Are We Teaching Reading Comprehension? Part I

Ann-
Decoding instruction is very important and needs to be there... but these days we are seeing confused school people emphasizing phonemic awareness in high school and decodable text in 2nd grade -- ignoring any research on the matter. If decoding is well taught, most students still will be benefit from reading comprehension instruction. You are absolutely correct.

tim

Tim, this is extremely helpful. I work with teachers who are exhausted and very busy teaching ELA, but this will help me help them understand that much of what they are doing is not teaching reading comprehension. I find that lack of teacher training in this area and lack of understanding by administrators and district leaders is my biggest hurdle. With these types of examples and more training, I think we could make a big difference. Our students spend a lot of time in class, but they are just not getting what they need to be better readers. I can’t thank you enough for your insight. You always break research down into layman’s terms, and you have been most influential on helping me teach teachers. Thanks again.

Seems to me that teaching text structure, transitions, and varieties of sentence syntax would be some means of teaching comprehension per your example. Additionally, these means are able to be assessed with independent reading to determine degree of transfer over time.

Very interesting article; I look forward to part 2.

I teach 3rd grade. I am concerned that we may often push our students harder than what they are developmentally ready to accept with many of these higher-order thinking skills, and all for the sake of high stakes testing.

I have been trained in the Science of Reading, and see many benefits of that research. However, I can’t help but consider how vastly different my own reading instruction and reading experience were like with whole language. I learned to read (spoiler alert to revealing my age) using Dick and Jane books. Where did the comprehension develop from that type of instruction? I don’t remember being specifically taught how to discern how sentences connected or how an author connected paragraphs, yet somehow I seemed to comprehend what I was reading. Additionally, I only recall one standardized test in elementary school — The California Achievement Test. I wonder if the lack of testing pressure allowed me to develop strong reading skills at the pace appropriate for me?

Mark Pennington - I agree with the different means of teaching comprehension you mention. I'm curious about the "assessments with independent reading" you mention. What would those look like?

Hi Tim
I mostly struggle with finding good rich texts for this reading comprehension! Right now I tend to make my own and try to find others. But if we are to really teach it right ..we really need rich texts and I struggle to find them. Would love your thoughts on this matter !

This new study has the potential to raise outcomes for kids as Hansford's Sold a Story has done. Kudos to Dr. Shanahan for featuring it.

A growing knowledge-centric movement suggests we cannot teach transferrable strategies. This claim is wrong. It suggests that strategies do not transfer across texts and cannot be used independently to build new knowledge. No research supports this claim.

The evidence overwhelmingly shows that students can learn and apply strategies to approach unfamiliar texts—even when they have little prior knowledge of the topic.

The key word here is transferrable.

Everyone agrees that knowledge-building and rich content are essential. (Though defenders of knowledge-centric try to paint anyone raising questions as anti-knowledge or anti-curriculum. This could not be further from the truth. Those defending strategies deeply value knowledge and curriculum.)

Over the past five years, a small connected group has been promoting the idea that foregrounding knowledge-building and downplaying the role of transferrable strategies is the bst way to help students navigate new texts.

- A side impact of this group has been that programs branded 'knowledge-building' sales have skyrocketed, yet with no evidence these perform better than all core program published after 2012 when the new standards dictated that all programs must be 50/50 informational texts / narrative stories. The differences between the new 'knowledge-building' branded and these core programs are unclear and there is no evidence swapping will raise outcomes. Yet, we've been swept away purchasing them in the hopes they will do better for kids. They don't seem to.

Knowledge-building helps incrementally over time, no doubt. But strategies help immediately—and significantly. Dismissing their importance risks weakening students' ability to navigate new material independently. We cannot teach all knowledge. Students must have a way to approach new unfamiliar topics.

Some argue that we have overdone comprehension instruction, but the reality is quite the opposite—we haven’t done it enough, or at least not well enough. Too often, strategy instruction is superficial, rushed, or ineffective, rather than taught deeply and practiced until it truly transfers. That is the point of this new landmark Capin study cited here.

This is not an either/or choice. Teaching both transferrable strategies and rich content is the most effective approach. Any programs published after 2012 can be used to do this. Shiney new branded knowledge-building that dismiss strategies are unlikely to have the impact that improving how we teach strategies will, according to this study would be a fair take on it.
Strategies must be explicitly taught and practiced until students can apply them independently—because transfer isn’t just possible, it’s the goal.

