Doug Lemov's field notes

Reflections on teaching, literacy, coaching, and practice.

09.23.16A Few Minutes with Literacy Expert Tim Shanahan

Image result for reading researchTim Shanahan is one of the country’s foremost experts on literacy. He’s a Distinguished Professor Emeritus at the University of Illinois at Chicago where he founded and directed its Center for Literacy. Though he understands the research on reading as well as anyone in the field, he’s also done plenty of real-world work, having been principal investigator for national Title I studies and Director of Reading for  Chicago Public Schools. He has also served–by presidential appointment–on the Advisory Board of the National Institute for Literacy.

He recently agreed to let us ask him a series of burning questions about reading and literacy here on the Field Notes blog.The interview will give most readers lots (and lots of research) to think about and we’re confident you won’t be disappointed.  Our questions and his answers appear below. If you want more, he blogs at

  • FN: It’s often taken as  an article of faith that giving kids leveled text that emphasizes readability is ‘research based.’ But you followed  the breadcrumbs on the research and found that this wasn’t so true, right?


TS: You can find educational experts claiming that students needed to be taught at their “reading levels” if they are going to make gains going back to the 1940s. What this claim meant was that there was a particular way to match texts to students… by estimating how hard a text was by its readability and by estimating how well a student could read by listening to the student read and having him or her answer questions about the text. The trick was to find a text in which students would only make 3-5 oral reading mistakes per 100 words and would answer 75-89% of the questions. It all appeared very scientific, and as a teacher I found it pretty convincing. Nevertheless, in all that time, there have only been about a half dozen studies done that attempted to test this and for the most part the findings suggest either that student text match makes no difference (that is, placing students with the just right book offers students neither a learning or motivational advantage) or it holds kids back, preventing them from the opportunity of working with those more complex features of text (meaning that placing kids like that actually lowers their achievement).



  • FN: Tell us a little about the term ‘guided reading.’  It seems like it’s taken on a meaning that’s different from what it originally implied. What would the tenets of effective guided reading be?


TS: Guided reading has also been around for a long time. The term first appears in the 1930s, but doesn’t become widely used until the 1950s. Originally the term was used to describe those textbook lessons in which students would read stories or chapters under teacher supervision with the goal of fostering successful reading comprehension. Basically, guided reading referred to the lesson plan proposed by basal reader publishers for leading kids in a communal reading. Different publishers referred to that lesson plan template as either guided reading or directed reading, and there were variations on the scheme from basal to basal, but usually the steps included… introduction or pre-teaching of vocabulary, background information review, motivation or purpose setting, stopping points where discussion would be held, and questions to ask the students.


These days guided reading certainly continues to carry the idea of supervised communal reading (though now more often with little books), but it also, in many people’s minds, refers to small group reading instruction and especially to such instruction delivered at students’ reading levels. The idea of guided reading, by its original definition, is still a good one, and small group instruction can certainly be beneficial and should have its place in a classroom for intensifying student learning and increasing the amount of student response and interaction. However, the idea that this work must be done at student reading levels has simply not held up under scrutiny.



  • FN: We think vocabulary is a hidden driver of long term outcomes in both reading and in school overall.  In Reading Reconsidered we advise teachers to invest more heavily in it.  But vocabulary has an assessment problem. It’s hard to tell how good students’ vocabulary is or how fast vocabularies are growing.  Any practical thoughts on measurement of vocabulary for schools?


