July 26, 2023 – Knowledge Matters Podcast
American education has a number of serious problems – and our failure to start building kids’ knowledge early is a fundamental one. By now you know that reading comprehension is complicated and, as you’ll hear, so is the explanation for what has gone wrong with the way American schools have approached it.
In our sixth and final episode, Natalie explains how we ended up in a place it’s not clear anyone wanted to go, in the grip of a reading crisis that goes far beyond the important issue of how we teach students to decode.
Let’s start with a quick overview of how we got here. For many teachers, and especially those in elementary and middle school, it might be hard to imagine a world where we don’t teach or test reading comprehension in the abstract, as a separate “subject.” That idea has become deeply woven into the system. But in fact, there was a time when “reading instruction” just meant teaching kids to decipher or decode written words. The term wasn’t also used as it is today, to mean teaching kids to understand the texts they read. That was taken care of through teaching them about history, literature, science, geography, and the arts.
We now use the word reading to cover two very different things: decoding and comprehension. And that has led to the kinds of problems we’ve been hearing about from teachers–trying, for example, to teach students the skill of making inferences without providing the foundation of knowledge that would enable them to actually do that.
A reading researcher named Alan Kamhi, has suggested that we redefine “reading” to just mean “decoding”, pointing out that would free up teachers to spend a lot more time on those other subjects, the ones that build the knowledge that’s actually crucial to reading comprehension. However, such a radical change isn’t likely to happen anytime soon.
Understanding how decoding and comprehension got tangled up requires a look back in time. Early in the 20th century, a surge in high school enrollment led to students being sorted based on reading comprehension and IQ tests. These tests aimed to gauge students’ supposedly fixed learning capabilities, not their progress.
Around the same time, textbooks for younger students began emphasizing comprehension. Initially, students answered questions about the text, but soon textbook authors developed workbooks designed to teach comprehension skills in the abstract.
By the 1950s, this method had become entrenched in the reading textbooks called basal readers. However, many teachers found the passages in the basals uninspiring and the questions too rigid. This dissatisfaction gave rise to the Whole Language movement in the 80s, which prioritized quality children’s literature and rejected traditional comprehension exercises. But when evidence showed many students needed phonics instruction, Whole Language faded out, and around 2000 Balanced Literacy took its place.
While Balanced Literacy was supposed to blend phonics instruction and exposure to good literature, its Whole Language roots led to continuing skepticism of systematic phonics instruction. On the other hand, Balanced Literacy embraced skill-based comprehension instruction, even though Whole Language had disdained it.
It’s not entirely clear how that happened, but Whole Language teachers had noticed that students often struggled with comprehension, particularly when reading nonfiction. Those teachers turned to then-recent research by cognitive psychologists on comprehension strategies, adapting the strategies to elementary classrooms.. While those metacognitive strategies were initially seen as quite different from the skills basal readers taught, the distinction between the two gradually blurred, and by the time Balanced Literacy emerged, many teachers were back to trying to teach the same skills as the basal readers.
Meanwhile, on the testing side, reading comprehension assessments became a lot more important. After Congress enacted the “No Child Left Behind” legislation in 2001, schools and teachers started being evaluated on the basis of their students’ scores on state reading tests. Those tests seem to be assessing skills like “finding the main idea” and “making inferences,” so teachers began spending more time than ever on that kind of practice.
Although standardized reading comprehension tests weren’t initially designed to measure student progress, they are now used for exactly that purpose. And whilehe doesn’t blame the people who created them, Hugh Catts says:
In the creation of them, we’ve produced something that’s really unfair for the student; it’s unfair for the teacher; and it’s unfair for the school system.
Hugh argues that comprehension tests are unfair to students because they essentially reflect knowledge that some of them haven’t had access to. They’re unfair to teachers because they force them to focus on test-taking and comprehension skills, since no one knows what content will be on the test. And they’re unfair to schools because they’re evaluated on the basis of scores that may not capture the knowledge their students have actually acquired.