Jackie-
Text is central when it comes to teaching reading comprehension. It is essential that the text convey content that we would want our students to know and sufficient linguistic/textual/cognitive demands that would allow us to teach students how to negotiate those. That means high quality, worthwhile text, that is sufficient difficult. (I encourage the use of programs because I think it makes it possible for teachers to work through content/text demands in systematic ways that could increase kids' chances of learning.

tim

Lesley--

I ungraciously didn't mention it in this blog, but your heads up on the issue is what led me to write it.

thanks.

tim

Annette-
Like you, I learned to read from a version of the Dick and Jane books. Those programs had very little that we would accept as comprehension instruction (as you'll see in my next blog). There is no question that we can successfully teach reading without much explicit direction in this area. However, let's not think of reading ability as a discrete skill that you either have or don't have. The issue is how well do we want our kids to read. Reading achievement in the US hasn't risen since 1970 though the demands for literacy in our society have increased considerably. Don't think of this kind of article as addressing how can we teach reading but as how can we teach reading so that larger numbers of kids get to higher levels of achievement.

thanks.

tim

Hi Tim,
I agree that teaching two sentence definition/explanation text structure supports comprehension, but I think teaching the practice of referencing text would also help. The student, who is not a practiced reader, might not notice when reading and likely would not return to text. They assume that ‘good readers' just know. Maybe the teacher could have asked the student to return to paragraph 2 of the text and pointed out the 2 sentence text structure.

But most of all the 2 page reading scenario screams the student’s vocabulary and how best to increase it. Pre-teaching the tier 2 word ‘erode’ and the tier 3 version ‘erosion’ would have helped this student understand the passage. Throwing in a little morphology would be nice too... Mary BH

Thank you for your thoughtful blog!!! I like that you are talking about explicit “how to’s” in teaching reading comprehension. I feel like there are so many materials that claim to improve comprehension, but I don’t totally buy into their proclaimed results. I liked how you described that an author often defines and gives an example and that we can teach kids to pay attention to this as a strategy of comprehension. I would like to know if you have other tactics you feel truly beneficial to comprehension. One strategy I use is telling students that there will always be a main idea of a chapter or section, there will always be details of who, what, where, when, why, and how, background information and vocabulary are critical. I also have a “plot chart” out and tell them fiction (and life) are made up of these components. I look forward to the strategies you suggest teachers should employ to improve comprehension.

Rosalie--
That kind of thing happens often when folks are typing on their phones. It is not carelessness as much as incoordination. I know it happens to me too.

tim

Leslie, I couldn’t agree with you more! Funny enough, I just wrote all about the points you make on my latest blog post, which I published the day before Dr. Shanahan’s insightful (as always) words.

Jw and Annette-

Thank you for your insights on how high stakes testing has changed. The tests now are not gaged for appropriate child development levels. I don't think you can accurately compare student achievement now with results from past assessments. As to the "Dick and Jane" books, maybe the key there was simplicity. This is a word that I think all educators should think about.
Maybe students now are forced to overthink and overanalyze everything they try to read to the point where the whole process is seen as something very unappealing. Maybe we are asking students to take on too much too fast, and the process of enjoying and understanding books becomes a lost.

Not knowing American literacy history, it is not possible for me to comment on what was done in the US , historically.However I do know what was happening here in NZ and what I am interested in specifically that NZ scored the highest in reading comprehension , in an international test in 1970.

After the introduction of Whole Language and Marie Clay, about 1970 , we now score the lowest in English speaking countries.

Curiously much of the comprehension material for students prior to 1970 in NZ before WL, came from America.
This included basal readers , Whole Word Ginn ,McKee etc with the readers and accompanying workbooks, the SRA , class sets , as well and other programmes like the Reader' s Digest 's Skill Builders. The main point was that there was much emphasis placed on comprehension and exercises were done everyday in classrooms. In fact some days more than one comprehension exercise was done daily since the topic of 'English ', which concentrated on written and spoken work, also included comprehension .

