Teaching with Complex Text: Haven't You Ever Heard of the ZPD?

  • text complexity challenging text zone of proximal development (ZPD) complex text
  • 22 August, 2020

Teacher question:

I’ve read what you’ve written about the instructional level. You claim that there is no such thing. Haven’t you ever heard of the “zone of proximal development (ZPD)?”

Shanahan responds:

I’ve heard of it, but if you think what I’ve written is contradictory to it, then I suspect you don’t really understand the ZPD construct or its relationship to this aspect of reading.

Let’s start with the “instructional level” idea first.

A century ago, it was common practice for reading teachers to place children in different reading books based on their abilities. For instance, one Wisconsin survey from 1918 shows that the majority of teachers were grouping children by book placements; this was the old bluebirds, redbirds, and crows plan. It was also more than a decade prior to Vygotsky’s ZPD and about three decades before the instructional level was operationalized.

In the 1920s and 1930s, various reading educators (e.g., Donald Durrell, William S. Gray) referred to the idea of their being an instructional level – that is an optimum student-book match that would promote maximum learning; but these references were informal, much like the actual classroom practices being used at the time. There was not any very explicit theory of it, and there were no criteria for determining what a good match would be… just the vague idea that texts could be too hard to support learning.

This changed in the 1940s with the publication of Foundations of Reading by Emmett Betts. Betts made explicit the theory of the instructional level (along with the independent and frustration reading levels). He also endorsed a specific measurement scheme that would supposedly allow teachers to place their students properly; that is, in texts that would lead to the greatest amount of learning.

Throughout the 20th century, university-based reading experts “tsked-tsked” about teachers who were not placing their students in easy enough books, because, of course, there were still many teachers who just used a single grade level reader without adjustment.

Lev Vygotsky’s work on the zone of proximal development took place in Russia during the late 1920s and early 1930s. However, this work had no impact whatsoever on American education until the 1960s when it was first translated to English.

Basically, the ZPD idea was Vygotsky’s attempt to cast learning into a social context. He was reacting to earlier schemes proposed by other psychologists (Piaget, for instance) who had treated learning as more of an isolated, individual pursuit. According to Vygotsky, there is a social transaction that takes place between a learner and someone more knowledgeable or skilled that leads to learning; that is, this series of social interactions take someone from a state of not being able to do something to being able to do it.

An example might help.

Mom wants Junior to learn to tie his own shoes. She says something like, “Junior, let’s tie your shoe. Watch. First, we cross the strings like this, and then we tuck this end under and pull it tight. Next we make bunny ears and then cross them and put this end under. Pull it tight again.”

Now, mom might be stressing that “we” are doing these things, but that is just a social device. Junior really cannot tie his shoes yet. This social interaction is getting him to participate in the task by watching closely under her direction.

At some point, mom changes this routine. “Junior, what do we do first when we tie our shoes?” The child responds that you have to cross the strings and tuck one end under, and she tells him to do it, he does, and she completes the task for him. Over several such episodes, Junior takes over more of the task until it is mastered.

Initially, mom knows how to tie shoes and Junior does not. Over time, however, Junior becomes able to do it, too, as a result of mom’s demonstration, accompanying language, explicit guidance, and so on. These social supports are the basis of learning in Vygotsky’s theory (and these supports are what Jerome Bruner and company later described as “scaffolds”). The scaffold metaphor is an apt one – notice that the shoes always end up getting tied – with mom providing only as much support as necessary.

ZPD refers to that space between when the learner can’t do some task at all and when he/she can do the whole thing independently.

Vygotsky died before he could fully rough out the implications of ZPD, but it is fair to say that it was a startlingly different concept than what Betts’ (and later Irene Fountas, Gay Su Pinnell, & Lucy Calkins) conceived the instructional level to be.

According to Betts, students would make the greatest reading gains when a text was matched to the child’s reading level. If the child were placed in an easier or harder text, then less learning would result. This makes the learning zone very narrow and very specific; you are either in it, or you don’t learn. But Vygotsky’s notion was much broader, more supple and dynamic; Junior was learning even when the only role he could play was to follow mom’s actions with his eyes.

The mechanistic and automatic quality of the instructional level is much more consistent with the kind of individual learning psychology that Vygotsky was challenging (with his more social teaching psychology).

That’s why those who embrace the instructional level idea are so gung-ho about independent learning. They encourage minimal teaching, because if the book placements are right then there should be little for the students to figure out and they should be able to do much of that on their own.

However, if you recognize that the ZPD is not just a narrow band of learning territory, but the entire distance between where the child is now (in terms of concepts and skills) and what full proficiency would be, then placing students in grade level text won’t seem so odd. We’re both trying to work in the ZPD, but instructional level fans want to zero in on a particularly slender slice of it.

And, if you accept that the ZPD can be so varied and expansive, then Vygotsky’s relative valuing of the social interaction that is teaching over the idea that students will learn mainly through their own individual explorations should make more sense, too.

Students who are working in a text that they can already read pretty well (the skinny little instructional level slice of the ZPD) will need considerably less scaffolding than those grappling with a relatively harder text.

The instructional level approach tries to minimize what needs to be learned in the hopes that kids will be able to figure those things out by themselves. By contrast, the idea of teaching with more complex text aims to expand the role of teaching in order to maximize the amount of student learning.

Aside from those differences, both approaches toil within Vygotsky’s zone of proximal development.


See what others have to say about this topic.

Marianne McCormick Aug 22, 2020 05:04 PM

So would you suggest that guided reading lessons incorporate more difficult texts-frsutration level-and independent reading time students would read books at their level? As a high school reading specialist, I find it difficult to suggest levels because students should be at a level z or beyond and giving a ninth-grader a books that is at his/her level, may be insulting to students. Any thoughts about this?

