There is a common misconception, says Timothy Shanahan, that it doesn’t matter what children and young people read, “as long as they’re reading”.

The distinguished professor emeritus at the University of Illinois Chi­cago explains that the rationale for this is that “if kids are practising, it’ll make them better readers”.

This is something Shanahan believed himself in the past. Yet over the course of his career, he has changed his mind about how much it matters what children are reading.

“Practice is important, but what you’re practising makes a big difference, and the evidence suggests that there are a lot of text factors that really do matter,” he says.

A former first-grade (equivalent to Year 2 in England) teacher, Shanahan became director of reading for Chicago Public Schools and president of the International Literacy Association, as well as serving on the advisory board of the US government’s now disbanded National Institute for Literacy under presidents George W Bush and Barack Obama.

He has written or edited more than 300 publications on literacy education, including on the topic of why children should have access to more challenging texts.

At a time when the recent curriculum and assessment review has reignited debates about diversifying the curriculum and which books exam boards should make space for, we sat down with Shanahan to discuss how teachers can select and scaffold the right texts for their classes.

Tes: How much does text selection matter for teaching literacy? Many argue that the pedagogy is more important than the text you are using

Shanahan: Over a long career, something like that is a moving target. You start out thinking one thing and then, through experience and, in my case, knowing the research, you change your mind.

It’s one thing to allow kids to read what they want to when they’re on their own, and I’m in favour of that. But within teaching, you have to worry about the content. You want kids to be building up knowledge about the world, and in ways that we adults think would be useful to these kids in their futures.

You want the text to have language features that will help the kids become better readers. And so the texts matter greatly.

What types of text features can help to make pupils better readers?

When we take on a text, there are those that are in our wheelhouse: the author has aimed it at us quite appropriately; we have lots of shared knowledge with the author; we know the author’s vocabulary and so on.

But if you’re trying to make somebody into a better reader, what you want are text features that give their comprehension some problems. A simple example is rich vocabulary. We want vocabulary that’s going to help kids build up a more extensive lexicon in their head, because that’s going to open up more ideas to them.

Similarly, we want complex sentences, with multiple clauses, or lots of internal punctuation, or passive voice - all these things we do in written English to communicate and keep our text interesting.

Pupils will not understand some aspects of these texts, and the teacher has to probe to find that out, because the kids will rarely volunteer this information.

How should the teacher approach that process?

What we have traditionally done is ask questions designed to help kids practise the types of questions that will be on the exams, but it’s a lot smarter to ask questions that reveal whether kids are understanding things that you think might be challenging for them.

If there is a sentence I think is rich and that I think is going to trip the kids up one way or another, I’ll have a question ready that they could only answer if they comprehended the sentence.

If the kids can answer my question, that tells me either the sentence wasn’t as hard as I thought it was or they’re more advanced than I thought they were.

If they can’t answer it, I have an opportunity to teach them something. And over time, as they learn to handle that kind of sentence or deal with that kind of grammatical construction, other texts open up for them.

Text structure is another element. Or symbolism in literature. Or how do you connect the graphics in a science book with the text?

By putting kids in books that they can already read, we avoid teaching them this stuff. For the ones who figure it out by themselves, wonderful. But lots of kids aren’t figuring it out.

And it helps if you can show them something within a text, rather than saying, “Oh, I’ve got a nice worksheet for you. You’re going to practise this, but you might never recognise it again.”

Does that mean it’s better for pupils to read whole texts, rather than extracts?

I’ll preface this by saying that I’m a textbook author, and the textbooks I work on use short stories, articles, extracts of longer works and so on. That makes this question tricky for me to answer.

But there is also no research on the use of extracts.

If I had some research to say, “Wow, kids learn more if they read a book,” that would be one thing, but as it is, I think there are trade-offs.

As an English teacher, if I want kids to see how an author uses symbolism or sets a mood, I could get one text and we could read it for the next six weeks, or I could give them four or five different examples, pulling chapters from these different books and so on. My hunch is that the kids are going to come away knowing more of that text feature from the multiple exposures. But I don’t have any research showing that.

It’s certainly possible that if you have to read the whole book for six weeks, there’s a memory demand that means kids still have to know what we read last week, and that’s beneficial.

Unfortunately, what happens in a lot of classrooms is that teachers end up reading the books to the kids, which drives me a little crazy, but that’s the only way they can get through it.

With extracts, I can expose kids to more authors and give them more concentrated practice with certain aspects of reading, but then you don’t have that notion of hanging on to an idea for a more extensive amount of time.

It’s possible that any of these things might matter. So my advice to teachers is to do a mixture, given that we don’t know.

What about using texts that have specifically been designed to help teach reading?

I’m not a big fan of writing text with the purpose of teaching. Decodables are OK, but only for a fairly brief amount of time, really early on. I don’t have a problem with that, as long as kids are reading other things simultaneously.

