Blogs About Reading
Shanahan on Literacy
Literacy expert Timothy Shanahan shares best practices for teaching reading and writing. Dr. Shanahan is an internationally recognized professor of urban education and reading researcher who has extensive experience with children in inner-city schools and children with special needs. All posts are reprinted with permission from Shanahan on Literacy.
Comprehension Skills or Strategies: Is there a difference and does it matter?
What’s the difference between comprehension skills and comprehension strategies? Are they synonyms or do we teach different things when we are teaching them?
I’m glad you asked.
Comprehension skills and comprehension strategies are very different things.
They are often confused; the terms are often used interchangeably by those who don’t understand or appreciate the distinctions they carry.
And, most importantly, these concepts energize different kinds of teaching.
The older of the two terms is “reading comprehension skills.” It was used occasionally throughout the Twentieth Century, but really took off in a big way in the 1950s. Professional development texts and basal readers were replete with the term and its use burgeoned for about 30 years before slackening a bit.
“Comprehension strategies” were rarely heard of until the 1970s. The term had wide use throughout the 1980s, both because of the extensive strategy research and because those promoting comprehension skills appropriated the newer, trendier label—old wine in new bottles. (Some of this due to honest ignorance and some to the frequent but idiotic claim of educators that any new practice is “what we were already doing.”)
Use of the term strategies finally overtook skills in 2000, but not necessarily because the practices of teachers actually had changed.
Basically, the term comprehension skills tend to refer to the abilities required to answer particular kinds of comprehension questions. Skills would include things like identifying the main idea, recognizing supporting details, drawing conclusions, inferencing, comparing and contrasting, evaluating critically, knowing vocabulary meaning, and sequencing events.
The old basal readers would make sure kids got plenty of practice with particular comprehension skills by making sure they practiced answering particular kinds of questions.
These days, the core reading programs do pretty much the same thing with the educational standards, with each standard being translated into a question type in lessons and assessments.
For most of these skills, there are no studies showing that they can be taught in a way that leads to higher comprehension, and even in those few instances where there is such evidence, the effects are quite small (and probably due to greater attention to reading the text than to practicing the so-called “skills.”)
In fact, there have been a number of studies and logical analyses showing that these skills lack any kind of psychological reality… they are indistinguishable from one another in test performance, though that hasn’t stopped instructional designers from trying to come up with programs that would teach these skills in a way that would benefit achievement (e.g., Wisconsin Design).
There has been a lot of research into question types over the past 50-60 years. Despite the claims, this body of research is a morass. There are so many variables that may be affecting any of these results, that it would be impossible to know what it all means.
Here are just a few of the variables found to affect how readers answer questions: poor/good readers; low/high knowledge readers; story/information text; the centrality of the information queried; verbatim vs. paraphrase; open-ended versus close-ended questions; reading or re-reading; text available or unavailable during answering; amount of text; factual vs. inferential; immediate recall vs. delay; and so on. (Of course, it is this complexity that undermines any possibility that one can teach question answering as a skill).
Reading has much more to do with being able to read particular kinds of texts and to deal with particular kinds of text features than to answer particular kinds of questions. The ACT (2006) concluded, for example, that if texts were easy, students could answer any kinds of question about them, while with sufficiently complex texts, they couldn’t answer any question types, no matter how simple.
An odd kind of skill the performance of which is totally dependent on the contexts in which it is used. Each text presents information in its own way, and reading comprehension is heavily bound up in the readers’ knowledge of the topic covered by the text. As such reading comprehension (different than decoding) is not a skilled activity, per se.
If comprehension is not a skill, then why has that been such a popular way to teach it? Initially, the concept fit the times. In the late 1950s when it “broke out,” B.F. Skinner’s version of behavioral psychology (e.g., stimulus-response, programmed learning) was in vogue. The idea that learning would result if we could simply induce particular responses to questions and then reward kids for their answers—rinse and repeat—seemed very convincing.
It has been harder to eradicate than a fungus, I assume, because it appears to map onto educational standards and the high-stakes tests. Principals and teachers assume it makes sense to practice the “comprehension skills” that tripped the kids up on the tests. So, they “use their data”: combing through test results to identify the kinds of questions that students failed on and then practicing those supposed skills over and over in the hopes the kids will be enabled to answer such questions on the next test. That it hasn’t actually worked doesn’t seem to dissuade them at all.
The idea of comprehension strategies is more recent, and it has a substantial body of research behind it. If the notion of comprehension skills emerged from behaviorism, then comprehension strategies is the child of cognitive psychology. Instead of repetition and automaticity as the watchwords to learning, intention and decision-making and thinking move to the forefront with strategies.
The basic premise of strategies is that readers need to actively think about the ideas in text if they are going to understand. And, since determining how to think about a text involves choices, strategies are tied up in meta-cognition (that is, thinking about thinking).
Comprehension strategies are not about coming up with answers to particular kinds of questions, but they describe actions that may help a reader to figure out and remember the information from a text.
For example, the idea of the summarization strategy is that readers should stop occasionally during reading to sum up what an author has said up to that point. Doing that throughout a reading and at the end has been found to increase recall… recall in general, not of any particular type of information.
Another frequently studied comprehension strategy is questioning. Students read, stopping throughout to quiz themselves on what the text says (and going back and rereading if one’s questions can’t be answered). The point isn’t to ask particular kinds of questions, so much as to think about the content more thoroughly, more actively than one would do if they just read from the first word to the last.
The same can be said about monitoring, visualizing, thinking about the way the text is structured or organized, rereading, and connecting the content with one’s prior knowledge.
These kinds of actions—these strategies—are used intentionally by readers to increase the chances of understanding or remembering what one has read.
Comprehension strategies need to be practiced too; however, they aren’t learned by repetition and reinforcement, but by gradual release of responsibility (including modeling, explanation, guided practice).
If you are serious about raising reading achievement, there is no point to teaching most comprehension skills. (Note: vocabulary is often listed as a comprehension skill and there are benefits to teaching that.).
On the other hand, there are very good reasons for teaching comprehension strategies, but there are at least three big problems with that kind of teaching.
First, studies of comprehension strategies have tended to be brief, usually about 6 weeks in duration (there are exceptions). Somehow that has been translated into substantial amounts of strategy teaching across students’ school lives. To be perfectly honest, no one, including me, knows how much strategy instruction is needed. But there is certainly no evidence that there are benefits to be derived from 8 to 10 years of 30-35 weeks of strategy teaching.
Second, the only point to using strategies is to make sense of texts that couldn't be grasped without that effort. Many texts are easy enough that a reader would not need to expend that amount of energy in comprehending. Unfortunately, most strategy instruction that I have seen takes place in texts that frankly are relatively easy for the kids to read. That means they have to pretend to apply those strategies in situations that wouldn’t benefit from such effort. If kids ever do apply these strategies to complex text, they are usually on their own. Most skip the effort since what such teaching conveys is that you don’t need strategies.
Finally, even major proponents of explicit comprehension strategy instruction (like the late Michael Pressley, for instance) argued that as important as it was to teach strategies, teachers needed—even when teaching them—to make sure the kids were actually learning the text content and not just the strategies they were using to think about that content. That principle largely has been ignored by teachers and publishers.
More on the teaching of comprehension skills and strategies next week; should we guide students' reading?