What is the Science of Reading?

Another insightful and thought provoking blog. Thank you! You’re spot on regarding references to SoR. I’m going through my own sites to check I am being clear in that regard. Even with years of data from thousands of teachers, there is no peer reviewed research. So also working on that.

I am going to print this quote.

‘If someone says your school isn’t aligned with the science of reading, they likely mean that you are not teaching phonemic awareness and phonics in the ways that they think you should.’

Nail, meet head.

Rudolf Flesch wrote in his 1955 Why Johnny Can't Read, "The only way to teach reading is by teaching spelling at the same time."

Ronald P. Carver in his 2000 The Cause of High and Low Reading Achievement wrote, "One very important way to learn how to pronounce more words accurately is sometimes overlooked, that is, learning to spell words accurately (Ehri, 1989a) Spelling is often considered a very important part of writing, but secondary to reading. In this regard, Gill (1992) noted that spelling was used to teach reading for almost 200 years, but 'by the beginning of the 20th century, the tide has so turned that learning to spell was seen largely incidental to learning to read.' However, Shanahan (1984) studied reading and spelling of second-graders and fifth-graders, and then hypothesized that '...spelling instruction would have the greatest impact on learning to read..... Evidence now exists which suggests that spelling words accurately is one of the most important parts of learning to decode words for beginning readers....In summary, teaching spelling and learning to spell words correctly is a very important way to increase pronunciation level for students below raudamaticiy. (178-179)

Using medical research as a model for reading practices is problematic. Having been involved in both it's clear that controlling variables is simple in medicine and all but impossible in classroom conducted research. Another is funding for research in new areas. My son help discover a novel method of detecting Parkinson's and 3 pharmas offered funding after one poster presentation. We have to rely on decades old research to valid practices while promising read research languishes.

"However, Shanahan (1984) studied reading and spelling of second-graders and fifth-graders, and then hypothesized that '...spelling instruction would have the greatest impact on learning to read..... Evidence now exists which suggests that spelling words accurately is one of the most important parts of learning to decode words for beginning readers....In summary, teaching spelling and learning to spell words correctly is a very important way to increase pronunciation level for students below raudamaticiy. (178-179)"

For more on the spelling-reading connection, I recommend Gentry and Ouellette's 2019 book Brain Words: How the Science of Reading Informs Teaching.

Dr. Shanahan,
I always look forward to reading your blogs and comments on Twitter. I appreciate the relevance of topics as they are usually spot on with my now. I am a reading specialist providing interventions in a K-5 school. There is so much information around about SOR. I was also trained in SBRI back in the days of Reading First. My undergrad focus was Whole Language. It can get a little overwhelming and we as educators need to read through a lot of information using our critical lenses. I have a question for you, we have a new curriculum director and he is advocating for the use of an online program for grades 3-5 called Reading Plus. We were asked to pilot the program last year. I found it too difficult for my challenged readers and lacking in any foundational skills instruction. I was also troubled by the strange eye training exercises. Have you looked at this program and could you share any insights as to whether this program has value? Thank you in advance.

Is the science of reading involve more than just the 5 components of reading?

Christina-
I don't know that program so could not comment on it. However, I rarely make comments on particular programs (though occasionally I will comment on a specific feature of a program if there is very specific research on that feature).
Although I don't know this program, I do know research has not been kind to programs that focus on eye exercises or eye movements.

thanks.

tim

Leah-
Indeed, science of reading is more than the 5 topics addressed by NRP. I listed several in the blog above and there is certainly more than that (and there are more specific information that science has uncovered since NRP in those five topic areas -- such as morphology in vocabulary or cohesion in reading comprehension).

tim

In my reading of what cognitive scientists have written, they do not describe how teachers should teach, but rather how the brain of learns and how it processes information. For application to instruction, cognitive experts often describe their findings as ‘guardrails’ or ‘boundary conditions.’
As one example, in his interview with Tom Bennett for ResearchED magazine, UVa. cognitive scientist Dan Willingham, who wrote "Raising Kids Who Read" and "The Reading Mind," advises:
"[E]ven when there’s something that scientists know with confidence that teachers should know,… it needs to be run through the filter of what teachers know about their classroom…. "
I have read Seidenberg and Dehaene, and I also thought their focus stayed on what neuroscience has measurably determined does and does not work in the brain. On those issues, they have credentials and expertise. I'd count their summaries of science as part of "the science of reading."