A study Hugh conducted some years ago showed that students learned what they’d been taught, including new vocabulary, but that didn’t carry over to improvements in standardized reading test scores.. That’s because it can take a while before students acquire enough vocabulary to enable them to comprehend text on subjects they’re not already familiar with. We now have some data indicating how long it might take for that to happen.
James S. Kim‘s team at Harvard tested the effects of a content-rich reading curriculum called MORE, which taught students about science beginning in first grade. By the end of second grade, these students were better able to comprehend texts related to what they’d learned than their peers in a control group could. However, they didn’t perform better on completely unrelated “far transfer” topics–which are the kinds of topics students are likely to encounter on standardized tests.. By the end of third grade, though, students did show improved comprehension on those “far transfer” topics, suggesting that a knowledge-building curriculum could improve general reading comprehension in about three years, with lasting effects.
A second study by David Grissmer, at the University of Virginia, found a comprehensive knowledge-building curriculum from kindergarten led to better scores by third grade, with sustained gains through sixth grade. The results were so significant that if extrapolated nationwide, the US – which is currently ranked 15th on an international reading test for fourth graders – would instead rank among the top five countries. Most notably, lower-income students’ scores matched those of higher-income peers, effectively closing the “achievement gap.”
More long-term research is needed, but such studies are costly and time-consuming, and it’s not clear there are any currently in the pipeline. If we await more definitive proof of the effects of knowledge-building on comprehension, many students may continue to struggle in the meantime. Given all the other evidence we have about the key role of knowledge in comprehension, does it make sense to maintain the status quo?
Delaying change is likely to lead to the perpetuation of two kinds of knowledge gaps:
The first is the gap between proficient and poor readers. Proficient readers, with larger vocabularies, can read more advanced books and more easily absorb new vocabulary—since knowledge sticks best to other, related knowledge. That enables them to read even more sophisticated books. But poor readers, with smaller vocabularies, are confined to simpler books and may struggle to retain whatever new words they contain because they have no prior relevant knowledge for them to stick to. The gap between these two groups grows wider every year, and by middle or high school it can be so vast that it’s almost impossible to narrow.
The second gap, typically evident at higher grades, is the one between what the curriculum assumes students know and what many actually do know. Especially at school where test scores are low, many students get a steady diet of reading and math through middle school, with little exposure to history or science, leading to knowledge gaps that hinder their reading comprehension.
If schools improve their early-grade phonics instruction without also building knowledge, they’ll only get a short-term increase in reading scores. : In the earlier grades, scores are likely to rise, but by middle school, when comprehension becomes more critical than decoding ability, the bump from better phonics instruction fades out..
Students who lack background knowledge often struggle not just on tests but also in class. And especially in high-poverty high schools, students may enter with significant gaps in their knowledge of the world, preventing them from engaging in grade-level work in subjects like history. For example, scores on national tests in American History hit a new low in 2022: only 14% of eighth graders scored proficient or above, and 40% scored below the “basic” level. Scores in civics are only slightly better.
Students may not acquire much more information about history at higher grades. One survey, for example, reported the shocking statistic that 11% of US adults haven’t heard of, or aren’t sure if they’ve heard of the Holocaust. For millennials, the figure is 22%.
Closing these knowledge gaps is important. It’s important for the untold numbers of students whose potential remains to be unlocked – students who might otherwise go through school and life feeling like failures, when in fact, it’s the system that has failed them. It’s important for society, which will otherwise be deprived of those students’ potential. And it’s important for democracy, which depends on a citizenry that can understand the world well enough to make informed decisions.
As Spring Cook, the former Texas teacher you met in Episode 1, put it:
It is a matter of equity, it’s a matter of democracy, and when we’re able to give students those skills and that knowledge at an early age, then think what a better society will have.
We hope you agree.
The challenge, of course, sometimes is, people want overnight success, and we recognize that it’s going to take time. We’re going to have stronger kindergarteners, stronger first graders, stronger second graders, and then over time, we’ll see the gains, they’ll take care of themselves. And so it was a matter of telling teachers and staff members giving them permission to not worry about what we call the STAR test.
Welcome to Episode 6 of the Knowledge Matters Podcast. I’m Natalie Wexler.
