Want better readers? Spend less time teaching kids to find the main idea, ‘Knowledge Gap’ author Natalie Wexler argues

In the average public elementary school, third graders spend nearly two hours a day on reading instruction, according to a recent federal survey. That far outstrips any other subject, with math coming in second at around 70 minutes a day, and science and social studies getting about half an hour a day each.

Teachers may think this approach is the best way to improve students’ reading ability. But in her new book “The Knowledge Gap,” journalist Natalie Wexler argues against skimping on science and social studies and emphasizing specific reading skills. She says that this approach, paradoxically, hurts students’ ability to make sense of what they read. (Find an excerpt of the book here.)

She builds her case with cognitive science that suggests that once students have learned to sound out words — “decode” — the key to understanding a text is having solid background knowledge on the subject.

In other words, if you already know a lot about education, you’ll probably have an easier time making sense of articles in Chalkbeat than, say, in Foreign Policy.

The implication, Wexler says, is that schools should start teaching science and social studies content early and often. Wexler draws from the work of E.D. Hirsch, the University of Virginia professor and prominent advocate of these ideas.

But what knowledge — and whose knowledge — should be taught? And what does Wexler make of the fact that there are relatively few studies directly linking a content-rich curriculum to better academic outcomes? Chalkbeat asked Wexler that and more. This interview has been edited for clarity and length.

Can you walk me through the main thesis of the book?

The title “The Knowledge Gap” is a reference to the achievement gap, which is basically referring to the test score gap. And I’m particularly focusing on the gap between those students from the top end of the socioeconomic scale and those at the bottom end — and that gap is significant. It hasn’t really budged in 50 years, or maybe it’s gotten worse.

We’ve been looking at that gap as a gap in skills. American elementary schools, and to some extent middle schools, have long approached reading comprehension as though it’s a matter of teaching generally applicable skills, like let’s practice finding the main idea and let’s practice making inferences. The theory is, it doesn’t really matter what content the kids are using to practice those skills; if they just get good at those skills they will be able to apply them eventually to any text that’s put in front of them, whether it’s on a standardized test or in high school.

That approach has been intensified in the last 20 years by the advent of high-stakes reading tests, because it looks like they’re measuring those skills. So teachers, policymakers, reformers have all assumed we should just double down on teaching those skills.

The problem is that, as cognitive scientists have known for decades, the most important factor in reading comprehension is not generally applicable skills like finding the main idea — it’s how much knowledge and vocabulary the reader has relating to the topic. So if we really want to boost reading comprehension, we should be doing the opposite of what we’re doing — especially in schools where test scores are low — which is cutting subjects like social studies and science that could actually increase students’ knowledge of the world and instead spending more time on these reading comprehension skills.

The way that it’s related to this gap, the test score gap, is that kids from better-educated families pick up a lot of academic or sophisticated knowledge and vocabulary at home. But kids who are coming from less educated families rely on school to get that kind of knowledge, and they’re actually the least likely to get it there.

Why focus on elementary school? Because it seems like that same logic would also apply in later grades as well.

Well, the elementary school day has long been dominated by reading and has only gotten more so. The theory is that you teach kids reading comprehension in elementary school and then when they get to middle school and high school, they can start acquiring knowledge through their own reading. So the high school curriculum has more content. It always has.

The potential power of what we do in elementary school is huge. Kids at that age, especially early elementary grades, they’re eager to learn stuff. So it’s a tremendous wasted opportunity.

[Note: It’s hard to pin down clearly whether time spent on reading instruction has really grown. One survey of school districts in the wake of No Child Left Behind found that about a third reported cutting science or social studies, while a majority said they had ramped up time on reading. On the other hand, federal data, based on self reports from schools, shows that reading instruction has dominated elementary schools since at least the late ’80s, and there hasn’t been a clear upward trend since, or a downward one for science and social studies.]

In the book you write that elementary schools are where “the real problem has been hiding in plain sight.” But on the federal NAEP exam, we’ve actually seen the biggest gains on reading tests in early grades. For instance, since 2000, black fourth graders have gained 16 points on the NAEP, which is not a trivial number. Now, there haven’t been gains since 2009, but those earlier gains have sustained themselves. And so I think if I were just looking at the NAEP I would say, actually, in elementary school we’re doing OK, and we really need to focus on those later grades.