TS: Vocabulary is important—especially as one moves up the grades and confronts texts that use a more diversified collection of words. The correlation of vocabulary and comprehension is surprisingly low in the earliest grades, but that correlation increases every year as students advance through school. Initially vocabulary isn’t that important because the word load of most beginning reading materials don’t exceed children’s oral language development (for years, publishers worked very hard at making sure, in fact, that the vocabulary demands of early textbooks did not exceed what children were likely to know in this regard). But, as this question notes, vocabulary assessment is challenging. If all that you want to know is whether a student is making progress in vocabulary development from year to year or what their normative level of vocabulary knowledge might be, then there are standardized instruments like the Peabody Picture Vocabulary Test that can be group administered. The benefit of such testing is that it should account for both the intentional and incidental aspects of vocabulary learning. Students certainly can learn words that are taught to them, but a substantial share of vocabulary development results from independent reading, media experience, and social interactions, when there may be no real intent to learn new words. However, what we often really want to know is how much impact our intentional efforts to foster vocabulary growth are having, and for that I would suggest simply keeping track of all the words that kids are exposed to through instruction and evaluating their knowledge of random sets of these words from time to time. Thus, let’s say across the curriculum, you were introducing/exposing/teaching 20 words per week. Perhaps at the end of the month you would randomly select 20 of these 80 to 100 words, to estimate what percentage of these kids were maintaining. Then at the end of two months, you’d have 160-180 words to choose from, and so on. This would not tell you how fast kids’ vocabularies were growing because it would ignore all the incidental learning that we know takes place, but it would allow teachers to estimate how effective their vocabulary teaching efforts were and if some kids were benefiting more than others.



  • FN: We’ve often argued that ‘reading strategies’ are founded on a confusion between correlation and cause. When we hear ‘good readers visualize when they read so we should teach kids to visualize what they are reading,” we often note that good readers also often relax with a  glass of wine but…..  Anyway, can you give a gloss on what we know about why reading strategies are problematic?


TS: I’ll challenge the premise of this question. The term reading strategies or more comprehension strategies refers to a small collection of actions that readers can intentionally take to try to increase their understanding or recall of information. Some experts bridle at the term “comprehension strategies” because they suspect the real meaning is closer to study skills than comprehension enhancement (in other words, they tend to be more similar to what someone might do to prepare for a test than to just read with understanding. Strategies usually need to be distinguished from skills in that strategies are to be more intentional and require greater adjustment to circumstances on the part of the reader.


The research does not just show that “good readers visualize or good readers ask themselves questions,” but in fact, there is a substantial body of research showing that, indeed, the teaching of such strategies has a causal impact on reading achievement. That is, there are more than 200 studies showing that the teaching of such reading approaches enhances reading comprehension on both teacher-made and standardized tests. Some of the strategies that have substantial research support are summarization (stopping during reading at various points to sum up for yourself what an author has told you up to that point), questioning (asking yourself questions about what the text said and trying to answer these questions—going back to the text if you cannot), visualizing (trying to visualize or get an image of what the text is describing), background knowledge activation (thinking about or trying to connect what you already know about a topic with what the text says), and monitoring (and paying attention to whether you are understanding the meaning of a text and if you are not to take some action to remedy the problem including rereading, looking up words, asking for help, etc.).


A big part of the success of strategies seems to me to be that strategies give students something to do with their minds while they are reading. Psychologists estimate that the average adult’s mind is wandering about 11% of the time during reading. One can imagine that kids might suffer this even more often, especially if they don’t like reading or are struggling to read. If I have to read and stop every section or every 3 pages to think about what the text says so far, I will probably end up learning more of the information. However, the research also finds that multiple strategies—whether taught one at a time or simultaneously—are more advantageous than single strategies which suggests that it is not just a paying attention issue, but that different strategies must provide some different affordances to making sense of a text (though attention still might be a big part of it).


Both the National Reading Panel and the What Works Clearinghouse, using different research review methodology, have determined the effectiveness and value of reading strategy instruction.


That does not mean that strategy instruction is without problem. For example, most studies of strategy teaching have shown that much of their impact can be developed in a relatively brief time (most studies have only been for about six weeks) and yet recommendations and curricula development have usually emphasized much heavier doses of strategy instruction—like all year, every year. Dosages that go well beyond any evidence. Most, if not all, of the major leading researchers have always emphasized the importance of teaching such strategies in the context of having students work with rich texts that would be worth comprehending and of ensuring that the lessons end up with students learning both the strategies and the text content. However, whether one observes in classrooms, reviews curriculum, or discusses teacher experiences, it appears that this dual focus is rarely the case (teachers often tell me, “well the kids didn’t get this story, but their summaries are getting better”—yikes). Often strategies are taught like skills, without attention to the varied contexts created by different texts (it was harder to ask helpful questions about this text because the author… or did you notice that this author used very little description, visualizing was almost like making up your own story, not very useful for this story).