I agree with most of what Shanahan writes here. In order to make claims that instruction improves reading outcomes you need to assess whether instruction improves reading outcomes, not rely on basic research on reading. Basic research can be useful for making suggestions about what sorts of instruction should be tested, and may help explain why some forms of instruction work, but in the end, you need to assess the impact of instruction.

What I disagree about is the strength of the evidence that phonics improves reading outcomes. Shanahan cites the NRP, and I've detailed a series of serious problems with this meta-analysis and others that followed. See.

Jeffrey S. Bowers (2020) Reconsidering the Evidence that Systematic Phonics is more Effective than Alternative Methods of Reading Instruction. Educational Psychology Review, 32, 681-705. (go to: https://link.springer.com/article/10.1007/s10648-019-09515-y)

and

Jeffrey S. Bowers (2021). Yes children need to learn their GPCs but there really is little or no evidence that systematic or explicit phonics is effective: A response to Fletcher, Savage, and Sharon (2020). Educational Psychology Review. Published online on 13 March 2021. (go to: https://link.springer.com/content/pdf/10.1007/s10648-021-09602-z.pdf)

Shanahan also cites the National Earl Literacy Panel (2008) and I'll simply paste a section of this report. Clearly, this meta-analysis does not provide the strong evidence that he is claiming in this blog.

"Finally, there were significant problems with the quality of much of the research in this area.
Many studies used simple pretest-posttest designs, which provide no causally interpretable
evidence, and studies often did not provide evidence that these groups were equivalent prior
to an intervention or represented the same population. Often, there was evidence for group
differences that existed before the start of the intervention. The panel was unable to rely on the
data drawn from such badly designed studies, and they were excluded from all of the analyses
reported here. These flaws do not allow appropriate postintervention differences to be attributed
unambiguously to the intervention; neither do studies in which the intervention is confounded
with other important factors that could be the source of any observed effects. Ultimately,
building a larger and more comprehensive knowledge base concerning early literacy skill
development and promotion will require more high-quality research."

In science, Researcher A proposes a hypothesis, and Researchers B-Z attempt to falsify the hypothesis. A hypothesis is supported to the extent that it survives attempts at falsification. That is how it is supposed to work, but not in the case of science of reading. Here, the strongest proponents of the science of reading refuse to engage with challenges, even when published in the top journals of the field. For a somewhat amusing illustration of the failure to respond to challenges see: https://jeffbowers.blogs.bristol.ac.uk/blog/key-lesson/

Shanahan is a striking example of someone who claims to follow the science, but he sure don’t like to be challenged. A few months back Shanahan did in fact respond to a post of mine, only to make a series of false statements that I pointed out. His response: “I have read about 1/4 of this and I'm out. I don't intend respond again” see: http://www.shanahanonliteracy.com/blog/what-if-there-is-no-reading-research-on-an-issue#sthash.Owpv40Te.dpbs
Tim has stayed true to his word ignoring other comments I’ve made, for instance see: http://www.shanahanonliteracy.com/blog/do-you-have-any-pet-peeves-about-reading-here-are-my-top-ten-pt-1#sthash.KAAmLk15.dpbs

And so far, not response here either. I expect I’m just seen as an annoyance by Tim and many others here. I’m not saying what you want to hear, and not sufficiently deferential. But my posts focus on data and arguments, they are not personal. I hope at least some people here will start to wonder why no one responds to straightforward challenges. For instance, the quote I provided in post above comes from a document that Shanahan himself wrote is straightforwardly at odds with the strong claim he made here. I look forward to a response.