And you would not be alone. That is a frequent mistake, because just like elementary school books don’t assume as much background knowledge as high school level texts, elementary level standardized reading comprehension tests also don’t assume as much sophisticated background knowledge and vocabulary as high school level tests. But if you’re not planting the seeds in elementary schools, you’re not exposing kids to those more sophisticated concepts and more sophisticated vocabulary, they may well not be able to absorb that in high school.

That’s another thing about knowledge. Knowledge doesn’t just help you understand a reading passage. It also helps you absorb and retain information if you have an existing framework to fit it into. Or you could say knowledge is like Velcro — it sticks to other related knowledge. So if a kid encounters a word like “labyrinth” for the first time or a phrase like “Achilles heel” for the first time in 10th or 11th grade, you might explain it to the kid then, but it may not stick.

I think that the danger is we thought elementary school is the bright spot in education because we’ve looked at those rising test scores at the elementary level and we thought, OK, everything’s fine there. And we’ve looked at the stagnant test scores at the high school level and we thought OK, the problem’s really in high school, and that’s where we need to work to solve it. I won’t say it’s too late, but it’s just much much harder to solve the problem if you wait until high school.

In your book you say, “There aren’t yet any reliable studies showing that [a coherent knowledge-rich] curriculum will outperform either a skills-focused curriculum or a content-focused one that lacks coherence,” but “it’s reasonable to assume that’s the case.” The fact that there aren’t any reliable studies about this seems like a really big caveat at the heart of your book.

There is evidence that focusing on content can boost kids’ reading comprehension scores. They’re not randomized controlled studies, but there is some evidence of that. What is harder to find evidence of is that you need a curriculum that builds logically from one grade to the next. And that’s hard to get because kids move around, especially in lower income levels, and there aren’t that many schools implementing that kind of a curriculum.

One that comes to mind — there’s a curriculum called Bookworms, and there was a study of a school district that implemented that curriculum. After just one year of implementation, schools implementing that curriculum did better than demographically similar schools in the districts that weren’t implementing that curriculum.

You spend a lot of time in your book making the case that schools should teach kids content, but you spend less time describing what content exactly should be taught. That seems like a really important question, and I’m wondering if you can address that?

I don’t feel like it’s my role to prescribe what content should be taught. And certainly it’s not either realistic or maybe desirable to try to do that on a national scale in this country, where curriculum has historically been very much a matter of local control. But there are now, for the first time, several curricula out there that have defined bodies of knowledge to be transmitted to elementary school students. Districts, schools, states can look at those different curricula and see which one fits their needs.

I would think generally there are different bodies of knowledge that could be taught and there are different ways to teach them. But we need to think about what is it that we want kids to know when they get out of high school that will enable them to read and understand a newspaper or news report or to vote responsibly or serve on a jury responsibly. So there are certain basics I think we can all agree on.

You mention in the book the incident with E.D. Hirsch at a press conference in 1987 where he was rolling out the specific things that he thought should be taught, and a reporter asked him, is Cinco de Mayo on your list? And he said he didn’t know what that is. That raises the question of whose knowledge is being taught. And I think that’s one reason why some people might be skeptical about teaching knowledge and figuring out who is going to determine what knowledge is taught. I’m wondering your thoughts on that?

I think Hirsch’s argument was really misinterpreted and I think one reason for that was he made the mistake of being very specific about what should be taught. He appended this list of 5,000 words and phrases that every American should know to his book “Cultural Literacy.” And people assumed that he just sat back in his armchair and his tweed jacket, being an academic and an old white guy and thought, “Well, what is it that people should know? What do I think is important?” It’s really not what he did. He did an empirical study with some colleagues: What is the knowledge that is assumed by newspaper editors, by trial lawyers arguing before juries? It turns out that a lot of that knowledge is Eurocentric because we are still primarily a Western society.

Of course, things change. The body of knowledge and information that is taken for granted among those who have what is called cultural literacy — for lack of a better word, the elites — that does change, and E.D. Hirsch did recognize that. I think he thought it changed pretty slowly and it may have been changing faster in the last 30 years than it did in the previous 30 years. But the Core Knowledge Foundation has continued to test the waters of what is the knowledge that is assumed. Cinco de Mayo, now as compared to 1987, yeah, you would expect people to be familiar with that, and believe me, E.D. Hirsch is now familiar with what that is. That doesn’t mean that people don’t need to know what happened during the American Revolution and the Civil War. There’s a lot of stuff that’s not changing.