That was Dr. LaTonya Goffney, superintendent of the Aldine School District in Houston, Texas.
As we’ve heard in previous episodes, reading comprehension is complicated – way more complicated than teaching kids how to sound out written words. It depends a lot on building students’ knowledge and vocabulary. And that takes time.
It also requires teachers to pay attention to certain things they may not be used to focusing on, like whether they’re building kids’ knowledge in a coherent way. And it requires them to worry less about things they’re used to paying a lot of attention to, like scores on state reading tests. The “STAR” test that LaTonya Goffney mentioned is the Texas state test.
Just as reading comprehension is complicated, so is the explanation for what has gone wrong with the way American schools have approached it. We could spend a lot more time talking about that, and what we can do to change our approach. But we only have one more episode – this one. So we’ll just hit some key points.
Let’s start with a quick survey of how we got where we are. For many teachers, and especially those in elementary and middle school, it might be hard to imagine a world where we don’t teach or test reading comprehension in the abstract, as a separate “subject”. It’s become that deeply woven into the system. But in fact, there was a time when “reading instruction” just meant teaching kids to decipher or decode written words. The term wasn’t also used as it is today, to mean teaching kids to understand the texts they read. That was taken care of through teaching them about history, literature, science, geography, and the arts.
We now use the word reading to cover two very different things: decoding and comprehension. And that has led to the kinds of problems we’ve been hearing about from teachers, trying, for example, to teach students the skill of making inferences without providing the foundation of knowledge that would enable them to actually do that.
A reading researcher named Alan Kamhi, has suggested that we redefine “reading” to just mean “decoding”. He points out that would free up teachers to spend a lot more time on those other subjects, the ones that build the knowledge that’s actually crucial to reading comprehension. But my guess is that such a radical change won’t happen anytime in the near future.
So, let’s look back to the past to see how these two fundamentally different concepts – decoding and comprehension – got so mushed together.
It seems to have started in the early decades of the 20th century, when the American high school population exploded. Before that, only the most academically inclined students went on to high school. Suddenly, high schools found they had a lot of students who, for various reasons, weren’t equipped to do the kind of work the curriculum expected.
Rather than trying to give all of those students a rigorous academic education, high schools – and eventually schools for younger students – developed different “tracks”. And to assign students to those tracks, schools started giving them reading comprehension tests, along with IQ tests. Unlike today’s reading tests, these tests weren’t aimed at measuring student’s progress. They just tried to assess what each student was supposedly capable of learning, and whether they should study Latin or something like carpentry.
At around the same time – in a shift that may or may not have been related to testing and tracking – reading textbooks in the lower grades started to focus on comprehension. At first, students were just supposed to answer questions about the passages they read in the textbooks. But soon, the authors of the reading textbooks came to feel that wasn’t enough. They developed student workbooks with passages that were unrelated to the passages in the textbooks themselves. The questions about the passages in the workbooks tried to break down the process of comprehension into different skills: finding the main idea, determining the sequence of events, drawing conclusions.
The new idea here was that you could teach those skills in the abstract – that if students practiced them on a workbook passage, they would get better at applying them to a textbook passage about something completely different. By the 1950s, that approach was deeply entrenched in the textbooks known as basal readers.
Many teachers didn’t like the basal readers. They felt the reading passages were insipid, and the questions too scripted. In the 1980s. That dissatisfaction contributed to the rise of an enormously popular movement known as Whole Language. Whole Language attracted attention primarily for its rejection of phonics instruction. The movement embraced the theory that most children pick up the basics of reading if they’re surrounded with good children’s literature – rather than the insipid snippets in the basal readers – and are allowed to follow their own interests. But Whole Language also rejected the comprehension activities in the basal readers. And the very idea of teaching reading comprehension at all.
Soon, growing evidence showed that most kids actually do need phonics instruction. And that evidence led to the demise of Whole Language. It was replaced around 2000 with something called Balanced Literacy, which is still the dominant approach. Balanced Literacy was supposed to include the best of both approaches: phonics instruction, and exposure to good children’s literature. But the leaders of Balanced Literacy came out of Whole Language, and they retained their skepticism about phonics. As revealed by Emily Hanford’s recent podcast series “Sold a Story”, Balanced Literacy hasn’t provided phonics instruction in a way that actually works for most students.