Eric--
Seidenberg's book makes very specific recommendations for teaching -- and instructional studies are barely mentioned. Likewise, much of the public discussion and debate of these issues turns on those cognitive and neurological studies and yet draw very specific conclusions for teaching.

thanks.

tim

Thank you for your fair and balanced approach to topics.
Somehow you always blog on something I have been trying to express accurately in the last few days. I'm sure others will feel the same, and are appreciative of the way you draw out points to consider.

As a first grade teacher it is so difficult to know what to do with all the science of reading information. I've read all the books by Seidenberg, Kilpatrick, Moats, Adams, Rasinski, Dehaene, etc., and instead of being more clear about what to teach I feel more anxious that I'm not doing everything I should be doing. I teach using Heggerty, Tools4Reading sound walls, Orton Gillingham materials, SWI and more, but am I really doing the right thing or are they fads that will swing out of fashion? The last blog talked about how sound walls weren't yet supported by research. I've also read that first graders typically don't need a full year of phonemic awareness yet Heggerty is a full year. Is there any way to know exactly what a first grade teacher should teach and for how long?

Dr Shanahan,
One of the finest reading teachers with whom I've had the privilege to work, recently remarked that she was reading, " All she could about the "science of reading"." I was floored.
I was a Reading Specialist, a strong proponent ,during my 23 years of teaching, of the NRP and taught many years in Reading First. I've pushed hard this year to return to the fundamentals of phonemic awareness and explicit phonics instruction. We've implemented a "word a day" school-wide challenge , made the National Spelling Bee a regular addition to our calendar ,and are attempting to utilize ongoing data to inform both teachers and students of specific individual skills that may be affecting their reading fluency and comprehension.
Yet, the fact that our school is 90 % second language and 100% dual language forces me to consider the neural pathways that must be formed before a child can become fluent in any language, whether spoken or written. For our children, most of whom speak a specific dialect of Egyptian Arabic, the key seems to be morphology.
However, my children and grandchildren have been very early readers. ( Age 3 -5 ) All have remained several years ahead in all areas of literacy. I've never been able to discover why, or how a 3 year year old would suddenly have the ability to pick up a book they'd never seen and read it to me except for the fact that, even before birth, we read to them.
Every night we read to them.
While we potty trained, we read to them.
Their bedrooms had shelves full of books. The playroom was downstairs.
They knew how to hold a book and turn the pages before they could properly handle a spoon.
Everyone in our home ended the night with a book in hand and flashlights were allowed in bed.
I apologize for my wordiness. But, the science of reading often leaves out what I've discovered to be key components necessary for a child to become a fluent reader with a vocabulary and comprehension strength that naturally lead to strong abilities as a writer-
a home full of books and parent(s) who read them.
As Strickland Gillilan so aptly stated:
" You may have tangible wealth untold;
Caskets of jewels and coffers of gold.
Richer than I you can never be -
I had a Mother (father, brother, teacher, etc. ) who read to me ."
The science of reading often ends at the door when a child returns home to a house where the magic of reading is absent.
Our school's greatest challenge now becomes to convince our parents to embrace literacy online and to become a part of that journey with their child.

Marie--

The point of teaching phonemic awareness is to get kids to the point at which they can easily segment words into phonemes and blend words into phonemes. Reading and phonemic awareness are reciprocal (they help each other), so the need for PA declines as kids learn to read. David Kilpatrick has hypothesized that some children (those with a phonological deficit) need more than this, though not likely in the regular classroom. Once your kids can fully segment, discontinue the PA work but continue with decoding, spelling, writing, and reading and that should do the trick. For those kids who just aren't clicking with decoding, it is possible that they will need even more PA than that but that should be a small number of kids given everything you are doing. My next blog is going to be on some of these issues, so watch for that.

tim

Feeling more informed - not to be too gullible about what is out there - because there is a lot under the Science of Reading banner. It does make my teaching life difficult! But thank you for the heads up!
I am wondering about this title "Reading at the Speed of Language" by Mark Seidenberg. Is it this one you were refering to "Language at the Speed of Sight..."?