By no means am I the first to say writing about a topic helps you understand the concepts better. In 'English' you were required to answer conventional direct comprehension exercises like : Why does ? Where is ? When ? etc and in then write in another section , which required research of the topic , to discuss more extensive and detailed aspects of the topic covered . If the topic was house building materials , you were asked to draw a plan of of your home and list all various materials used eg. Then direction was given to write two paragraphs describing the difference between brick houses and wooden ones.

Comprehension was clearly about acquiring knowledge.

I am unashamedly a traditionalist and believe there is a trend in the West to return to a knowledge rich curriculum , structured learning in not just literacy, along with handwriting and firm discipline and some rote learning in the basics and other aspects of traditional teaching . It is what I experienced when I was a child and I saw much superior academic achievement then, than what we have now .

Look forward to part 2 of this topic.

Deep comprehension requires cultural alignment: the reader's inferences align sufficiently with the author's implications to make sense. This requires deep knowledge by the teacher about authors' intentions so that rich learning dialogue can take place in the classroom. I'm particularly conscious of this from working with English as a Second Language students in cross-cultural classrooms. This knowledge goes much further than asking 'who, what, when, where, why'.
For example, here is the opening sentence from 'The Little Red Hen': 'Down the road and over the hill lived a little red hen.' You might ask 'where' and 'who', or you can explain to students that this is the sort of sentence that is typically part of a story orientation section: it begins with two 'where phrases' that describe the setting before we reach the main character; it includes the generic verb 'lived' to introduce the character; the noun group used to describe the character begins with the indefinite article 'a' to signal that we haven't previously met this hen, the adjective 'little' and the choice of 'hen' create a diminutive, female character with whom the reader can empathise. (It wouldn't work if it was a rooster.) I don't know why the hen is red except perhaps for assonance: i.e. 'red' and 'hen' share the same middle vowel that links them together.
These are examples of the sort of meaning-making that might be shared during whole class close reading, to develop common, transferable literate knowledge that students can subsequently apply in independent reading.

The end of year testing that students need to take has changed. They continue to get more and more difficult. This year our state is starting cluster tests. They look at the passages on the left side of the screen and then they answer questions with drop down answers. If they get one wrong if can effect other answers. Answers are not "right there." The majority of adults would not get the correct answer. How can people compare reading comprehension now from the past when the questions use to be choose a-d? The question now are sooo difficult I struggle getting them correct.

One aspect that connects both knowledge based curriculum focus and comprehension focus is vocabulary. They are reciprocal concepts in a way, especially for English language development.

Let's not put all decodable texts in the same basket. Sure, the contrived tongue twistered ones have no place in the teaching of reading, let alone comprehension, but there are phonically controlled texts out there for children at all stages of learning the code that are perfectly comprehensible with natural language patterns. So very useful for students learning the hows of understanding a text in the early years.

Berys-

Research has found only a small benefit for a brief period of time for decodables of any kind. I think their use is being overdone by people who are ignoring the "science of reading."

tim

Half the time we use decodables as dictation exercises for spelling .

Miriam P. Trehearne

It is important for classroom teachers, administrators and parents to understand that oral language (the foundation of all literacy learning), comprehension (word and world knowledge being key components), and writing are all "curriculum gap areas"
(Teale, Paciga, and Hoffman 2007) that in general deserve much more focus at school and at home!

Thank you again and again Tim for sharing what the research really tells us!

Dear Tim,

Thank you for this article on a critical subject. I really look forward to the next article, which I assume is going to be practical in the sense of sharing the key skills we need readers to have and how to teach/model these in the classroom. It may be a very rough generalization but that switch from 'learning to read' to 'reading to learn' is real and needs careful consideration.

