Why We're Teaching Reading Comprehension In A Way That Doesn't Work

The key factor isn't "skills and strategies"--it's knowledge.

---

There’s been a lot of concern about phonics instruction in recent months, sparked by an illuminating new audio documentary. But there’s another aspect of reading—comprehension—that is equally crucial, and teacher training in that area is even more problematic.

As the documentary details, many teachers—and professors of education—are unfamiliar with the overwhelming evidence that systematic phonics is the most effective way to teach children how to decode written language. While there's been some pushback, quite a few teachers who have listened to the documentary or an accompanying piece on NPR—or read the New York Times op-ed by the documentary’s producer, Emily Hanford—have expressed dismay that they were never given this information as part of their training.

But there’s been little discussion of the even more widespread problems with training in comprehension instruction. True, compared to phonics, teacher-education programs are more likely to say they cover reading comprehension. But what prospective teachers learn about comprehension in those courses is dangerously inaccurate.

One reason is the influential 2001 report of the National Reading Panel. The report endorsed five “pillars” of reading instruction, including phonics, phonemic awareness, fluency, and vocabulary. The fifth pillar was instruction in strategies designed to boost comprehension, such as learning to summarize or make a graphic representation of a text. While many educators challenged the report’s findings on phonics, they embraced its endorsement of comprehension strategies.

What the report failed to mention was the strong evidence showing that the most important factor in comprehension isn’t mastering strategies: it’s how much knowledge a reader has of the topic. In one well-known experiment, students who scored poorly on a reading test but knew a lot about baseball outperformed “good readers” who knew little about baseball—when the reading passage was about baseball. Similar studies have yielded the same kind of result, and scientists agree on the key role that relevant knowledge plays in comprehension. In fact, the comprehension strategies endorsed by the panel all rely on activating prior knowledge—which means they only work if a reader has enough background knowledge to understand the text in the first place. But that’s one of many things prospective teachers never learn.

In a typical comprehension lesson, a teacher focuses on a supposed skill or strategy, like making inferences or determining an author’s purpose. But most of the things teachers spend hours on every week were never endorsed by the National Reading Panel and have little or no data behind them. As reading expert Tim Shanahan has observed, teaching such “skills” is like pushing the elevator button twice: it might make you feel better, but it won’t make the elevator come any faster.

Even when teachers focus on a strategy that is backed by evidence, they don’t implement it in the way supported by research. Rather than putting a difficult text in the foreground and modeling whatever strategies might help students extract its meaning, teachers put a strategy in the foreground and choose simple texts that lend themselves to demonstrating it, without regard to their topics. And they teach comprehension day after day, year after year—sometimes through high school. But studies have shown that after only two weeks of strategy instruction, students stop getting benefits.

After the teacher explains a comprehension “skill,” students go off to practice it on books at their supposed individual reading levels—easy enough for them to read on their own or with minimal help. But there’s no evidence that this system of leveled reading boosts comprehension, and studies have found that kids can learn more from a text above their supposed level—if a teacher helps them understand it. Plus, leveled reading does little to build knowledge, a process that generally requires staying on the same topic for at least a couple of weeks. As with the texts teachers use to model “skills,” the books children use to practice them aren’t organized by topic.

Another pervasive and dangerous misconception is the belief that students need to learn to read before they can “read to learn”—that is, before they can start acquiring knowledge of the world, through their own reading. As a result of this assumption, the elementary curriculum has long been heavily weighted toward reading. That has become even more true in recent decades, due to the advent of high-stakes testing in reading and math. Especially in schools where test scores are low, subjects like history, science, and the arts have been squeezed out of the curriculum, sometimes through middle school.

But the idea that kids don't need to acquire knowledge until after they’ve learned to read ignores the fact that gaining knowledge is part of learning to read—or learning to understand what you read. Even while they’re learning to decode, children need to listen to adults reading aloud from sophisticated, knowledge-rich text. Otherwise, they’ll lack the knowledge and vocabulary that will equip them to understand that kind of text once they’re able to decode it themselves. As any parent knows, children can take in far more sophisticated concepts and language through listening than through their own reading. That remains true, on average, through middle school. And the longer we wait to start building kids’ knowledge, the harder it becomes to close gaps between those lucky enough to acquire knowledge outside school and their less fortunate peers.

In most places, the advent of the Common Core State Standards has only made things worse. Previously, elementary students got a steady diet of fiction. In an effort to build knowledge, the standards have called for at least 50% of their reading to be nonfiction. Most teachers, however, have continued to focus on skills, adding new “nonfiction skills” like identifying different “text structures.” But nonfiction assumes more background knowledge than fiction. It’s one thing to make an inference about a fictional character’s thoughts, based on your knowledge of human nature. It’s quite another to make an inference about, say, Brazil if you’ve never even heard of the place.

Still, there are signs things are changing. Just a few years ago, there were no elementary literacy curricula designed to build students’ knowledge. Now there are several—and two large urban districts, Baltimore and Detroit, have each started implementing one, while the state of Louisiana has been encouraging their adoption.

Ideally, prospective teachers will start getting accurate information about reading comprehension during their training. But that may not happen anytime soon. Education schools have historically been disconnected from scientific research on the learning process; their lack of interest in or familiarity with phonics is only one example. The good news is that even once they’re on the job, teachers can learn how to provide students not only with the skills they need to decode words but also with the knowledge that can unlock their meaning.

This post originally appeared on Forbes.com in a slightly different form.