Why Comprehension Instruction Goes Bad

Last week, I posted a blog that described how effective comprehension strategy instruction works. I said that students won’t use strategies forever and that I didn’t believe that strategies eventually morphed into skills (at least not skills that look anything like the strategy). I think strategies work more like true scaffolds; they operate as a temporary support that allows kids to read on their own more effectively, but not in the same way they will need to read on their own later. The problem is that strategies are cumbersome and no one will use them for long. Frankly, when a student is reading with a teacher, he or she does not need to use strategies because the teacher will do the scaffolding (by previewing the material, setting reading purposes, breaking the reading into pieces, and following up with questions and discussion). Strategies can play this role temporarily when kids are reading on their own, but eventually they have to have enough experience in reading and understanding challenging texts so they can handle it without strategies.

The problem with comprehension instruction is twofold: on the one hand, teachers do not teach strategies very well and so lower readers tend to find independent reading difficult, even when they are receiving strategy teaching. Furthermore, most comprehension instruction (whether aimed at strategies or not) is pretty weak gruel. Students can escape with a pretty shallow processing of text as they may be guided by weak questions, superficial discussions, and no real engagement with the difficult or subtle aspects of the text.

Good comprehension instruction should push kids to think more deeply about a text than they would if they were reading on their own. In stories, kids struggle with character motivation and psychological elements; these aspects of a story may not be stated explicitly and they require an understanding of how human behavior works. Kids tend to focus on behavior itself rather than thinking about underlying reasons for the behavior (they struggle with this in life, too). Good comprehension instruction leads kids to think about those aspects of a text that they might normally neglect.

Good strategy instruction is tougher. It both has to lead kids to a deep processing of text (so the discussion of character motives had better take place even when you are teaching kids to summarize), but this teaching also needs to guide kids to develop intentionality, so they will try to think about the text even when they are reading on their own. If a teacher pushes strategy teaching too hard, kids won’t get much guided reading practice with the teacher. However, if a teacher pushes the story too hard, and ignores strategy teaching, then low readers will find it nearly impossible to read independently, and they will not be likely to progress in reading.

If teachers can run a good discussion that guides kids to think about the important ideas in a text and to come away with a real understanding of what the author said and how he said it, then the students will have a real chance of becoming good comprehenders. But this will often be insufficient for poorer readers, because they usually need more of this than the teacher can provide.

Strategy teaching is meant to be a lifeline for such students. If a strategy gets kids to interact with a text more than they would on their own, then this could help strengthen them in the same way that participating in a discussion group with the teacher does. If we can afford to provide substantial doses of teacher guidance to reading comprehension, then strategy instruction won’t be necessary. If we can’t afford this, then providing what we can and supplementing it with strategy teaching is the way to go.

A couple of other things:

Last week, Drinda raised some questions about my reading comprehension column, her questions and my answers are below:

1. Are the gains made by students who benefit from learning strategies sustained once they stop using the strategies?

If the students engage in a sufficient amount of reading on their own in which they use these strategies, then they should read better even when they are no longer using the strategies. (That’s the theory anyway. We have no long term studies of this.) The difficulty is to know how much meaningful, engaged reading (under the teacher’s supervision and on their own) that they must do to accomplish this goal.

2. When they are using the strategies, do the use them with all reading for a while--or only when they specifically choose to use them (or are prompted by a teacher)?

Ideally, we would want kids to use strategies a lot, so I’d say all the time. But research indicates that kids don’t use strategies much. That is partly due to the low quality of much strategy teaching and partly because it is hard for poor reader to sustain a reading conversation in their heads when they are new to this. We’d like kids to use the strategies especially when they are reading without the teacher.

3. If these are of most benefit to low performing students, should they only be taught for small groups?

The benefit of strategies is mainly for low readers. They don’t hurt the high readers, but they don’t help much either. Grouping for comprehension instruction tends to be a good idea (both because it allows you to vary text difficulty based on student needs). However, whether strategies are part of the mix or not, it is still important to engage kids in meaningful reading with interactions with others.