You shared that one of the challenges in this area is researchers differing categorization of what is reading comprehension instruction and what is assessment of reading comprehension. I would love to know if there is more general consensus on what constitute the key reading comprehension skills that we need to teach students. That sort of practical listing is important for teachers and school leaders if we are to do a better job of teaching reading comprehension. The example you shared is a good case in point, because it is the sort of skill that can be practiced in every classroom. It seems to me that when we discuss transferable skills we are making the implicit statement that students transfer them from classroom to classroom. No doubt that's true, but it is most true I would guess when they have been mastered to some degree. The other way of looking at transferable skills is to think of their instruction as being transferable - I guess is what we mean when we say 'Everybody is a reading teacher'. The "Grand Canyon" example you gave in terms of content could appear in a text in a reading lesson, a social studies lesson or a science lesson. And while in many cases, particularly elementary classrooms, the same teacher may be teaching all subjects, there may well be situations where different teachers are involve, such as my school. This makes me think that we need every teacher to have some grounding in knowing and teaching effectively reading comprehension strategies. For that, we need a common understanding of what reading comprehension is and what teaching it looks like. Are we there yet? What I see among my colleagues in our school is a much richer understanding of learning to read than reading comprehension. Our collective efficacy as teachers of reading is tremendous and we take great pride in being intentional about early reading skills. We want to practice the same level of intentionality with reading comprehension.

As teachers, we aim to teach for comprehension? If so, that makes sense to me. Thanks for the blog post.

I'm a big believer in Reuven Feuerstein's approach to teaching. He talks about 'transcendence.' Your example of talking about how to infer the author's intent, even when she doesn't state it explicitly, is an example of 'how' to read it. Giving several different examples over a period of time, is 'transcendence.' It also is more transcendent if you talk about how we infer meanings in speech. For example 'Good job!' stated in one way is friendly, but you can say it in a sarcastic way, which makes it a condescending put down. Kids know about this at every level of elementary school, but they don't always realize how this same skill (understanding the intent of a speaker/writer) is needed in reading books. I believe that all of the reliance on 'comprehension questions', instead of the teacher choosing a skill like the one you mentioned, and reading informative passages on curricular areas (like an extended unit on geology, for example).

Thanks for this Tim,

I wanted to highlight this point from Lesley Laud:

"We cannot teach all knowledge. Students must have a way to approach new unfamiliar topics."

It seems to me that this notion of developing instruction that results in student learning that affords the capacity for self-directed learning of untaught knowledge is basically the primary goal of education.

My supervisor John Kirby along with Michael Lawson edited an excellent book published in 2012 titled, "Enhancing the Quality of learning: Dispositions, Instruction, and Learning Processes." In their introduction they offer definitions of high quality and low quality learning that I think just about everyone will agree with and speaks directly to this point. They argued:

Low-quality learning will result in knowledge that is narrow in scope, fragmented, and does not lead to other learning.

High-quality learning results in knowledge that is extensive, integrative and generative so that it supports transfer.

Peter and Lesley--
I agree with both of you that depending on knowledge building alone to increase comprehension would be a huge mistake -- and one that seems to be growing in popularity. Nevertheless, I remember when I was a young teacher saying dumb things like, "It doesn't matter what they read as long as they are reading." Knowledge matters -- and it matters not only in reading. The reading of texts within reading instruction should be taken as an opportunity to build knowledge and not just reading and language abilities.
tim

I just wrote a blog on this topic today before seeing this post. I sent you a link to that blog via X. I think the key for improving comprehension remains directly teaching research-based comprehension strategies to students. That means teaching them in a way that results in students internalizing and actually using the strategies. Naming a strategy, describing a strategy, or even practicing a strategy is not enough. By the way, your views on teaching strategies have affected the way I talk to teachers about what to do. You pointed out that asking them to use a strategy they really don't need for that particular piece is not advisable. Based on that, I now ask teachers I in-service to teach the new strategy and then use it for a while. Then, they can come back and talk about the how and why of the strategies they've used lately. If memory serves, the NRP said that teaching about multiple strategies concurrently pays off. So, I advise allowing the class’s conversations to include all the comprehension strategies covered to date. The issue of how much time teachers actually spend directly teaching comprehension has not been answered by recent observational research that I am aware of. If such research exists, I would love to hear about it. Thanks for all that you've done and are doing for our field. Dr. Sam from St. Louis.

Chris Such has provided an alternative perspective to this blog post on building reading comprehension on Twitter: https://x.com/Suchmo83/status/1901729405455814767

Because his latest book (Primary Reading Simplified) is based on the perspective that he outlines in the Twitter post, it would be great if you could give us your thoughts on his perspective. That way we can be clear if he is on the money or if you are on the money or both of are on the money! There are many teachers and schools in Australia and the United Kingdom who use Chris Such’s suggestions for teaching reading comprehension so many students will be affected by this.