Dr. Shanahan,
Happy to know that you will address the issues of phonemic awareness and how much to teach to which student. I feel sometimes I may be doing too much with all students but not sure how to differentiate. I am with Marie, on this post, and like her I have also read a lot about SoR to understand what I can best do for my students in the face of so much information. I wish for every student of mine to be a successful reader and so naturally I must do the best. I am looking forward to your post. Could you guide us to some highly effective practices (must dos) that would make our teachig robust?

Anila--
Indeed, that is what I meant -- I'm terrible with names and titles. sorry.

tim

Wow! It has been decades since I heard from anyone whose commentary on literacy resonates with my own explorations and teaching experiences from the 1990's. How do I keep my comments brief enough that I can actually hit send, and be ready to go back to the classroom tomorrow morning? I will have to leave out so much background to do this, but here goes . . .

1) Unlike teaching algebra, which is more repeatable and "scientific", ELA is about half the "art" of teaching. Developing English language arts involves building a community of learners, exploring complex communication skills, and encouraging learners to express thoughts and opinions that might not be considered "correct". How do you test those parts of language arts and critical thinking? Naturally most of the research tilts towards replicable results.
2) Even though effective independent thinking is the underlying goal of education, it sure is discouraged in the classroom, even when it works. I know this because of my own experience. After years of substitute teaching, I earned my Multiple Subject and High School English credentials at the age of 45. I was hired to teach a self-contained classroom for students in Mendocino County, who had been unsuccessful in regular high schools. In the 1990's. I could teach math, science and history with the available materials, but algebra and literacy, the higher level thinking spheres, were more difficult. An apprenticeship-style approach is much more effective in these areas. I spent four summers attending institutes with the California Reading and Literacy Project. I also traveled out of state for NCTM math conferences, but that's another story. When CAHSEE exams were introduced in the late nineties, we found that only 35% of our MCOE students passed the math exam and about 60% passed the ELA exam. In my classroom, 90-95% passed both exams and graduated. I attribute these replicable results, year after year, to "teaching as an art" as well as a science. My superiors did not agree with me. I was eager to share what I learned with a five-part literacy program, but my approach conflicted with a standards-based, but untested, Character Based Literacy program they had purchased for the next five years. When this program became mandatory, despite the lack of results, and I was directed to abandon my successful English program around 2013, I chose to retire and took time off to dance with cancer, then returned to substitute teaching. I am now, at the age of 70, teaching full-time at Willits High School, with wonderful young teachers and more enlightened administrators.
This brings me to my last point for now, 3) The "science of education" flows from the questions asked as well as who is asking, and whether the results support marketable programs. This paradigm can be very limited and biased. I am gratified by the studies exposing the underlying cultural prejudices in SAT tests, and inspired by the move into Common Core standards, doing their best to incorporate what we know about supporting more diverse learning skills. Surviving the challenges of Covid-19, climate change, peak oil, and an unsustainable economic system will require unprecedented levels of community building, environmental and political awareness, communication and literacy.

The NELP Report (2008) that Tim cites is an absolute must read. Download it and read it slowly. It provides the finest lesson in what a "synthesis of the scientific research" looks like and a foundation for helping us become critical, savvy consumers by understanding scientific process/requirements. NELP, like science, includes detailed clear definition of terms, statements of what is simply not known and delineates many specific aspects of how studies are problematic. What's left is not so much which heightens the challenge but this report delivers. Many of Tim's top-commented blogs are about letter names, sounds, phonemics, language development, etc., etc. This report shows why these topics remain difficult, ambiguous, and even contentious. Note: NELP is about EARLY literacy skills in children ages 0 -5 (birth through 5 years or kindergarten). Topics such as phonics play out differently at this early stage. Read it!

Thank you Tim. Like Sam Bammarito, my take away as well is:

"The lack of research supporting an instructional approach is NOT proof that an approach does not work. It may work, even if it hasn’t been tested yet. Lots of time it’s necessary to stretch research findings beyond what was directly studied. There is no other information to go on."

