My take on Dr. Paul Thomas’s idea that the decline in NAEP reading scores shows it is time for a new story by Dr. Sam Bommarito
The Science of Reading star has been rising for quite some time. See this chart from the Jan. 21st edition of Education Week LINK.
A quick glance at the chart shows that since 2019, sixteen states have mandated SOR practices. By 2021, another 14 states joined, bringing the number of states with three to five-year commitments to SOR practices up to 30. Given the promises made by SOR proponents and the number of districts adopting SOR practices in the past few years, one would expect substantial gains in reading scores nationally. Instead, NAEP scores have DROPPED to below pre-pandemic levels. Remember that those scores provide a long-term longitudinal look at national reading scores. Paul Thomas, a major critic of SOR, has written some posts on this very topic. Let’s look at some highlights from his recent posts.
Paul’s post provides a lot of well-researched information. This post is a must-read, and I suggest that my readers take the time to read the full post. I want to call attention to several important points he makes.
- “reading achievement on NAEP was flat during the balanced literacy era and now has dropped steadily during the SOR era”
- “NAEP “basic” is approximately grade level (although even that claim is problematic since no standard exists in the US for “proficient” or “grade level”), and “proficient” on NAEP is high:”
This point is especially important in light of a point he made in an earlier post LINK.
“As I have noted, NAEP achievement levels are confusing since “proficient” is well above grade level and “basic” tends to correlate with most state metrics for “proficient” (see here for a full explanation and state/NAEP correlations).”
MY TAKE- The bottom line is that the claim often made in the media that about 2/3 of our students are reading below grade level is completely inaccurate and greatly exaggerates the actual problem. Paul labels this as one of the “Big Lies” being told by SOR advocates.
- But a key element of the SOR story is often overlooked: “One of the excuses educators have long offered to explain America’s poor reading performance is poverty.”
- “In other words, the SOR story argues that the US has a reading crisis that is entirely the result of in-school policies and practices, that SOR-based reading instruction guarantees 95%+ of students will achieve reading proficiency.”
MY TAKE- The relationship between poverty and low reading achievement is well established. Yet SOR advocates downplay/ignore the impact of poverty on reading achievement. In addition, the claim that SOR-based instruction guarantees 95% of students will achieve reading achievement simply is not borne out by the longitudinal NAEP data In addition, England’s experience with synthetic phonics for over a decade has demonstrated that synthetic phonics is not the cure-all some make it out to be LINK.
I agree with Paul that it is time for a new story. It is time for the WHOLE STORY. It is time to look at ALL the research, not just that research that supports one side or the other’s point of view. My own opinion is that if we stay the course with the brand of SOR currently being mandated, the result will be yet another very expensive swing of the pendulum LINK.
I’ll say it again- it is time to stop taking sides. It is time to end the “Big Lies”. It is time to talk about what we can all agree on so we can find a path to improving the literacy performance of all students (see my blog last week, LINK). Dare to dream!
Happy Reading and Writing,
Dr. Sam Bommarito (aka, the guy in the center taking flak from all sides)
Copyright 2025 by Dr. Sam Bommarito. Views/interpretations expressed here are solely this author’s views and do not necessarily reflect the views of any other person or organization.
PS If you found the blog through Facebook or Twitter, please consider following it to ensure you won’t miss future posts. Use the “follow” entry on the sidebar of the blog.
“ I agree with Paul that it is time for a new story. It is time for the WHOLE STORY. It is time to look at ALL the research…”
I think you and Paul are right, Sam. But the above 3 sentences stood out to me because in the late 80’s or early 90’s Brian Cambourne did just that – and the title of his book that set out his theory and research is entitled The Whole Story. I believe Brian’s use of the word “whole” captured several ways of looking at literacy. The idea that the brain operates most effectively given “holistic” opportunities, the idea that “whole language” was an appropriate descriptor of the body of research, and the fact (in my opinion) that Brian presented a “whole story” that brought together not only his research, but also the work of the likes of Don Holdaway, Frank Smith, Ken Goodman, and several others was a major accomplishment. The book is anchored by a natural learning theory Brian called The Conditions of Learning along with a corollary called The Principles of Engagement. Many educators such as yours truly, Regie Routman, and many many others successfully used Brian’s learning conditions to effectively teach children regardless of the economic or cultural environment — and the result was that mutually supportive communities of learners were created and whole classes “joined the literacy club.”
Thanks. BTW I did write about one of Brians workshops- have a look https://doctorsam7.blog/2021/10/23/cambournes-model-of-learning-gives-students-what-they-deserve-informed-empowered-teachers-who-provide-them-with-literacy-instruction-that-fits-their-needs-by-dr-sam-bommarito/
Yes, that was a great piece you did re: Cambourne’s workshop and his conditions of learning theory.
The decline in NAEP scores would be magnified if states weren’t permitted to retain students in oder to inflate scores. Statistics show that kids retained are more likely to drop out of high school. If current policies are decreasing scores and increasing the drop out rate, one must ask who is benefitting from continuing such policies?