Sam-
One correction here. NRP found that teaching multiple strategies was more effective than teaching a single strategy. However, teaching those strategies concurrently (like reciprocal teaching) was no more effective than teaching those strategies consecutively. Some experts prefer the former (with no supporting data) and I tend to lean towards the latter (with no supporting data) because I prefer its clarity.
tim

Tracey--

Chris Such's position seems to be that as long as students are reading text and comprehending it there is a possibility that they will learn something from this and that will transfer to some other small number of texts -- and that there is nothing one can do more powerfully to improve comprehension in ways that transfer to other texts.

I agree with him on the idea that comprehension improvement tends to be gradual and that learning about erosion will help you to read about other texts on erosion and related topics. However, there is a substantial body of research showing that it is possible to teach students things that do improve their ability to read a much wider range of texts (including the hundreds of studies on comprehension strategies). Also, evidence is accumulating that text selection matters (we tend to try to teach comprehension with texts that are too easy for the students -- which by making comprehension learning too gradual reduces that accumulation of abilities).

tim

Hi.

Professor Shanahan: You appear to have misread (or entirely missed) parts of what I said in my tweet thread. I most certainly do not think - nor did I say - "that there is nothing one can do more powerfully to improve comprehension in ways that transfer to other texts" than developing understanding of written language through text experience. Nor did I ignore the value of comprehension strategies.

In fact, my central argument was that there were things we could teach that were more powerful (or transferable, as I described it) but that the accumulation of less transferable knowledge about written language and the wider world mattered too and didn't have zero effect on comprehension ability, as implied in your blog.

Here are some quotes from what I wrote:

"The perfect example of this (highly transferable things worth teaching) is comprehension strategies, the use of which might theoretically be applied to all texts.

And, yes, learning about text structures, cohesive devices, conjunctions, etc is especially useful because of how many texts it applies to."

I was simply arguing the point that the gradual accumulation of knowledge about language - including from teacher explanations of words and concepts - *does* potentially improve comprehension abilities as well. Specifically, I said this:

"Of course, we need to engage pupils in analysis of texts and discussion that builds strategic reading habits and increases pupils' awareness of common sentence structures, text structures, etc. But we also need to provide pupils with a lot of pacy, meaningful, varied reading experiences, not only because of pupil motivation, but because of the gradual accumulation of knowledge about written language and the world that it provides."

I'm not sure how you have reached the conclusion from this that I think the comprehension strategies or the selection of texts doesn't matter.

I would very much appreciate a response.

Best wishes

Chris

Just to add:

My thread made no mention about text difficulty.

To be clear, I agree that all pupils are entitled to - and benefit from - teaching using challenging texts. The whole purpose of teaching is to support pupils to understand - and learn from - texts that might not be as fully appreciated or understood without the support of the teacher.

Christopher--
You are correct that incidental learning occurs -- not only during reading but during any direct or vicarious experiences that students have. However, teachers need to aim their efforts at eliciting learning intentionally. That's why we need to focus on lessons most likely to generalize. The most talented kids will learn incredible amounts on their own from their experiences (that's consistent with definitions of intelligence), teaching gives the rest of us a chance to enhance our learning. Just having kids read texts that they can comprehend may have some unmapped impact on future reading experiences but my teaching responsibility is to aim for more than that. It is a continuum of course, but the distinction matters.

tim

Professor Shanahan:

I have not stated - nor do I advocate - that kids should "just read texts that they can comprehend". What I *do* advocate is that one consideration of our instruction should be increasing the breadth of pupils' reading experience. Some pacy experiences with challenging texts - including some well-timed teacher explanation and brief questioning - is, I think, a useful component of this, though it cannot replace slower, analytical reading any more than it can replace fluency practice.

I also think that it is incorrect to suggest that only "the most talented kids" learn about language - written or otherwise - from the sheer breadth and variety of their experiences. In fact, I would argue that often what makes these kids appear to be "most talented" is that they have received a breadth of reading experience outside of the classroom, either through family support or independent reading. Thus, one component of effective reading instruction is to ensure that pupils leave our school having read *a lot* of challenging texts with us, something that I believe requires us to complement slower, analytical reading with some meaningful, pacier reading experiences. (It is if course possible to find a middle ground between these extremes that offers both, something like 'questioning the author' seems to do this where I have seen it implemented well.)