I would point out that many strategy opponents seem more annoyed by particular strategies than about strategies in general. For instance, David Coleman argues against strategies and then openly espouse that readers should try to imagine what the counterarguments might be when reading a text like Martin King’s Letter from Birmingham Jail. Seems like a good strategy to me, though it hasn’t been studied. I would argue that something like “the leapfrog read,” if kids were learning to use it as a tool, would also be a strategy—an intentional effort to confront and think about the ideas in texts in a particular way. But again without research. Or my own strategy for teaching kids to construct themes from stories… Just because most strategies have been general—meaning they have their greatest value for younger and poorer readers and provide less payoff for better and older readers (since they are already usually reasonably good at paying attention) doesn’t mean that we shouldn’t teach kids how to intentionally make sense of text.


  • FN: One of the big challenges of helping teachers to Close Read more and better is defining it.  We’ve seen quite a few of teachers doing what looks a lot like the usual reading strategies business but calling it close reading for example. What’s your definition and what should or should not be a part of it?


TS: There are a lot of different ways to approach reading, close reading being just one of them. Some alternatives include our traditional notions of teaching reading comprehension (the idea that readers simply read the words and remember key information), or reader-response in which the reader more tries to experience a text than to analyze it and which leads to more of a self-appraisal of the connection of the text to one’s own life and experience; or post-structuralist reads where one tries to read against the text, perhaps even using a particular ideological stance to try to blow up the text (including Marxism, feminism, Christianity, etc.). They all have different values and benefits, and they all overlap in some ways.


For example, in close reading it is important that you grasp what the events were or what the information was that was described literally or explicitly in the text (as well as grasping those inferences the author likely intended). Thus, if you read a newspaper editorial, it would make sense that you come away knowing the editorial board’s position on the issue and the reasoning or evidence behind their position. Or, if you were reading a story, you should be able to tell who the important characters were and summarize the plot. Or, if you were reading a scientific paper, you would need to grasp the multiple parts of the author’s theory or the steps of the experiment. However, while being able to comprehend that kind of information is important in close reading, it is important in every other kind of reading, too. That idea that one needs to  be aware of what the author is “saying” or “implying” is common to all kinds of reading. Thus, if a teacher has gotten the idea that since close reading requires that kind of learning, then I have always been teaching that and I’ll just keep doing what I have been doing.


There are other dimensions of close reading that are important or more helpful because they distinguish close reading from those other approaches. I think basically that there are two or three things that make close reading stand out from other approaches to reading. First, is the idea that what we need to learn and think about in a text is the information that is in the text itself. Many, perhaps all of the other approaches to reading, emphasize the use of other tools to make sense of the text. Close reading is hermeneutical, however. It focuses heavily on the analysis of what is in this text, rather than going outside to other knowledge sources to make sense of the text. For example, readers use background information to interpret what is in a text. Some approaches to reading emphasize background in reading comprehension (by getting readers to review what they already may know about a topic, to use background to make predictions about what the text is likely to include, etc.). Close reading asks one to be more of an anthropologist, setting aside what one might already know to try to focus more on the author’s words and structure. Or another example: it is not uncommon to try to read a text in the context of other pieces written by the same author. Thus, students might read several Hemingway stories with the idea of figuring out the threads that go from text to text (“Well, I wasn’t sure that in this story he meant that the woman was… but given this other story I’m sure what he must have meant was…”). Close readers eschew such tactics. They try, as much as possible, to set aside any information that is not in the text itself. (That even means close readers tend not to be big fans of book blurbs… information provided by a publisher about the text that might shape one’s understanding. The only information that matters in close reading is in the text itself.


