Some Thoughts on the 2019 NAEP Reading Decline

2019 NAEP scores have been released, and the results in reading… aren’t good. As the New York Times reports:

Two out of three children did not meet the standards for reading proficiency set by the National Assessment of Educational Progress, a test administered by the National Center for Education Statistics, the research arm of the Education Department.

The dismal results reflected the performance of about 600,000 students in reading and math, whose scores made up what is called the “nation’s report card.” The average eighth-grade reading score declined in more than half of the states compared with 2017, the last time the test was given. The average score in fourth-grade reading declined in 17 states. Math scores remained relatively flat in most states.

Only 35 percent of fourth graders were at [or above] proficient in reading in 2019, down from 37 percent in 2017; 34 percent of eighth graders were at [or above] proficient in reading, down from 36 percent. Overall student progress in reading has stalled in the last decade, with the highest performers stagnating and the lowest-achieving students falling further behind.

Despite attempts to pin the blame on poverty and other social ills, the fact that math scores have not declined anywhere near as much as reading scores suggests that the problem lies in the schools—in considerable part, at least. Intuitively, at least, it does not make sense to suggests that socio-economic factors could suddenly have such an outsized impact on one area of the curriculum while having almost no effect on another. In fact, math scores actually improved in some states.

Most glaringly absent was any mention of the fact that 1) many schools are failing to teach phonics, either effectively or at all; and that 2) once students can decode well, reading comprehension becomes a proxy for knowledge. Thus, if schools are not making an explicit effort to systematically develop students’ general knowledge, no one should be surprised when their reading skills suffer. (Notably absent as well was a serious discussion of the effects of technology on students’ attention spans and ability to sustain focus on complex material, obviously a huge factor as well, but that’s not my main concern here.)

Part of the problem is that “reading” is treated as a single entity rather than the result of many separate components all working together. Failure (or perhaps willful refusal) to understand what these components are and how they interact makes effective remedies impossible.

According to the simple view of reading (Gough and Tumner, 1986)—a model that has been more or less universally accepted by researchers for several decades—the ability to read depends on two main factors: 1) decoding ability, and 2) comprehension of spoken language (which comprises myriad factors, including syntax, vocabulary, background knowledge, etc.).

Weakness in one or both of these areas leads to a reading problem.

So, for example, a student may be able to understand sophisticated oral language but have weak decoding skills; conversely, a student may have strong decoding skills but weak vocabulary and background knowledge.

More commonly, the two deficits present in tandem—this is what researchers often refer to as a “garden-variety” reading problem.

Obviously, there are many complexities within those two factors, but at the most basic level, these are the components of reading.

So if students are struggling to read on a large scale, then either one or (much more probably) both of those factors are the culprit, and they need to be taken as the starting points for any serious attempt at remediation. While other, external issues obviously play a role in what students know when they enter the classroom, as well as in their attitude toward school and reading, ignoring the factors laid out in the simple view—factors that schools can control— is an act of stunning neglect.

For most students, the two abilities do not fully converge until around eighth grade—before that, children can generally understand spoken language more sophisticated than what they can read. But if the ability to translate what one comprehends by ear into an understanding of written marks on the page is not properly developed, or if students’ general language skills and knowledge base remain weak, then a gap between the two sides will persist, and students will be unprepared to read at a more advanced level in high school.

The fact that such a steep drop-off occurs in eighth-grade reading scores, precisely at the point when all the various factors involved in proficient reading typically come together, very strongly suggests that the two sides of the equation are not being dealt with effectively throughout elementary and middle school, and that, moreover, curricular deficiencies become more pronounced and cumulative as students advance through the middle grades.

Until the middle grades, students who have not learned to decode well but who have memorized lots of sight words can often disguise their difficulties because the texts they are expected to read remains relatively simple; beginning around seventh grade, though, the volume of new words becomes overwhelming, and the deficits impossible to hide. This alone goes a long way to potentially explain why the decline in eighth-grade reading scores was so more widespread than the decline in fourth-grade scores.

For students who do become proficient decoders, background knowledge becomes the dominant factor in comprehension. For that group, the obvious implication is that reading deficits become more pronounced between fourth and eighth grades because schools are increasingly failing to furnish students with the knowledge they need to understand a wide range of topics. One possible explanation for the 2017-2019 decline is that more experienced teachers are retiring or are being effectively pushed out of the profession in many areas, and are getting replaced by younger colleagues more likely to drink the edu-koolaid about skills trumping knowledge.

Alternately, the students in eighth grade now might have been exposed to a more skills-oriented, test-centric curriculum all along, and so have actually gained less subject-based knowledge as compared to classes just a couple of years ahead of them. Indeed, this group of students would have been exposed to Common Core for essentially their entire school careers.

No doubt, the test-and-punish regime, and its accompanying obsessive focus on mastering formal skills (finding the point, inferencing, etc.) at the expense of actually learning about things like history and science, has ironically played a key role in exacerbating the very problem it was designed to solve. But the persistent disparagement of learning “mere, rote facts,” and the belief that critical thinking skills can be developed in the absence of a substantive body of knowledge, should not be underestimated either. Add in the fact that many teachers themselves have a shaky knowledge base in their subjects, and you have a recipe for a disaster.