Perhaps this is somewhat lost in transatlantic translation and we are talking at cross purposes. If so, I hold my hand up and accept my share of the responsibility for that. I tend to read your blogs with an eye on how they will likely be interpreted by the teachers in the English education system. For perspective, it is very common in English schools for pupils to read no more than a couple of pages across an entire week of instruction (i.e. 2-3 hours of reading lessons), analysing them in depth and practising how to answer different types of questions. As a result, an entire year's worth of reading instruction might involve pupils experiencing 15,000-20,000 words, the amount we might see in a single relatively short book aimed at a 9-year-old. When I talk about needing to increase pupils' breadth of reading experiences, it is this sort of instruction that I (partly) have in mind. Perhaps, instruction in the US always involves much more engagement with written text than this. Or perhaps the vast majority of pupils are doing lots of reading at home. But I hope this explains a little more about my perspective.

Chris--

I can't see how you or anyone could read my blogs and conclude that I don't support the idea that kids should be reading more than a couple of pages during reading instruction. I think you're missing the value of the distinction that I'm making here. Very confusing.

tim

Just a quick aside on the subject of incidental learning, the breadth of reading experience and "most talented" pupils:

While I've found relatively few studies on this subject, Sutherland et al's (2019) paper on the subject actually saw by far the most gains for *lower-attaining* pupils when instruction included intentionally pacier reading experiences. In contrast, higher-attaining pupils gained considerably less.

Do you know of other papers I might read that suggest otherwise (or just other papers on this subject more generally)?

Professor Shanahan:

I understand the distinction you are making. My point is that your blog, in part, portrays this distinction as a dichotomy between things that support comprehension of other texts (i.e. analysing common linguistic structures) and things that don't (i.e. explaining concepts to pupils). In contrast, I think *both* support comprehension of other texts. The former is a slower process that teaches something widely transferable; the latter is a quicker process that teaches something less widely transferable. I think that both have their place, and I think your blog (perhaps unintentionally) might give the impression that only the former does.

Chris--

I don't know of such papers.

tim

Dr. S.,

A powerful and timely topic to put forth. Thanks. I'll read your next blog after pointing out what makes me wonder if there isn't one missing artifact of the exchanges on teaching comprehension in this blog. I wondering what you think. I wonder what our collective comprehension is regarding what comprehension really is, and more importantly then, how to teach it.

You pointed out initially, with the emergence of the cognitive sciences that there has been a shift in the way we think about teaching comprehension. So true. Yet, throughout this blog there has not been one word in the discussion about what the underlying brain-based cognitive mechanisms are for language comprehension, not to mention memory, the presumptive undergarment of comprehension. As we all know, the underlying brain sensory-cognitive mechanisms (phonetic and orthographic processing- cognitively perceiving sounds and letters) that are foundational to decoding and encoding (reading print and spelling) have powerfully led to advances in teaching decoding and spelling (done correctly). What are the sensory-cognitive mechanisms associated with oral and written language comprehension? What parallel brain processing architecture (coding mechanisms) serves comprehension such that the codes of phonemic awareness and alphabetic-based symbol imagery serve decoding and spelling? The catalyst for comprehension has to begin in the brain right? Or, can we just move on to pedagogies? Language is an invented artifact of culture to share information/knowledge, and communicate, as Schopenhauer intonated, "to get the world outside the head into our heads". By what underlying nuronal code(s) is the language code dependent when reading a book, processed in the human brain to "get meaning" and once understood. More importantly, it it entact for all children and if not how might it be be metacognitvely be awakened, taught, like we teach sounds and letters? Could it be that strategies are but a proxy for us trying to find ways to really uncover the way the brain needs to be taught? There is an emormous corpus of historical knowledge and current neuroscientific applied memory, and cognitive comprehension research to answer these questions.

A hint. Einstein said "If I can't picture it, I cant understand it", or maybe a more urbane reference would offer a better clue - a picture is worth a thousand words.

Reading instruction does need to focus, not on the text of the day, but the text of the future. Explicitly teaching text structures, like the example given, can help students analyze the purpose and the meaning of the text.