But something strange happened on the comprehension side of reading instruction. Even though Whole Language had rejected the idea of teaching reading comprehension as a set of skills, Balanced Literacy came to embrace it.
How did that happen?
Those teachers didn’t want to go back to the comprehension skills in the basal readers. And apparently, it didn’t occur to them that the reason kids were having trouble with comprehension was that they lacked the background knowledge assumed by nonfiction texts at higher grade levels.
From what I can tell, in the waning days of Whole Language, teachers started feeling like they needed to do something to help kids learn to read. Some had noticed that, even if kids learned to decode more or less on their own, they had trouble understanding what they were supposed to read, especially starting at around fourth grade, and especially if the text was nonfiction. But some of them came across recent research by cognitive psychologists on comprehension strategies – things like asking yourself questions about what you’re reading as you go along. Psychologists had determined that expert readers were likely to monitor their comprehension in this way, when they had trouble understanding a text.
At the time, psychologists weren’t at all sure you could actually teach these strategies to kids. But that didn’t stop teachers. Many embraced this new approach with enthusiasm – and they saw the strategies as very different from the comprehension skills the basal readers tried to teach. The strategies weren’t just different types of questions for students to answer. They were things you could teach kids to do consciously.
Psychologists did eventually start doing research on whether you could teach strategies. When a commission of experts called the National Reading Panel surveyed the research in 2000, they found evidence for teaching several comprehension strategies, but not for teaching skills. Over time, though, the distinction between skills and strategies got blurred, and teachers started using the terms interchangeably. Many went back to teaching the same skills that were in the basal readers.
Meanwhile, on the testing side, reading comprehension assessments got a lot more important. After Congress enacted the “No Child Left Behind” legislation in 2001, schools and teachers started being evaluated on the basis of their students’ scores on state reading tests. Those tests seemed to be assessing skills like finding the main idea and making inferences, so teachers began spending more time on those kinds of skills than ever.
Here’s what all this tells us: Reading comprehension tests weren’t designed to measure student’s academic progress – but that’s what they’re used for now. The Whole Language Movement, which gave birth to Balanced Literacy, rejected the idea of teaching comprehension skills – but that’s what Balanced Literacy teachers do now. And as we heard in Episode 2, there’s really no evidence that teaching skills or strategies year after year in the abstract, will improve comprehension. But that’s the assumption that a lot of our education system is based on.
In other words, we’ve ended up in a place that it’s not clear anyone really wanted to go.
I’m not blaming the people that created them, but in the creation of them, we’ve, we’ve we’ve produced something that’s really unfair for the student. It’s unfair for the teacher, and it’s unfair for the school system.
That’s Hugh Catts, the reading researcher at Florida State we heard from in Episode 2. He’s talking about the state-mandated standardized reading tests given towards the end of the school year, usually in grades 3 through 8.
Those tests are unfair for students, Hugh says, because they assume knowledge that some kids are able to pick up outside school, and others aren’t. It’s virtually impossible to test a skill like “making an inference” without also testing knowledge. If a student doesn’t happen to have enough relevant knowledge to understand the passage on the test, she never gets the chance to demonstrate her skills.
The tests are unfair for teachers because they don’t know what knowledge will be assumed by the test passages, so all they can really try to teach are test-taking and comprehension skills. And the tests are unfair for schools because they get rated on the basis of their students’ test scores – and those can be slow to change, even if a school is providing kids with the knowledge and vocabulary that will eventually enable them to do better on the tests.
That’s because, as Hugh discovered when he conducted his own study on reading comprehension several years ago, kids will learn what you actually teach them, including new vocabulary. But, it can take a long time before they acquire enough new vocabulary to enable them to understand passages on subjects that no one has taught them about.
How long? That’s been an open question. But now we have a couple of studies that shed some light on the answer.