Thank you Tim!

Thank you for this well articulated summary of what is truly best practice. So often educators get sucked into half truths, and latch on to practices that may not have the most positive impact on their students' learning. I don't see much of this happening in reading in my district, but more in writing instruction right now. We have a self-proclaimed "literacy guru" who travels the state and she is boasting some writing practices that are not rooted in research. This is unlike her, but we have some teachers who have bought in hook, line and sinker and I fear for the young students they are impacting as many things are not developmentally appropriate. Would love to hear your thoughts on "the science of writing". In particular your thoughts on written responses in reading.

Tim, Thank you for continuing to share about what is (and what isn't) the Science of Reading. There is a group working to "define" the science of reading in a comprehensive way with a common definition. The Reading League and partners are rolling out their work one section at a time. Their work can be found at: Science of Reading: A Defining Moment (https://www.whatisthescienceofreading.org/?mc_cid=bd933354dc&mc_eid=648f23fb92) I'm not asking you to critique colleagues, but as their definition stands now, is it one that you could support?

I would love your feedback on my dissertation entitled A Mixed Methods Study of Teachers Conceptions of Benchmark Reading Assessments West Chester University, Pa.
Dr. Derby
Inspired by the work of Dr Gavin Brown

Where does classroom management fit into all this science discussion?

Title 1 junior high teacher here and only when I had that down - thank you Fred Jones Jr., Ph.D. - did any significant and consistent growth happen. And that's just not eliminating chaos, but truly creating an environment in a public school classroom that matches the safe, relaxed, and motivated student environment in which these profound investigations into reading science occur.

Frankly, in the populations who need the greatest interventions, strong practice in appopriate classroom management is woefully overlooked while being much harder to master than many instructional strategies these reading science studies reveal.

Marie-
Sorry, I wouldn't be able to provide that.
good luck.

tim

Where does classroom management fit into all this science discussion?

Title 1 junior high teacher here and only when I had that down - thank you Fred Jones Jr., Ph.D. - did any significant and consistent growth happen. And that's just not eliminating chaos, but truly creating an environment in a public school classroom that matches the safe, relaxed, and motivated student environment in which these profound investigations into reading science occur.

Frankly, in the populations who need the greatest interventions, strong practice in appopriate classroom management is woefully overlooked while being much harder to master than many instructional strategies these reading science studies reveal.

Christopher--

When I speak of reading instruction, I organize the information into three categories: amount of instruction, content of instruction, and quality of instruction. Something like classroom management conceptually can be included in either the amount category or the quality category. Indeed, there is research showing that off-task behavior reduces the amount of learning. Yes, it is a part of the science of reading. In fact, when I was director of reading in the Chicago Public Schools, I paid a substantial amount for training in classroom management for our coaches to impart to their schools.

Tim

Yes, in adult basic education, "Lots of time it’s necessary to stretch research findings beyond what was directly studied" rings true because relative little research into the neuroscience of learning, reading, and literacy instruction is available. Adult Basic Education in this country is built upon the instruction of well meaning teachers who get their instructional strategies straight from publishers staffing booths at a local or national conference (and I am one of those unwitting teachers.) But a few years ago I had data to say that a a particular student was not making progress in my classroom, and I was embarrassed that I did not know why.

I spent a year of independent research, as Carol Gibson Busching describes above. I know better, so now I do better, but I often feel alone in the world of adult literacy - save for those standing out in the field.

Thank you, Mr. Shanahan, for your straight talk.

As always, insightful communications on the subject Tim. Well done, to point to out the complexities in the world of reading research as so many are condensing their flavor of science about reading into the well accepted phrase, "the science of reading". Indeed it's a morass to sift it all out and where I've ended up through all the neuroscientific reading intervention research that we've done with Georgetown, MIT, the University of Washington, and currently Stanford, is that getting to the core of the global scientific view of reading points to the need for a more comprehensive nomenclature. Maybe the phase that should be used is "A Science of Reading".

