage, or identifying cause & effect, are "strategies." To argue that effective readers neither use nor need strategies such as summarizing or identifying cause & effect defies logic and the very definition of what it means to comprehend.

On the other hand, many comprehension strategies are more general comprehension or language comprehension skills, rather than skills related to the process of reading. For example, identifying cause & effect is a strategy that could be used with a read aloud passage or lecture. It's not "reading-specific." So, if we take a closer look at text-based comprehension strategies, such as re-reading passage or looking up unknown words, maybe we'd be able to conclude that, with those strategies, effective readers neither use nor need them? I'd argue no. Take the strategy of re-reading text that wasn't initially comprehended. Would it be reasonable or possible to argue that effective readers don't do this?

Here would be my challenge to you: Identify a class of comprehension skills specifically that you think are unhelpful. I'm sure we could also identify an isolated skill or two that may or may not be helpful to teach. However, which class of comprehension skills are suggesting are neither used nor helpful?

Finally, when doing some initial research, I searched my notes for "comprehension strategies," and actually came up with a blog post you had made a while back entitled, "What is the biggest educational change promoted by the common core?" In that, you stated,

"The same can be said about teaching students to comprehend text. The standards don't require you to teach comprehension strategies, but research suggests that if you do you will be more likely to get the students to the standard."

What research were you referring to here? It seems that you have identified research to support instruction of comprehension strategies?

Tim Shanahan said...


Yes, I did point out that there is no convincing evidence that good readers use strategies for comprehension, and the research shows that teaching such strategies to poor readers is helpful, but not with better readers.

I agree that it is important to define terms, but we have more than 40 years of research literature and writings about strategies. Making up your own definition now is a bad idea. A strategy is an intentional behavior towards saving a problem. Thus, when someone intentionally turns headings into questions and then reads and tries to answer the questions, he/she is engaged in some action aimed at increasing understanding or memory for the information. The same can be said of the reader who tries to visualize a description or who stops and summarizes what they are reading every few pages or after each section. Strategies do not include comprehension itself. Nor do they include automatic behaviors that a reader may carry out without any conscious awareness. Thus, the reader who can answer a cause and effect question has not necessarily used a strategy, but the reader who starts from the premise that science chapters will describe causes and effects and who then sets out trying to identify and link those in some way is using a strategy.

The point here was to point out that as useful as decoding and comprehension strategies are, there are also language interpretation skills that we have not been teaching. Muddling the definition of strategies will neither help teachers to make the distinction, nor is it likely to encourage teachers to teach kids to think more effectively about language.

I’d say a good starting point for your reading on comprehension strategies would be the Report of the National Reading Panel. We reviewed more than 200 studies of the teaching of reading comprehension strategies.

Ed said...

Thanks for the reference Dr. Shanahan - I'm familiar with the NRP report, but will review again.

I must have misinterpreted your original comments. I thought I had read that you thought that there was no evidence that teaching comprehension strategies was effective. It seems that you're saying it is effective, at least with struggling readers. We'd be on the same page with this. I'd point out that many strategies - across reading and other subjects - aren't necessarily helpful for kids who already know how to perform the task successfully. I'd follow up with the thought that this is, of course, expected, and not evidence that the strategies aren't useful, though they may indeed be less useful in a Tier I/general education setting.

Thanks for your clarification about the definition of "strategies" as well. I might broaden the conversation to talk about "skills" rather than strategies, and argue that the definition of "skills" does include processes that occur automatically, and are what comprise the general definition of "comprehension," at least in part. If we consider this, then the idea of teaching comprehension "skills" becomes a very different conversation. There would indeed be support that the use of comprehension skills occurs with effective readers, even if no overt, intentional "strategies" were used per se. This isn't just semantics, but gets to the underlying point: Teaching kids overt methods of trying to make sense out of text, especially if they aren't already making sense out of text, is a good idea.

Finally, I'd say that I definitely agree with you that there are underlying general/language comprehension components that contribute to reading comprehension, and that those skills are probably a lot more helpful to children when it comes to making sense out of text. I fully agree that our curricula should include instruction that helps children develop rich language skills, analytical skills, etc. to help them more successfully comprehend.

More broadly, I'm not a big fan of the "this or that" approach to educational methodology. We tend to either like things or not, fall in love with them or let them fall out of favor. We go through too many fads. While I'm not a fan of artificial balance (trying everything under the sun just because we don't want any approach's "feelings"), I do believe in using what works, regardless of whether it fits neatly under a particular paradigm.

Jessie Grant said...

I have always found that students do best with explicit teaching of a skill--even comprehension strategies. Through modeling and guided practice, students are able to learn how thinking during reading occurs. Dr. Shanahan brought up a good point...CCSS does not fully address comprehension strategies yet requires students to read complex text with an emphasis on vocabulary instruction (meanings, nuances of words, etc.). I think that ignoring the instruction of comprehension strategies (simply because they are not included in CCSS) will be a detriment to students down the road. They need to learn how to think during reading and work through the ideas that are included in complex texts. I've enjoyed everyone's thoughts and comments--you got me thinking!

Penny Love said...

The first point made really struck a nerve, on how in America we are constantly chasing fads and not building on the past. I have noticed that todays society does not focus how to have students think effectively. Working in the content of Social studies, when it comes to having students reading classroom material, they tend to limit themselves and not engage in asking questions, or thinking critically. If they do not understand something, instead of stopping and trying to they would rather skip ahead. This could possibly play a role into why teachers are just passing students rather than actually challenging them. We do not have the opportunity to challenge students anymore because we are so focused on making sure we complete a certain test or material before a deadline. If we are to incorporate strategies in the beginning and not what to the end I think it would be more effective for students. I do not feel students are given easy texts per say, I feel they just are not challenged and teachers are focused on making them easier for them to handle and manage.

James Jennings said...

Dr. Shanahan stated that "readers need to be meta-cognitively aware." I think this is a very important statement, and I am very interested in what it actually means and how is it developed, nurtured, etc. I am not certain how much attention or emphasis needs to be placed on whether or not CCSS promotes the teaching of strategies. CCSS promotes "reading," and there is nothing in CCSS that prohibits teachers from using various comprehension strategies. If we can teach strategies that will produce favorable outcomes for CCSS, then both parties win.

Tim Shanahan said...

There is little research on the effectiveness of the various "comprehension skills" that we teach. I would suggest that you shift your attention from some of those question-answering skills towards guiding kids to make sense of challenging text (improving their ability to read the texts silently for extended periods, to figure out the meaning of unknown words, to untangle and make sense of complex sentences, to link the ideas across text through cohesion, etc.). I suspect you'll accomplish a better result.