A team led by a researcher named James S. Kim, at Harvard, created a curriculum that builds kids’ knowledge of science in the early elementary grades. The research team then tested the effects of the curriculum on comprehension. In first grade, kids studied how animals survive in their habitats. In second, they learned about how paleontologists study dinosaurs. Students got about 30 hours of these lessons each year. A control group of similar students got “business as usual”, the standard kind of curriculum.
Tests designed by the research team showed that by the end of second grade, students in the treatment group had clearly learned what they’d been taught, just like the kids in Hugh Catt’s study. Compared to the control group, they were much better able to understand a passage about paleontologists using fossil evidence. They were even able to apply that knowledge to understand a passage on a topic that was only somewhat related to what they had learned. It was about archaeologists studying human fossils to research life in ancient Pompeii. But when the kids were asked to read a passage about genealogists studying people’s ancestors, they did no better than the control group.
Then, in third grade, the children learned about how astronauts keep their various body systems healthy in space. At the end of the year, the kids not only did better on passages related to what they learned, but also on one that contained none of the words they’d actually been taught – a passage on how the anatomy of a skyscraper is like a human body system. That study suggests it can take three years for a knowledge-building curriculum to produce the kinds of gains in comprehension that show up on a standardized test. And it suggests that those gains last. At the end of fourth grade, students still did better than the control group, even though they hadn’t gotten the science curriculum that year.
A second recent study led by David Grismer at the University of Virginia, used actual state reading tests as a measure – not tests designed by the researchers themselves. Measures that aren’t designed by researchers are generally considered more reliable.
That study found that children who got a comprehensive knowledge-building curriculum starting in kindergarten, did a lot better on the state tests in third grade, when compared to a control group that got the typical elementary curriculum. Those gains persisted in fourth, fifth, and sixth grade. In fact, the gains were so large that if they were translated to American students as a whole, the US would move up to a position among the top five countries on an international reading test given to fourth graders. Currently, we rank 15th out of the 58 countries that participate. The study also found that children from lower-income families got the biggest boost, scoring as well as their peers from higher income families. In other words, the curriculum seemed to eliminate what is sometimes called “the achievement gap”.
Of course, that’s only one study – and there were only 16 kids from lower income families in the sample, although the study as a whole involved more than 500 students. Some argue that we need more studies like that before we can say with confidence that a knowledge-building curriculum can boost reading comprehension. But it took 14 years to get the results from that one study, and it cost a huge amount of money. As far as I know, there are no more such long-term studies in the pipeline – and even if there are, we’ll probably have to wait years to get the results. Meanwhile, untold numbers of children might continue to get the kind of literacy curriculum that doesn’t seem to work. Given all the other evidence we have on knowledge and its relationship to reading comprehension, that doesn’t seem fair.
It’s not just state tests that are unreliable measures of progress. Schools give internal standardized reading comprehension tests throughout the year to monitor progress and assign students to their individual reading levels. And those tests can be unreliable too. Here’s what Abby Boruff, the first grade teacher in Des Moines, said in Episode Three about giving reading level tests:
The level C expository text was the book called “Shopping”, and they go to the grocery store, and half of the food words in there – yeah, if kids didn’t have those in their, in their vocabulary, they weren’t they weren’t going to read that book.
Hugh Catts points to a study in which researchers took a sample of about 1000 students and gave them all four different reading comprehension tests. The tests didn’t agree with each other.
And they identified the lowest 10th percentile of children on one of the measures and looked to see what percentage on average, fell – of those kids – fell within the lower 10 percentile on the other measures, and it was only 43%. Less than half of the kids who were identified as poor readers on one measure, turned out to be a poor reader on another measure. That has implications for using these measures in a diagnostic form.
The tests were equally unreliable in identifying kids as good readers. Hugh says he saw the same thing when he was conducting his own study years ago.
And it wasn’t uncommon to find a child who got all the questions right on one passage, and less than half of them right on another passage. Sure, that child had some average particular level of comprehension, but it was quite variable, depending upon what they were reading.
Hugh says no single test can be an accurate measure of a student’s comprehension ability. It’s not just a question of background knowledge – maybe a kid was just zoning out or having a bad day.