Sandy backlund Aug 22, 2020 05:24 PM

Bravo for shining a light on Vygotsky....may we all build scaffolds for students
to build strategies and “reading muscles” for challenging text! Leveled texts build necessary fluency and confidence and are just the shoestrings.....

Janet Rehmert Aug 22, 2020 06:39 PM

All we have to do to see leveling books doesn't work is to look at the reading scores of the students of the United States for years and years and years. We now have 50 years of reading and brain science. It's time we use it, and help kids read. Reading programs need to get on the research train.

Sharon Aug 22, 2020 06:41 PM

This made me think about read alouds which are typically well above independent reading level for early readers. Should we be displaying/highlighting the text for students as we complete the read aloud? Would this be beneficial? This could be done easily now that we are virtual.
I’m worried about the limited time that I will have with students. I want to make sure that I’m making the most of the time I have with them.

Tim Shanahan Aug 23, 2020 01:06 AM

Marianne—indeed, instructional is meaningless at high school. Studies are showing kids do as well or better with grade level texts. Look at my publications on my site, print and PowerPoint and you’ll get some guidance on this.



Tim Shanahan Aug 23, 2020 01:08 AM


I wouldn’t go there. I think the oral interaction around those complex text read alouds would be best.


Thomas Franklin Aug 23, 2020 05:38 AM

I agree with the aspect of highlighting particularly difficult sections over time and scaffolding methods of reading, analysis and thinking routines. We use on the line, between the lines and beyond. This is effective with a wide range of text levels. There is always learning, sometimes it is simply independent thought about meaning, other times it is structure, vocab, grammatical features.
Great article, thank you!

Sue Testerman Aug 23, 2020 04:30 PM

Like Natalie Wexler wrote in The Knowledge Gap: "Leveled books lead to leveled lives."

Cez Aug 24, 2020 03:04 AM

Am I correct in associating the concepts of ZPD, scaffolds, and instructional level to comprehensibility input plus 1?

Amy Rountree Aug 24, 2020 02:29 PM

Re: text levels and high school
Many texts that high school students would not be ashamed to read are technically well below level Z. It is always interesting to run texts through the Lexile program on Scholastic's site. I became very skeptical of the text leveling system when I realized that, based on text complexity, the Twilight series (830 Lexile, or Level T) is at a lower level than Diary of a Wimpy Kids series book (950 Lexile, though also a T according to Guided Reading Levels). There's very little hard science behind calculating these levels -- under any system you use. Scholastic uses a strict number of words per sentence, number of syllables per word algorithm. But those are not the only things that make a text complex. I think you have to think about background knowledge and topical interest. I do not know how the Guided Reading Levels are calculated. I think that, once you have most of the Phonics basics down, the level doesn't matter as much as the background knowledge, interest level, scaffolds, etc.

Tom Gunning Aug 24, 2020 05:09 PM

Betts' instructional level had a fair amount of flexibility. Instructional word recognition ranges from 95 to 98 %. For a 500-word selection that amounts to between 15 and 25 problematic words. Betts also distinguished between immediate and basic instructional levels. An easily discouraged student might be placed at a level where her difficulty was minimal. Once confidence was built, the student might be moved up to the basic level, where up to 5% of the words were unknown.

Tim Shanahan Aug 25, 2020 01:11 AM

The problem with that is that he made it up, and didn’t evaluate its effectiveness. That’s why studies are now finding that kids learn more from harder texts. Betts May have been flexible but it was in a very narrow range that basically meant the student could already read a text well.


Gream Lee Sep 11, 2020 08:14 AM


Lise (pronounced Please without the P) Sep 12, 2020 01:22 AM

It's always a joy to read your blog. You make me think; challenge my beliefs. I don't always agree but I sure appreciate the learning opportunities I get from your thoughts.

I absolutely love your example of teaching a child to tie his shoes (I will borrow it, if you don't mind). I teach teachers the fundamentals of the French and English orthographic systems in Quebec (all kids learn both). I sometimes have a difficult time convincing teachers to guide their students, not just let them "discover" the word or make them read it over and over again until it sinks in. For LD kids, it doesn't. But it sure contributes to their disdain of reading.

French is harder to write, English is harder to read, and together, they are the most opaque european writing systems (lucky us!), and over 20% of them are struggling, one in five! I always wonder what is being taugh to teachers when I have to explain that writing the instructions for LDs, as precise and clear and colourful and cute... as they might be, is quite useless for students with READING difficulties.

I'm convinced your example will help some of them see the light! ;-)

Thank you.

Stay safe!
Ca va aller! (soft C)

(Website is in French only, sorry)

Tony S Sep 13, 2020 03:02 AM

Can you explain the difference between your explanation of ZPD (lower bound in difficulty of tasks is 'can do on own' and upper bound is 'proficiency') and the one that's often taught today, where the lower bound is, 'can do on own' and the upper bound of difficulty is, 'can do with help?' That's where I'm getting stuck. I think it has to do with whether a skill is viewed as a discrete procedure that can either be successfully completed or not (like tying a shoe), as opposed to a progression of increasingly complex procedures (like learning to knit). Can you help?

Dr. Gwen Lavery Sep 20, 2020 09:32 PM

This is so true! I had teachers to teach more complex text in guided reading, and we moved from a D to B. Within the guided reading, we provided a mediated learning experience and reciprocal teaching!

What Are your thoughts?

Leave me a comment and I would like to have a discussion with you!

Comment *

Teaching with Complex Text: Haven't You Ever Heard of the ZPD?


One of the world’s premier literacy educators.

He studies reading and writing across all ages and abilities. Feel free to contact him.