Let's be clear, the use of the phrase "the science of reading", while understandable is tacitly a form of legitimation for reading program developers, as well as state DOE's, legislator's, and purveyor's of political driven policy, using the terminology "the science of reading" is in the main a compromise when trying to make real what it means to get meaning from squiggles (letters), representing sounds, and ultimately concepts (vocabulary and comprehension ) on the page. While I get the need to settle on something that scientifically legitimates practices and expenditures, looking at "reading research" in the abscence of "learning research" (the much broader spectrum of the cognitive and behavioral sciencific disciplines) at large is akin to looking through a microscope at cells to determine how the host organism functions at large in the environment. For lack of a better reference I have to think of reading research within the framwork of "The Cognitive Science of Learning". And luckily, there are well-researched theoretical models that get closer to placing knowing about reading into the more legitimate scientific universe of knowing about learning.

As always, insightful communications on the subject Tim. Well done, to point to out the complexities in the world of reading research as so many are condensing their flavor of science about reading into the well accepted phrase, "the science of reading". Indeed it's a morass to sift it all out and where I've ended up through all the neuroscientific reading intervention research that we've done with Georgetown, MIT, the University of Washington, and currently Stanford, is that getting to the core of the global scientific view of reading points to the need for a more comprehensive nomenclature. Maybe the phase that should be used is "A Science of Reading".

Let's be clear, the use of the phrase "the science of reading", while understandable is tacitly a form of legitimation for reading program developers, as well as state DOE's, legislator's, and purveyor's of political driven policy, using the terminology "the science of reading" is in the main a compromise when trying to make real what it means to get meaning from squiggles (letters), representing sounds, and ultimately concepts (vocabulary and comprehension ) on the page. While I get the need to settle on something that scientifically legitimates practices and expenditures, looking at "reading research" in the abscence of "learning research" (the much broader spectrum of the cognitive and behavioral sciencific disciplines) at large is akin to looking through a microscope at cells to determine how the host organism functions at large in the environment. For lack of a better reference I have to think of reading research within the framwork of "The Cognitive Science of Learning". And luckily, there are well-researched theoretical models that get closer to placing knowing about reading into the more legitimate scientific universe of knowing about learning.

Hi Dr. Shanahan,

I am wondering, what are your thoughts around districts who are simply walking away from the leveled reading approach based on SOR approach? I am very concerned that many schools are not using a balanced approach and rather throwing out so many proven practices that children do need to be successful readers. Help?

Hi Dr. Shanahan,

I am wondering, what are your thoughts around districts who are simply walking away from the leveled reading approach based on SOR approach? I am very concerned that many schools are not using a balanced approach and rather throwing out so many proven practices that children do need to be successful readers. Help?

Thanks for your input. I'm wondering if you oppose the term "evidence-based" reading pedagogies--instructional approaches suggested by sound empirical research?

Dr. Shanahan, what are the recommended time blocks for teaching the components of reading - shared reading, phonics/word study, guided reading, writing; or which components do you recommend are essential for an effective reading block? We are starting to look at our master schedule and we always struggle with fitting it all in. Also, what about the differences in grade level focus on those components...does that make much difference in the time allotments? As you know COVID has set many students back and we need to focus a lot more on the foundational skills, so that might take more time in the schedule. What are your thoughts?

Lucy--
There is no research on the topic that you asked about, however it is a decision that every practitioner has to make. First, I recommend that you get activities out of your mind and focus on what you are trying to teach (thus, we want kids to decode and spell words, read text fluently, comprehend what they read, and write their own texts). Guided reading or shared reading are activities not outcomes. Second, divide the instructional time among those learning goals that you have. In my scheme, I devote 25% of the time to each of the following: learning words and parts of words, oral reading fluency, reading comprehension, and writing. In some situations, I might add oral language development to this list and divide it into 5 parts. Of course, the times for each would not need to be equal.
Good luck.

tim

Hi Tim,

Where can I find the Science of reading programs' list?
Thank you.

Sevil

Sevil--

Such a thing does not exist. Science of reading is not a program. Sorry.

tim