What can we do about this problem? State reading tests probably aren’t going away anytime soon. But there are ways to make them fairer. Hugh Catts points to Louisiana, which is experimenting with a reading test where the passages are related to topics covered in the English Language Arts, and social studies curriculum. But Louisiana is unique in that most schools are using its state-created curriculum. In other states, schools are using all sorts of different curricula. So those states can’t ground their reading tests in any particular body of knowledge.
But all states have academic content standards, laying out what students are supposed to learn about social studies and science at different grade levels. The passages on their reading tests could be grounded in those standards. Hugh says that kind of test would make more sense.
And what I would suggest that we move toward, is something like that at a state level, to where we actually test kids knowledge of a particular subject matter, a science test, or a humanities or, or a social studies tests that would provide teachers with a body of knowledge to teach, it would allow us to prepare kids to do better in their reading and writing. And we can actually make a difference.
And what about those internal tests schools give during the school year? That’s actually an easier problem to solve, at least if the school is using a knowledge-building curriculum. Those curricula come with their own assessments that are grounded in the content that’s been taught. And they don’t just ask kids to recite facts. They also ask them to do things like make inferences about those facts.
In Episodes 3 and 4, we heard from Deloris Fowler, the instructional coach in Tennessee, whose district uses a knowledge-building curriculum called CKLA. Deloris says the teachers sometimes do feel pressure to do test prep. But she tells them the tests that come with CKLA actually do a good job preparing kids for standardized tests,
At the end of each of the units in CKLA there is a unit assessment, and it’s, they have to read two passages, and there are questions about the passages and those passages are aligned with what the unit is. So if they’ve read about vikings, then the two passages or something about Vikings. But the actual question types are exactly the same types that you would find on the state tests. They’re multiple choice, some of them are multiple select, some of them are short answer. So we’re telling teachers like, if you’re doing those assessments, and you’re teaching your test-taking skills through those assessments, then your students should be able to do that on any test. I mean, knowing to go back and look in the text to you know, look for the answer that kind of thing. So that that’s kind of the way we’ve been approaching it.
Knowledge-building curricula also teach comprehension skills and strategies in a way that works by putting the content in the foreground.
You know, there’s no knowledge-building curriculum out there that doesn’t also teach skills and strategies. Of course, it does, it has to. But again, it’s like what comes first. And the strategies-first approach doesn’t necessarily transfer.
That’s Brent Conway, the assistant superintendent in Pentucket, Massachusetts, who we heard from in the last episode.
A knowledge-building curriculum will guide teachers to ask kids to do things like activate the knowledge they already have about a topic, or make inferences about a text – even if that’s not labeled “strategy instruction”. And because all students will have had a chance to acquire the knowledge to be activated, or the information that will enable them to make an inference, it’s a fairer and more accurate assessment of their ability to do those things.
Building knowledge is a gradual, cumulative process that goes on for years. In fact, it never really ends. That explains why it can take a while to see results on standardized tests. But it also means that if schools keep using a knowledge-building curriculum, they’ll begin to see better results as time goes on, not just on tests, but in the classroom. Here’s what Brent Conway says about that.
A knowledge-building curriculum, you’re not going to necessarily see the benefits of it in one year. And in fact, you may not see the benefits of what you’re doing. But when students get to third grade, after they leave first, second, it’s first grade, second grade, third grade, all of a sudden, in third grade, they’re reading about something and they remember you discussing it back then, right? And they’re building this volume, time after time after time, where things begin to make connections.
I’ve heard from lots of teachers that this happens, and I’ve seen it myself. When I was doing the research for my book, I followed a class of second graders who had been getting the CKLA curriculum since kindergarten. They were constantly making connections between what they were learning in second grade and what they had studied in previous years, sometimes prodded by the teacher but sometimes on their own. And the failure to build knowledge also has a cumulative effect. That’s why it’s so important to begin building knowledge and sophisticated vocabulary as early as possible. As time goes by, two kinds of gaps – knowledge gaps, if you will – get increasingly larger.