Paul makes the point that retaining kids gives a skewed picture of what is going on. And kids are being retained at a very high rate in places like Mississippi & Florida. He asked the question if SOR is working, why is there a need to retain this many kids? BTW thanks for your comment.
I have not ever understood why there has to be one or the other mindset in teaching, particularly reading. I taught elementary students for 41 years, 27 of those as their direct classroom teacher, moving among grades 1, 2, and 3. I used whatever worked with particular children, as I learned after a few year, that one method/strategy/mode of teaching does not work for all children. Fortunately, the kids taught me that! Also, I was fortunate in doing most of my teaching in an era where we still had freedom to teach as we thought best rather than have a curriculum strictly determine what we did. I was appalled in my later career when distrcit officials wanted all classrooms at a certain grade level across the district to be on the same page everyday -reasoning that, if a child moved schools, they would suffer no “loss”. I could not agree to do that; I teach children, not curriculum, and no curriculum has ever taught a child to read. Anyway, I know I’m in agreement with you that teachers should have the freedom to choose among all strategies, methods, and resources to teach the children in front of them. No curriculum can adapt to meet the needs of all the kids, but an informed, astute teacher can certainly do so. Thanks for continuing to bring these things to educators’ attention and give them research, wisdom, and determination to do what’s best for learners!
Glad to hear you agree. An important * is that teachers should have the freedom to choose among strategies, methods and resources within the confines of the districts guidelines.
It’s time for a complete re-set on how we think about literacy teaching and learning. I propose a new paradigm: Enlightened Literacy Beyond the Reading Wars: Providing Hope for Democracy, Global Citizenship, and the Human Condition. My book will be out within the next couple of months — describing my decades-long journey as a teacher, literacy specialist, literacy education professor and researcher. It’s time to end the fixation on phonics.
Sounds very interesting. Please e-mail at bommaritosam@yahoo.com if you would be interested in doing an interview around tha book.
My thoughts about recent NAEP results: https://www.youtube.com/watch?v=cwAvZNus9N0
The education establishment public closed schools to in person learning during COVID.
Parents and education advocates said this would cause learning loss.
The education establishment denied this.
Now national test scores are way down.
But the education establishment denies those tests are even valid.
Okay, fine. It’s clear that accountabiltiy isn’t possible.
So I send my kid to private school and get private tutoring.
But if you’re worried about inequality guess what just happened?
It got worse.
Congratulations.
The folks in the “educational establishment” I am following aren’t saying the tests are invalid. They say that most of the tests used in the “miracle gains” reports do not measure comprehension the way state tests do. So, you get great gains in what those tests measure- DECODING. In addition, some studies have shown that programs focused on decoding alone are implemented in ways where teachers learn phonics, but the kids don’t learn to use it in real reading. In addition, those tests do not improve comprehension scores. Pouring money into programs that raise decoding scores but not comprehension scores doesn’t make sense. There are programs that teach decoding in a way where the students learn the phonics knowledge AND how and when to use it. Those programs do raise BOTH reading and comprehension scores. Unfortunately, most of those programs are banned under the current laws being adopted.
Your reply changes the subject rather than addressing the central issue.
No serious Science of Reading advocate argues that decoding, by itself, is the goal of reading instruction. The question is whether explicit, systematic phonics instruction is the most effective way to build accurate word recognition so that students can ultimately comprehend what they read.
Instead of engaging that question, your response moves through several different claims:
* that some tests primarily measure decoding,
* that some implementations teach phonics poorly,
* that decoding gains do not necessarily produce comprehension gains,
* that some other unnamed programs improve both decoding and comprehension,
* and that those programs are “banned.”
Even if each of those claims were true, none answers the original question: compared with alternative approaches, what evidence shows that less explicit or less systematic phonics instruction produces better reading outcomes?
From a logical standpoint, I think your argument has several problems.
*Red herring. The discussion begins with whether explicit phonics instruction is effective. Your response shifts to implementation problems, test design, and legislation. Those may be worthwhile discussions, but they don’t answer the original claim.
*Straw man. Your reply treats the issue as though advocates are proposing “decoding alone.” That isn’t the position of the mainstream Science of Reading. Reading comprehension has always been understood to depend on language comprehension, vocabulary, background knowledge, syntax, and knowledge building in addition to decoding. Critiquing “decoding alone” doesn’t address what most proponents actually advocate.
*Hasty generalization. You cite “some studies” showing poor implementation and then suggest that spending money on explicit phonics programs doesn’t make sense. Poor implementation of some programs is not evidence that the instructional principle itself is ineffective.
*Unsupported assertion. You state that there are programs that successfully teach both decoding and comprehension and that these are banned under current laws. That’s a significant claim, but you don’t identify the programs or provide evidence that the laws actually prohibit them.
Ultimately, I think we’re still avoiding the central empirical question. If explicit, systematic phonics is inferior to the approaches you favor, the strongest way to demonstrate that is with comparative evidence showing those approaches consistently produce better reading outcomes—not by changing the subject to labels, implementation failures, or policy debates.
