Some parents are having trouble making sense of their kids’ reading assignments while schools are … [+]
There’s been heated debate over reading instruction within the education world lately—and some confusion. Now that the coronavirus has closed most schools, parents may be experiencing confusion too. Here’s some help.
“This is the remote reading lesson I taught to my kindergartner yesterday,” a New York City parent named Michael LaForgia recently tweeted. “I am a Pulitzer Prize-winning writer, and I could not correctly identify the ‘main idea’ in this passage—though the instructions assured us it was in there.”
The passage, sent home by the teacher, wasn’t an excerpt from Kierkegaard. It was a paragraph briefly mentioning sharks, whales, dolphins, and other sea creatures. The teacher eventually revealed, via a recording, that the main idea was “ocean animals”—which also happened to be the title.
Welcome, parents, to the wacky world of reading instruction—or more specifically, comprehension “skills and strategies” instruction. The ocean animals activity was intended to teach kindergartners the “skill” of finding the main idea. There’s a host of such supposed skills, including “comparing and contrasting,” “determining the author’s purpose,” and “making inferences.” The theory is that once kids learn the skills, they can apply them to understand any text put in front of them—including, eventually, the books they’ll be expected to read in high school. Acquiring knowledge of things like social studies and science, it is believed, can wait until later. If that doesn’t make sense to you, you’re not alone.
But first, a little background. There are two basic components to reading instruction: decoding—or sounding out words—and comprehension. Lately, there’s been a lot of debate over the first component, centering on how important it is to teach phonics. That issue has been dealt with well elsewhere (see, for example, here and here). The bottom line: evidence shows that many kids—probably most—need systematic phonics instruction in order to become good readers, and virtually all can benefit. This evidence has been around for a long time, but, for complex reasons, many educators are unaware of it or resist it.
Back to comprehension. It gets less attention than phonics, but the problems with the way it’s taught may be even more pervasive. The prevailing theory, outlined above, has led to many instructional hours—beginning in kindergarten and sometimes lasting through middle school—spent having kids read disconnected passages like the one on ocean animals. Students practice the “skill of the week” on books or texts at their own “reading levels,” which may be well below their grade level. Often the texts are fiction, but even when they’re nonfiction—as with the one on ocean animals—the focus is on the “skill” rather than the content. Content-rich subjects like social studies, science, and the arts have been pushed aside to make more time for this kind of comprehension instruction. This approach began before the era of high-stakes reading tests, but the tests—which seem to call for comprehension “skills”—have intensified it.
But evidence from cognitive science indicates that the main factor in comprehension is how much knowledge and vocabulary the reader has relating to the topic. So the best way to boost comprehension is to build kids’ knowledge through the very subjects schools have marginalized. Rather than asking students to repeatedly locate the main idea of a paragraph (assuming there is a main idea), it would make sense to teach them about things like sharks and whales and dolphins. It would also be way more interesting. To be clear: individual teachers aren’t responsible for this misguided approach. It’s the result of a system that includes teacher training, curriculum, and policy.
While the vast majority of elementary schools focus on so-called skills, an increasing number of educators are questioning that approach. Across the country, schools and districts have been adopting new literacy curricula that focus on systematically teaching kids about things like sharks and whales rather than “finding the main idea.” Perhaps because that movement is gathering steam, there’s now pushback—pushback that relies on evidence from the many studies indicating that teaching comprehension strategies is effective. Yes, such evidence exists. The question is how to square it with the evidence showing knowledge is crucial to comprehension.
One of the prime pieces of evidence put forth by advocates of building knowledge—myself included—is something popularly known as the baseball study. In that experiment, conducted over 30 years ago, researchers found that kids who scored low on a standardized reading comprehension test nevertheless did quite well on a comprehension test when the topic was something they knew about—in this case, baseball. In fact, the “poor” readers who knew a lot about baseball did significantly better than the “good” readers who didn’t. That experiment strongly suggests that knowledge of the topic trumps general comprehension “skill.”
The significance of that study has now been questioned by a prominent reading expert, Dr. Timothy Shanahan. Over a distinguished career, Shanahan has championed phonics instruction, debunked the theory of “leveled reading,” and become a trusted source for many educators trying to make sense of conflicting claims about literacy instruction. And in a recent blog post, he dismissed the baseball study as a one-off, arguing that there are no other studies that “pit reading up against knowledge in this way.” So we really shouldn’t pay too much attention to it.
In fact, there are other studies quite similar to the baseball study, both in terms of methodology and results. Researchers in Germany, for example, did a “soccer study.” They found that “poor” readers who knew a lot about soccer outscored “good” readers who didn’t by a significant margin, when the topic was soccer; in fact, they were three times as likely to make accurate inferences about the passage. Another study involved preschoolers from different socioeconomic groups. First, researchers read all the kids a story about birds, a topic they had determined that some—those from wealthier families—knew a lot more about. Those kids did better on a comprehension test on the story. Then the researchers read all the kids another story about made-up animals called wugs, equalizing background knowledge. This time, when they tested comprehension, there was no significant difference between kids from higher and lower socioeconomic groups.
But Shanahan’s argument doesn’t really rest on the supposed lack of replication of the baseball study. He acknowledges there’s lots of evidence that background knowledge is important to reading comprehension. He says there’s no evidence that building knowledge causes an increase in comprehension—just “lots of correlations.” Therefore, he says, it makes no sense to reduce the amount of time schools spend on comprehension instruction.
How would you demonstrate causation rather than correlation? The standard method is to conduct a randomized controlled trial: take at least two randomly selected groups of people; “treat” one group with whatever you’re testing, and have the other one be a “control” group; and see whether what you’re testing makes a difference. If it does—and if there are enough similar studies with similar results—you can reasonably draw the conclusion that the “treatment” is causing the effect. That basically describes the studies supporting comprehension strategy instruction.
It’s true we don’t have a bunch of such studies showing that boosting knowledge causes better reading comprehension. One reason is that building knowledge is a gradual, cumulative process. Especially when measured with standardized tests, which aren’t tied to any particular body of knowledge, it can take years to see results. And it’s hard to follow a particular group of students for years because so many switch schools. It’s also hard to ensure that any curriculum is being delivered well, or at least in a similar way, by the many different teachers students will encounter. Not to mention that knowledge-building elementary curricula have only been in existence for a few years.
On the other hand, there are some studies showing that, compared to a focus on content, strategy instruction makes no difference and can even have a negative impact. And some have questioned whether strategy instruction itself—which originated in observations of techniques that “expert” readers use unconsciously—is based on correlation rather than causation. “Good readers often relax with a glass of wine,” Doug Lemov, author of the best-seller Teach Like a Champion, once told me, “but I don’t think we’re talking about that being an instructional strategy for third-graders.”
Then there’s the question of whether the kinds of things that strategies try to get kids to do—summarize a text, or ask themselves questions about it—could be taught through the content areas rather than in a separate reading class. History teachers could ask kids to summarize a chapter on the beginning of the Civil War. Science teachers could have students question themselves as they read a description of cell division. Those are good techniques for getting kids to understand, absorb, and retain content. They might also accomplish the goal of strategy instruction—getting kids to do these things when they’re reading independently. Maybe there’s no evidence to support that approach, but as far as I know it hasn’t been studied.
There are a lot more problems with the way comprehension is currently taught—problems that Shanahan’s post on the baseball study overlooks, even though he’s called them out previously. I’ll address those in my next post. So if you’re still confused, stay tuned.