One is the gap between good readers and poor readers. The good readers start out with bigger vocabularies, which means they get to read more sophisticated books. And they’re able to retain new vocabulary from those books – because, remember, knowledge is like Velcro: it sticks best to other related knowledge. That additional vocabulary, in turn, enables them to read yet more sophisticated books. So, those kids can zoom ahead.
The other kids, meanwhile – the ones who start out with less of that other half of the Velcro – are limited to books with simpler vocabulary. They’re also less likely to be able to retain whatever new vocabulary might be in those books, because they may not have the other half of that Velcro for new words to stick to. So they can fall farther and farther behind their more advantaged peers. By the time students get to high school, or even middle school, the gap between these two groups can be very wide, and very hard to narrow.
The second kind of knowledge gap is also most likely to show up at higher grade levels. That’s the gap between what the curriculum assumes a student knows, and what that student actually does know. Many students get to middle or even high school, never having had any systematic exposure to history, science or anything but reading and math. That can leave them, through no fault of their own, with large gaps in their background knowledge that make it difficult or impossible for them to understand what they’re expected to read.
All of this explains why improving phonics instruction in the early grades, without also building knowledge, produces only short term effects on reading test scores. Better phonics will yield a bump in the elementary grades, because scores at those levels are more dependent on kids ability to decode words. But, by middle school, the test passages assume a lot more sophisticated knowledge and vocabulary. That means scores become more dependent on comprehension. And at that point, the bump in scores from better phonics instruction fades out.
It’s not just about tests: students who lack background knowledge often flounder in class.
So you had these students who maybe had some mild disabilities in some areas, but basically were pretty good, knew how to read and knew how to do all the things they were expected to do in elementary school. But because of those gaps that had not been filled in early on, they hit the wall in middle school, and they didn’t know what to do.
That’s Roberta Jacobson, a former special education teacher in Texas who now works for a publisher of knowledge-building curriculum.
The situation gets even worse as kids are passed along to higher grade levels. The problem is most obvious at high-poverty high schools. Teachers at those schools have told me they’ve had students at all levels of ability. But they’ve also said that many students are unaware of some basic things about the world, like the difference between a city and a state, or a country and a continent. They may not know anything much about historical events, or have a sense of historical chronology. And yet, we give them a world history textbook and expect them to understand it. That’s likely to be a frustrating experience, both for those students and for their teachers.
It’s not impossible to make up for these gaps at higher grade levels. But it’s a lot easier to just start building kids’ knowledge early. So this is a problem that’s much bigger than elementary school, and much bigger than just “reading”.
Let’s take American history. Scores on national tests in that subject have been declining for years, hitting a new low in 2022: only 14% of eighth graders scored proficient or above, and 40% scored below the “basic” level. Scores in civics are only slightly better. Maybe you assume students learn more about the subjects after eighth grade? Not necessarily. One survey, for example, found that 11% of US adults haven’t heard of, or aren’t sure if they’ve heard of the Holocaust. For millennials, the figure is 22%.
You might think middle and high schools just need to spend more time teaching subjects like history. But that might not be enough. If students have never been exposed to a subject before, and the curriculum assumes they have, the information may go right past them. That’s because it doesn’t have anything to stick to. Remember: knowledge is like Velcro – it sticks best to other related knowledge. So we need to start building knowledge of subjects like history earlier in elementary school to build that other half of the Velcro.
Another problem is the way history, civics and other subjects get taught. The American education system hasn’t just overlooked the importance of knowledge in reading comprehension instruction. It’s also overlooked or downplayed the importance of knowledge in general. Often, teacher training and curriculum materials focus on things like developing students’ critical thinking skills over ensuring that they’ve actually understood and retained information that’s been covered. A graphic widely used in teacher education shows a pyramid of learning objectives with “knowledge” and “comprehension” at the bottom, and “synthesis” and “evaluation” at the top. Teachers are often encouraged to just skip those “lower order” processes and focus on the “higher” ones.