You missed the key point of my answer. For SOR folks, especially SOR folks on social media Systematic phonics is code for Synthentic Phonics. If you disagree find examples of SOR folks teaching any of the other 5 forms of phonics listed by the NRP. Synthetic phonics does not produce gains in comprehension. See the NRP and see my blog today. SOR advocates are not using the the most effective way to teach phonics. The most effective way is to teach comprehension strategies and decoding strategies concurrently. One way of doing that is to use Rasinski’s concept of building a bridge between phonics and decoding. See my blogs today and tomorrow for more details.
“You missed the key point of my answer. For SOR folks, especially SOR folks on social media Systematic phonics is code for Synthentic Phonics. If you disagree find examples of SOR folks teaching any of the other 5 forms of phonics listed by the NRP. Synthetic phonics does not produce gains in comprehension. See the NRP and see my blog today. SOR advocates are not using the the most effective way to teach phonics. The most effective way is to teach comprehension strategies and decoding strategies concurrently. One way of doing that is to use Rasinski’s concept of building a bridge between phonics and decoding. See my blogs today and tomorrow for more details.”
Your position has narrowed. It started as “decoding alone.” Now it’s “synthetic phonics specifically.” That’s progress. It concedes that systematic, explicit phonics is not the problem. The dispute is over which kind.
On that narrower claim, the National Reading Panel doesn’t help you. The NRP compared the systematic approaches it identified: synthetic, analytic, analogy, embedded, and the others. It found that all of them outperformed non-systematic or no-phonics instruction. It also found they did not differ significantly from one another in effectiveness. If synthetic phonics were clearly inferior, that is where it would have shown up. It didn’t.
The NRP also found that systematic phonics improved comprehension, with the strongest effects in kindergarten and first grade. That is where the instruction is aimed. So “synthetic phonics does not produce gains in comprehension” needs a citation to a specific table or finding. “See the NRP” is not enough when the NRP’s own summary points the other way.
Second, teaching decoding and comprehension concurrently is not an alternative to the Science of Reading. It is the Science of Reading. Scarborough’s Rope and the Simple View of Reading both treat word recognition and language comprehension as parallel strands that develop together. No mainstream SOR curriculum postpones vocabulary, background knowledge, or read-alouds until decoding is mastered.
Where we actually disagree is what “decoding strategies” means. If it means teaching children to guess unfamiliar words from pictures, context, or the first letter, that is three-cueing. That practice is what the current laws restrict, and the evidence against it is substantial. If it means something else, name the program and the study.
On Rasinski: as I understand his work, the bridge runs from word recognition to comprehension, and fluency is the bridge. That depends on accurate, automatic decoding. It is an argument for solid phonics, not against it.
The question from my last comment still stands. Which approach, compared head-to-head with explicit systematic phonics, produces better reading outcomes, and what’s the evidence?
You said “The question from my last comment still stands. Which approach, compared head-to-head with explicit systematic phonics, produces better reading outcomes, and what’s the evidence?” That is a valid but misleading question. I’ll start by saying I’ve always taken the position that students need systematic phonics intruction. The question is how much and what kind. The answer to how much hasn’t been established to my knowledge. The question of what kind is the key question. My point is that ALL FIVE FORMS OF THE PHONICS LISTED IN THE NRP CAN BE TAUGHT SYSTEMATICALLY AND THAT INCLUDES ANALYTIC PHONICS. So all five forms should be availalbe to students. But under the curent laws they are not.
Where did the myth that the three cueing system was a failed practice come from. Read this RRQ article carefully https://www.linkedin.com/posts/ila-literacyresearch-scienceofreading-share-7373751634981736448-iJR7/
The key takeaway is this “study highlights how creators use narrative trends to position SoR as the only “right” way, often oversimplifying complex literacy debates.” Advocates of the three cuing system have pointed out many time it is not a system of intruction. It is a screening device. Those folks treating it as a system of instruction were not doing BL. Suppose I found some folks implementing SOR in a way contrary to what mainstream SOR folks do. Now I use that set of folks to argue against SOR. THat is using a strawman. That is exactly what happened in the case of BL. So my contention that the critics are using a strawman stands.
As for evidence that practices like letters don’t improve comprehesion, see this study.
Shanahan wrote a blog where he pointed out that the NRP did not say synthentic phonics it said systematic phonics. Yet many SOR advocates act is if it said synthentic. In that same post Shanahan said that those folks saying research calls for years of instruction are just “making that stuff up”. That is why I call your question misleading.
The other part is where the myth that the three cueing system failed came from. It came from the cauldron of the debate in social media (see other comments in this thread). The evidence that LTRS and other simliar programs do not improve comprehension is substantail. Here is one example On its own, it has not been shown to raise comprehension scores. https://riveteducation.org/a-timely-reminder-there-are-no-silver-bullets-even-with-professional-learning/. Direct, explicit instruction in synthetic phonics raises students’ decoding scores. It also improves teachers’ phonics knowledge. But as analysts like Johnson and Thomas have pointed out in both their books and blogs on the subject, LTRS training improves students’ decoding scores, but does not significantly improve their comprehension scores.