Does the article discuss any of this? Of course not. And will the powers that be take it into account? Almost certainly not as well.

Just as telling as the article itself is the comments section, however; as a snapshot of educated Americans’ attitudes toward Why Johnny Still Can’t Read, it’s remarkably illuminating.

As I read through the comments, a few major themes stuck out for me. They shed light on some common misconceptions about reading and help explain why there isn’t more pressure on schools to adopt programs/curriculums that might make a real difference.

First, and unsurprisingly), many if not most commenters seemed to be operating under the assumption that becoming a good reader is a largely an emotional process—that is, if students could only be taught in such a way that would result in their loving to read, then the problem would be largely resolved. (Never mind it’s hard to love reading when you can’t figure out the words on the page.)

As one astute commenter (teacher) pointed out, however, no one wrings their hands over whether students are sufficiently enamored of equations! (Or, I would add to that, how much time their parents spend with them on numeracy-building activities.) But because proficient reading is viewed as a fuzzy, romantic process rather than the culmination of many discrete skills, people have difficulty conceiving of it in intellectual terms.

Second, the belief that students would naturally become readers if given books on topics that interested them. Now, obviously children are more likely to enjoy books about topics in which they have an established interest; and yes, some children will use such books as steppingstones for new interests. However, on a broad scale this solution is exactly backwards. Assuming that children can decode competently, they need to be deliberately exposed to—and explicitly taught about— new topics so that they can acquire a broader range of knowledge and be able to understand texts on a wider range of subjects. If children are allowed to stay only in their comfort zones, that’s a lot less likely to happen.

When you think about it, the narrative is actually quite contradictory here: on one hand, people burble on endlessly about children’s natural curiosity, but on the other hand insist that reading problems can be solved by NOT introducing children to new things. But if teachers are averse to providing students with sufficient knowledge to understand texts about new topics, then of course students won’t enjoy them.

I mean, seriously, duh.

I did find some general references to the narrowing effects of standardized testing on curriculum, but these were generally cast in terms of destroying students’ interest in reading rather than on the fact that it’s hard to understand texts on topics you don’t know anything about.

Third, I couldn’t help but be struck by many commenters’ sentiment that reading can’t really be taught—or at least that schools can’t really be expected to teach it—because students’ fates are entirely sealed by the home environments. (As one commenter put it, she got the “reading gene” from her parents.) If parents don’t read to their children or provide a literacy-rich home environment… Well, schools can’t really be expected to do anything about that now, can they? 

As Betsy Devos is quoted as saying, “Think about the mom or dad who cannot read, and so does not read to their own children at bedtime,” Ms. DeVos said as she released the scores. “Think about what that portends for their lifelong learning.” The amusing part is how many commenters—people no doubt in horror of Devos’s agenda—agree.

Let me point out that we are talking about middle-schoolers here. How many parents still read to their 11- to 14-year-olds at bedtime? (And how many kids that age still want their parents to read to them?) Furthermore, the decline between fourth and eighth grades—just when students outgrow the “being read to” years—implies that there is something else going on.

One plausible explanation is that students—particularly disadvantaged ones—are increasingly either plopped in front of computers and left to “self-direct” their learning, or put in groups where they largely spend their time talking to their classmates. What they are not hearing enough of is structured adult speech—the kind of speech that comes far closer to written English than anything heard in casual conversation. The problem isn’t (just) that parents aren’t talking to children; it’s that teachers aren’t talking to them, or at least not in ways that can be translated to the page. But the taboo against direct, teacher-centered instruction is so extreme as to preclude consideration of that factor. If anything, the decline in scores will be taken as a signal to double down and insist that the problems stem from schools being insufficiently student-centered.

The “environment-is-destiny” argument also stands out to me because it comes from both the left and the right: the former asserts that poverty is the sole cause of low scores, and that no progress can possibly be made without economic gains; the latter that “cultural factors,” poor work ethic, familial dysfunction, etc. have rendered a subset of children effectively uneducable, and that these pathologies are so entrenched that any attempt to counteract them is bound to fail.

Now obviously, both of these views draw on socio-economic realities to some extent, but they have also been hashed to death elsewhere. What concerns me, rather, is how they converge to produce a kind of fatalism; if schools are so thoroughly unable to carry out a central part of their mission because of external factors, then one is left wondering what exactly they are for. Social-emotional learning perhaps?

But perhaps that is the point.

Essentially, schools are damned if they do, damned if they don’t.

If they do attempt to provide students with the broad knowledge necessary to become competent readers across a variety of fields, then they are derided for teaching disconnected, rote, pointless facts, unnecessary for twenty-first century learners who can look everything up on Wikipedia.

If they don’t provide students with the broad general knowledge they need to become competent readers across a variety of fields, then test scores sink, fueling the narrative that schools are failing and paving the way for even further “reforms” and destabilization of the system.

Anyone want to offer a way out?