But that’s not what that graphic – which is called Bloom’s Taxonomy – actually means. Instead, it means that you have to go through knowledge and comprehension to get to those higher order processes. That’s what cognitive science has found as well. In other words, it’s impossible to teach critical thinking in the abstract, just as it’s impossible to teach comprehension in the abstract. Both of those things depend on having knowledge of the topic you’re trying to think critically about or comprehend.
So, fixing what’s wrong with American education isn’t just a matter of adding more content to the curriculum. That’s hugely important, especially at the elementary level. But it’s also crucial for teachers to know how to teach that content in a way that works for all students: when to explain things directly, when to ask questions that assess whether kids have understood what’s been taught, when to ask them to analyze it or come up with their own ideas. A good curriculum can help guide teachers in doing that complex and extremely challenging work.
American education has a number of serious problems – and our failure to start building kids’ knowledge early is a fundamental one. But I want to make it clear that teachers aren’t to blame for these problems. In fact, no individual is to blame. The problems are systemic, and teachers are the victims of a flawed system as much as students are.
I’m convinced the system can be fixed. But it will take time, patience, and an understanding of root causes that are sometimes hard to discern.
Especially in the wake of the pandemic, when schools are under so much pressure to address the myriad problems it has led to, it can be hard to decide what to focus on.
Some argue that it’s more important to address students’ mental health and emotional well being, then what is often called “COVID-related learning loss”. But when I raised that issue with LaTonya Gaffney, the superintendent of a high-poverty Texas district that suffered from COVID more than most, and that adopted a knowledge-building curriculum during the pandemic, she was firm. You have to do it all, she told me:
Well, you can’t choose. And I think that’s a missed opportunity sometimes when you’re serving in districts that have some challenges – many times you want to focus on one piece, you want to focus on our food insecurity, or you want to focus on the fact that we don’t have health care as we’re educating the whole child all those different things.
But the one thing that I don’t think we’ve done a great job in Texas or across the country, is focusing on the fact that we don’t know how to teach kids how to read. And so we need to put just as much emphasis on getting this right as we do on getting other things right, the other ones you can get, or you can feel good. In the moment, when you do a food drive, you do a toy drive, you do a clothing drive, you feel good, because you see people happy – you see them smiling.
But there’s nothing that would wipe my smile away quicker than to know that kids are graduating and cannot read.
At the same time LaTonya cautions that we won’t be able to change the system just by mandating that schools use a knowledge-building elementary curriculum – something the federal government doesn’t have the power to do, and something no state government has chosen to do. It’s not even enough to make one freely available online, as Texas has.
Teachers and administrators need, first of all, to understand why this change is necessary, so that they’ll buy into it. And once they have a curriculum, they’ll need support in figuring out how to deliver it well. Without the “why” and the “how” things may not go well.
I just know, based on my own homework based on the work that we’ve done, that this is the right direction for the country. And I know that we’ve gotten it wrong so much, that we really got to get it right. And so we can’t mandate it – we can’t just say, “Okay, it’s free: go out” because people will say, “Oh, I tried it, it didn’t work”.
It is important to get this right. It’s important for the untold numbers of students whose potential remains to be unlocked – students who might otherwise go through school and life, feeling like failures, when in fact, it’s the system that has failed them. It’s important for society, which will otherwise be deprived of those students’ potential. And it’s important for democracy, which depends on a citizenry that can understand the world well enough to make informed decisions.
Here again, is Spring Cook, the former Texas teacher we heard from in Episode 1, talking about the changes she saw after her elementary school adopted a knowledge-building curriculum. She said that teachers, students and parents were excited about what was going on, that kids were having fun learning. And she also said this:
Because it is a matter of equity, it’s a matter of democracy, and when we’re able to give students those skills and that knowledge at an early age, then think what a better society will have.
I hope you agree. Thanks for listening.
For more information about this episode, visit the Knowledge Matters website linked in the show notes. This podcast is produced by the Knowledge Matters Campaign. You can learn more about their work at knowledgematterscampaign.org and follow them on Twitter, Instagram and Facebook. Search the #knowledgematters hashtag and join this important conversation. If you’d like to get in touch with me personally, you can contact me through my website, nataliewexler.com.