A second dimension that is special to close reading is the idea that the way that an author chooses to convey information is as important to the message as the message itself. It is not enough to “comprehend” a text or to be able to summarize it or answer basic questions about it. A close reader will understand how the author has powerfully communicated the ideas and where that power emanates from. That means close readers not only read for that literal recall and low level inferences that have been common in reading instruction for 100 years, but they also want to think about the author’s word choices, and why particular images or ideas were juxtaposed, or who the telling affects the message. When you read The Great Gatsby did you notice the women’s names? Do you think it was an accident that Tom’s wife was named Daisy or was their some deeper relevance to that? The reason why symbolism, ambiguity, metaphor, juxtaposition, and other literary devices are so important in close reading is because author’s compose or construct texts to do more than just tell a story or describe an image. The author intentionally obscures information or “roughens the surface” of a text so that we have to dig for the meaning—so it can become important to us, so that we will remember it or care about it or so that we will grasp it more deeply. Good readers—in the close reading world—are those who are able to figure out those deeper meanings through a more careful analytical consideration of all the clues in a text. Close readers keep their eyes out for odd comparisons or peculiar choices of words or repetitions and so on. And, then they read and reread these passages to try to figure out how the text worked. (Perhaps this analytical requirement is a third dimension or perhaps it is just implied by the emphasis on how a text works).


Teachers who just teach kids to get more facts from the story or who do traditional reading comprehension lessons are not understanding what close reading is. I see lots of comprehension instructional materials that look like that too: they are labeled close reading, but they probably mean literal meaning or careful reading (getting some facts straight) or thorough reading (getting a lot of facts from the text). However, none of those descriptions have much to do with close reading.



  • FN: One of the most disparaged terms in teaching is “teacher talk.” Doubly so, with literacy classrooms.  Do you agree that teacher talk is bad? How would you think about it?


TS: Teacher talk is not only not bad, it is imperative. Our ability to motivate, to instigate, to explain, to critique and so on are most often going to be express through teacher talk and all of those are essential parts of teaching. The problem that leads to the disparagement of teacher talk, of course, is that there is so much of it (and I am a big offender in this). I suggest there are two problems with teacher talk: the first, being we talk too much not leaving space for student talk, student involvement, or even nonverbal demonstration or models provided by the teacher, or any other kinds of instructive action that doesn’t require or depend upon our yakking away. I try to combat this by scheduling places for kids to do the talking or by setting a time for the introduction (they have to be in the activity by 9:10, so that means I have to stop explaining by 9:05 so the kids will have a chance to ask questions and get started—or more accurately to get started and then to ask some questions).


The second problem is that many teachers do not speak effectively. There are studies showing, for instance, that it is possible to teach teachers to provide better explanations. With such training kids come away knowing more. I see many teachers who use too damn many pronouns, so many that kids struggle to follow what the teacher is explaining. When are examples needed? When will an analogy help or when will it block understanding?  Sometimes it is important to get in your listener’s face to make eye contact or to get close, but I see teachers sitting at their desks or rooted to the white board. I’d like to see better intentionally better or more effective talking (audio tape yourself with your phone and play it back

, , , , , ,

5 Responses to “A Few Minutes with Literacy Expert Tim Shanahan”

  1. Mike G
    September 24, 2016 at 6:22 pm

    Excellent Q+A.

  2. Camille
    September 27, 2016 at 8:20 pm

    Wow! This was a fantastic interview and it helped clarify ideas that were swirling around in my head. Thank you.

  3. Christina
    September 28, 2016 at 5:00 pm

    Great Interview with lots of things to think about! I would be curious to learn his take on teaching reading to Dyslexic students.

  4. November 3, 2016 at 6:49 pm

    Yes! The most important points for us were questioning the validity of leveling, whose research base is dubious though its popularity is widespread, and balancing reading strategies with the development of content knowledge.

  5. February 26, 2017 at 9:33 pm

    Christina I would second your question teaching reading to the dyslexic child

Leave a Reply