Finally, I will be posting no new info here for the next two weeks. Cyndie and I are off on vacation. I'll put some new material up in mid-July when we return. Enjoy your summer.

Why Teach Comprehension Strategies?

There is no doubt research shows that reading comprehension strategy instruction works. The National Reading Panel said so. Although comprehension studies have been short-term, there are just so many of them (more than 200 such studies).

That doesn’t mean everybody agrees with strategy teaching. Isabel Beck and Margaret McKeown have argued strenuously against such teaching. They claim teachers would be better off having kids read text and engaging in deep discussion of the ideas.

I respect Isabel and Moddy, but how can you ignore so much research? I think the disagreement lies in a basic misconception about the purpose of strategy instruction.

There are three ways to think about strategies—but only one of them can be right.

One view is that strategy teaching leads kids to actually use the strategies while they read. And that they'll then use them for the rest of their lives—reading better right into the grave. This simplistic notion is not held by any researchers I know, but some teachers buy it. This is part of what gets a negative response from Isabel and Moddy (let’s face it: good readers know they don’t usually use strategies).

A more widely held view is the one most research big-shots accept. David Pearson, Scott Paris and Peter Afflerbach claim strategies are just a stage of learning. Their idea is that strategies eventually morph into become skills. They are kind of phonics of ideas. You learn phonics, but decoding eventually get so skilled that you read without thinking about decoding. So, the idea is that comprehension strategies do continue to be used by students, but without conscious awareness.

I don’t know about you, but I find it hard to believe that I’m predicting, questioning, and visualizing away, and not even knowing it.

And yet, unlike the strategy critics, I teach comprehension strategies.

Strategy teaching works… but mainly with low readers. Consequently, I think of it as a kind of “pay-attention-and-think-about-the-text” instruction. It helps poor readers to do something with their minds during reading, which is incredibly important. Isabel and Moddy are right about teachers guiding kids to think deeply about text. Strategy teaching, when it is good, just provides a kind of scaffolding that allows kids who have trouble thinking while reading to do so, even when they read on their own. Strategies are temporary, but not because they become automatic skills, but because they are too cumbersome to sustain. The benefits of strategy teaching decline once reading is able to stimulate normal language processing responses.

If the “Shanahan surmise” is correct (that strategy teaching is beneficial because it provides a temporary scaffold supporting independent thinking during reading for kids who can’t do such thinking automatically), there are implications for teaching reading comprehension. I’ll deal with that in my next column.

RtI on Steroids, or Why I Believe in the 9-Tier Model

The latest rage in the schools is RtI. Special education money (about 15% of it) can now be used for improving classroom instruction and installing preventative intervention programs. I'm a big fan of this movement for several reasons: First, because the best way to determine if someone has a learning problem is to offer really good teaching and if the struggling continues then you know. Second, special education programs simply haven't worked very well for most kids, and the learning disabilities label has been over applied, and those programs are getting expensive.

But even though I like RtI, I have problems with it (as do others, perhaps most notably, Dick Allington--however, his problems emanate from concerns about who will deliver the prevention services). My concern is that RtI is often so mechanistic that nothing good is happening for the children. Schools buy an instructional program and/or a regimen of professional development and they think they have a good Tier 1 response... they set up a reading class a couple of times a week for groups of struggling students and you can check off Tier 2 as well. That won't work and so if RtI is going to cut the case load in special education schools are going to have to be real aggressive about meeting students' learning needs in reading.

This weekend I met with a group of educators from across the country in Santa Fe and I told them about my 9-Tier Model. The first reaction was 9 tiers instead of 3, is this guy crazy? However, once they saw what has been missing from their 3-tier plans, they were more than willing to consider building up their efforts (not by going to 9 tiers, but by implementing a richer set of responses across the three tiers they have been doing). Good for them--and, more importantly, good for the kids whom they are responsible for. Here's my powerpoint on the 9-tier model:

http://timothyshanahan8.googlepages.com/rtionsteroids