One issue there might be with surveying newspaper editors and talking to trial lawyers is that their conceptions of things that are important may be biased. Trial lawyers and newspaper editors likely are disproportionately white and disproportionately come from an affluent background.

It depends on how you define bias. We’re not talking about knowing the rules of playing polo. Sure, there are certain things that people from wealthier backgrounds are more likely to know. Some of those things are going to be important to functioning well in society and some of those things are not going to be. But we want kids’ destinies not to be determined by their ZIP code.

That’s been the rallying cry of education reformers, and a lot of that has translated into, we want kids from any socioeconomic background to be able to go to college and thrive there. And if you get to college and you don’t know who Winston Churchill is or you don’t know what Stonehenge is you’re going to be at a disadvantage. So it’s not just cultural bias. It’s also what is the knowledge that any student, any person in this society, needs to have a fighting chance of succeeding and contributing to society and to our democracy.

This may seem like a silly question, but a lot of cultural references that people may encounter may be non-academic in nature. I’m thinking of an example for myself where I have never watched “Game of Thrones” and I actually encounter references to it all the time. Should schools be teaching students cultural references that they’re likely to encounter that go far beyond science and social studies?

I’m with you on “Game of Thrones.” We must be the only two people left in the country that haven’t watched it. We could be teaching those things in school, but I bet you students are a lot more familiar with those cultural references than they are with Achilles heel or the War of 1812. I think we also have to think about what is likely to survive in the culture and be used as a touchstone 20 years from now that people will need to have stored in their long-term memory so that they can easily understand a text or or whatever is put in front of them. We just don’t know exactly what it is, but there’s certain things that have stood the test of time, and maybe we should go with those.

You say at one point that education is our “only hope” to reduce multigenerational poverty. The implication is that fixing the knowledge gap is our only hope to reduce multigenerational poverty. That seems like a really strong claim. And it also suggests that direct anti-poverty programs are not a way to reduce multigenerational poverty.

We’ve had a war on poverty going on since the 1960s and we haven’t really reduced multigenerational poverty. I’m not saying we should do away with anti-poverty programs but I don’t think they’ve worked to break that cycle for most people who are trapped in it.

You can no longer get a good-paying job with an eighth-grade education going to work in a factory. That’s just not really there anymore. I’m not saying everybody needs to go to college, but if we want to try to level the playing field to some extent and enable kids to move up in society through the socioeconomic levels, I don’t really don’t see any alternative to education.

Sure, kids need food. They need a stable place to live. Health care, all of those things. But that’s not necessarily going to enable them to get a really well-paying job or live a really fulfilling and rich life. I mean, they may, but a lot of them are deprived of that opportunity because they are deprived of a meaningful education.

I’ve seen this, for example, in the history of women’s rights. Look to what preceded the push for women’s rights, that 1848 Seneca convention and all of that — a tremendous explosion in women’s education in the earlier part of the 19th century. And what that did for women was it changed their self concept, changed their idea of what they were capable of doing. I think that that is a pretty universal experience. So I think that is ideally what education can do. It can change your sense of who you are, what you deserve from society, and what you’re capable of.

[Note: The official poverty rate fell steeply starting in 1959, but hasn’t budged much since 1970. But research that more fully accounts for welfare benefits suggests that such programs have helped cut the poverty rate from 26% in 1967 to 16% in 2012. “We’ve done a better job of cushioning the impact of poverty by expanding the social safety net,” Wexler said in a follow-up email. “There’s certainly lots of evidence of growing inequality in U.S. society and low rates of social mobility — that’s what I have in mind.”]

Can you talk about what you think the takeaway from your book is for teachers and also for policymakers?

For both, the initial thing is just to understand why what we’ve been doing hasn’t really been working and why. Secondly, it really has to be a multi-pronged effort if we want things to change. It probably has to start with the adoption by policymakers or central office officials of a content-focused elementary curriculum. But if it’s just imposed on teachers from the top down as so many education initiatives are — without teachers understanding why it’s a good idea — then it probably won’t really get into classroom practice in the way we want. Historically teachers have been able to just close the classroom door and do what makes sense to them, and they still do that.

We need to also help educators understand why this new approach makes sense and help them understand how to implement it well because it is a big change, especially at the elementary level for most teachers. So they need support, coaching, and time to adjust to it.