But to show the kinds of features I was just describing, I’d much rather the kids be reading something real that has some real value in terms of its content, over creating a text that’s going to show those things.

You mentioned being frustrated by seeing teachers reading books aloud to the class. Why do you see this as a problem?

I’ve been in schools in a professional way now for more than 55 years, and what I see is that if texts on the curriculum or exam syllabus are going to be hard, what most schools do is try to find easier texts to put the kids into, instead of teaching them how to handle the difficulty.

The alternative is that they read the text to the kids, or they don’t use the text; they just tell the kids what they need to know. I even see this in literature classes, which is shocking to me. I hear, “We can watch the movie,” as though it’s the same thing.

Those options solve a problem for the teacher: how do I get through this lesson? But they don’t have the outcome of making kids better readers; of helping them to be more able to read a text like that - or more likely to read it.

There are places for reading to kids; I certainly read to the kids as a teacher. There are even places for watching the movie - but not in place of the kids reading themselves.

If we do that, we’re cheating them. We’re undermining kids’ ability to learn to do what they need to do. And that’s really unfair.

If you’re teaching maths, you want the students to be doing some maths, not just watching their teacher doing maths on the chalkboard.

There has recently been debate about the need to diversify the curriculum. Do you think it is important for pupils to be exposed to diverse voices and to see themselves represented in the texts they read in class?

This is another area where there’s no research. There is lots of theory, and the theory is very strong that it really matters, although I don’t know if it matters in terms of learning to read.

I did a study once where I was studying how the kids thought about authors, so I was sharing text with them and then asking them questions. I didn’t tell them anything about the author; I just read the text, then interviewed the kids, asking things like: “Who wrote that? Was that a man or a woman? Were they Black or white?”

There was this one African American boy in seventh grade (so around 12 years old) who was just adamant that this particular book had to have been written by a white man, because it was “well written”. I messed up this kid’s data for the study because I showed him that it wasn’t, and he was stunned.

So I do believe this matters on some level. It’s not necessarily a “learning to read” issue, but it is an issue of self-development and of kids trying to understand where they fit in society. I do think it’s important that kids see themselves in what they view, what they read and what they see adults being capable of doing around them.

Outside the classroom, is it important to give children autonomy over what they read?

I think we should be letting kids make their own choices when they are reading for their own interest. This is one of the problems I have with “book levelling”, in that a kid might go into a library and have someone tell them they can’t take anything off the blue shelves because those books would be too hard for them.

My advice is: let the kids try.

Don’t be afraid to suggest books to them. Of course, there are certain topics that we protect children from, and that’s fine, but beyond those obvious things, let the kids choose.

Does that include letting children choose to read things that aren’t books, such as comics or magazines, for example?

I think magazines can be a great way to get kids interested not just in reading but in the world. There are some wonderful magazines out there.

The one concern I have is when we get into the reading that isn’t reading. I get letters from parents all the time about how their child will only read graphic novels. And I think you could argue that you want the child to have wider exposure than that.

The problem with comic books is they just don’t present kids with very many words. Sometimes young people like reading them because it means they don’t actually have to read.

What advice would you give to a teacher selecting a text for their class?

The first thing I would focus on is content: what’s the quality of what I’m going to share with them? For example, are you trying to expose them to a certain genre or a certain body of information? That would be my first criterion.

The second one would be, is the text hard enough? Is it going to trip up enough of my kids? Will there be something to learn and something for me to teach? I don’t want it to be something they can already read well.

A third thing I might think about is whether it’s me who has to make the absolute choice of text. For instance, if it’s about genre and I want a certain kind of story, can I get two of those?

If I am able to go to the kids and say, “I’ve got these two terrific stories and I don’t know which one we should read; what do you guys think?”, that can be a really good thing. Research shows that giving children agency helps them take on harder texts because they have had some say in it.

And then, once the choice is made, I need to identify those things that I think might trip the kids up. Where are the problems going to be, so I can question them in a way that will tell me what to teach?

It’s also not unreasonable for the teacher to say, “Gee, we haven’t read any non-English writers lately. Maybe for the next text I pick, I’m going to pay attention to that,” and I’ll bring the kids’ attention to it somehow.

Those are the kinds of things that you have to think about. And there are always going to be trade-offs.

Is there anything else teachers or subject leads need to consider?

The idea of selecting one text always scares me a little bit. The notion that you just read one thing and then get on to reading something else doesn’t seem as powerful to me as selecting texts (whether whole books or extracts) that are connected somehow, because they’re going to give pupils five different examples of a genre or three different treatments of some information.

It’s that notion of not just reading one thing, but reading a series of things that - through content, style or literary feature - helps you learn about the others.

It would be easy to imagine a class reading 26 really good things across the year, none of which had anything to do with each other. And that might not be the most powerful experience for kids.

I still want you to pick 26 really valuable things, but can you find some reason why you would read those right in a row or work on them simultaneously? I think teachers need to keep that in mind as well.