BigConceptsinDI < Kitchen

The Best Model on How to Teach Children (and Adults): Direct Instruction (DI)

We have been doing a lot of floundering the dark lately at KTM trying to figure out what works in education, what doesn’t work, and why. The one thing we lack is a solid benchmark for an effective instructional program. So let’s put this floundering to an end right now by setting the benchmark for one of the most, if not the most, effective and efficient educational programs out there: Direct Instruction (DI).

(NB: This post will be the foundation work for understanding DI. In later posts I’ll describe DI’s research base, its underlying philosophy, and put some more flesh on the skeleton. DI is just too big a topic to cover in one post)

The best metaphor for DI is a staircase -- a staircase that begins in the pre-K level (for low performers) and ends at about the sixth to eighth grade. Each step in the staircase is the foundation for subsequent steps. Each step is constructed so that it is in reach by a student on the previous step within about 30-45 minutes of instruction time.

Each student is initially placed on the staircase based on his ability level. The student must have the knowledge and skill repertoire needed to take on the next step and reach it. Every student firmly on a particular step is the same with respect to the program sequence even if they are not the same age, learn at precisely the same rate, have equal intelligence, or exhibit the same style of learning.

Now we’re ready to see how the DI staircase is constructed and why the staircase works so well.

Big Concepts of DI

Student Placement: It is critical that students are placed properly in the instructional sequence. Placement is ongoing and always referenced to student performance. Major regrouping is done at least three times per year in K-3; less frequently thereafter once the implementation has stabilized. Initial placement is done with a 5-10 minute placement test.

Rule: Students are placed where they already know 70% of the material introduced for the first time.

Students must be familiar with most of the material; it must not be entirely new. Placement is where students make some mistakes but not too many. Students should not be struggling with the material; they should not be placed at the edge of their knowledge. When students work too hard, they make lots of mistakes, and making lots of mistakes means the material won’t be mastered.

Application of Rule: On day one of kindergarten, students should not all be started with the same lesson as they are in traditional classrooms. This is the first big mistake of existing teaching programs. Students should be grouped so they are familiar with most of the material that will be taught in their first lesson. This may mean that some low performers are initially placed as many as 300 lessons behind their middle-class peers.

Frequently, low-performers need to be explicitly taught skills, like how to behave and follow directions, that we take for granted in their middle class peers. Nothing is taken for granted in DI. For example, introductory DI lessons teach students how to touch a letter on signal in their book because once phonics instruction begins they will need to point to each letter in their book and then sound it out. If the student isn’t firm on touching on signal, the activity becomes sloppy and the students become confused, hindering learning.

Take a look at the Effective Behavior Management, Use of Praise, and How to Set Up a Reading Group Carefully (especially the end) videos on this ADIHome page to see some of the pre-skills that need to be taught and how the classroom is set-up for effective student management before effective instruction can begin.

Firming: Material introduced in previous lessons and repeated in this lesson is being firmed.

Rule: Students must be at least 90% correct on material being repeated from a previous lesson.

This firming process is really just spaced repetition or practice which we know is necessary for mastery. Gifted students need 8-12 repetitions for mastery. Average performers need 25-35 repetitions. Low performers need even more. The amount of practice each group needs for mastery dictates the rate they can progress through the program. The amount of practice given is keyed to student performance.

For example, average students usually receive two periods of English language arts (ELA) instruction per day to complete the program by about the sixth grade. High performers move at a faster pace (two to four times the average rate) and once they get to about a year ahead of the average group, they typically receive only one period of ELA instruction. Low-performers, on the other hand, move at the same pace as the average performers, but they typically need some extra firming sessions and may need time to repeat some sessions. Otherwise, they move at the same speed through the material as the average students.

Take a look at the Why Correct Errors Immediately and Why Teach to Mastery videos on this ADIHome page to see the firming process in play and the importance of error correction during the firming process.

Mastery: Students must master all tasks and activities by the end of each lesson.

Rule: For material taught this lesson students should be 100% firm on all tasks and activities by the end of the lesson.

The firming process is repeated for at least three lessons until all students are firm with the material. Nothing is considered mastered unless it has been repeated in three lessons. Once mastered the material is continually used in problem solving and other tasks. In this way, students understand that they are expected to retain and use material they have learned.

Automaticity: Material is not considered to be mastered until the student can automatically recall it. Automaticity is determined by first time performance. If students are making many errors on their first time performance, they have not yet mastered the material and more firming is needed. Subsequent correct responses are not as good an indicator since these responses may only indicate that the student knows the information now, after they’ve been corrected. Automaticity comes when the student knows the material as well as he knows his own name.

Error Rate: The students' error rate must be sufficiently low that the lesson can be completed in the allotted time. Since the lessons have been carefully scripted in advance and field tested, it is known that the material can be adequately learned by even the low-performers in the allotted time. So, an error rate that is too high usually means that the students are not placed correctly in the instructional sequence and/or the material is not being taught correctly.

High error rates indicate that students have not and are not learning the material properly. Even many repetitions won’t solve this problem. If excessive repetitions are needed, the students tend not to retain the material as well. This applies to low performers as well as high performers.

Take a look at the Why use Scripts video on this ADIHome page.

Pacing: pacing should be as fast as possible. Students should be moved as quickly and reinforcingly as their performance permits. If one year’s lesson is completed early, the class should begin the next year’s lesson without waiting.

Frequent Feedback: In DI, students’ thought processes are made overt with constant monitoring and frequent responses. This information is then used along with frequent delayed testing, every two weeks, to determine if students are mastering the material and whether remediation may be needed. The teacher also uses this feedback to judge whether the presentation rate is appropriate.

All Students Held to the Same Standard: In DI, all students learn the same material regardless of their ability level. The only difference between the ability groups is the amount of practice or firming needed to bring each group to mastery.

Homogeneous Grouping: Think of these as temporary skill groups. Students at the same level in the program are treated the same. If students are at the same level, they are sufficiently familiar with the material to understand what will be taught in the upcoming lesson. In addition, the errors the students make will be similar, so corrections will be easier to effect. Homogeneous grouping is, thus, necessary so that communication between the student and teacher is clear and productive.

These are the big concepts of DI. These are the reasons why I believe DI is such an effective and efficient instructional program. In my opinion, the ultimate reason why DI works is because it recreates the learning environment of a high–performing student (the only students who really learn much in other programs) for average and low performers. Think about it, high performers succeed because they know a lot going into school and they have the cognitive ability to comprehend what is being taught no matter how poorly or inefficiently it is being taught. They never really fall behind and can always keep up with the pacing. DI recreates these conditions for the rest of the student ability curve.

Here is an exercise for the readers: Pick your favorite pedagogy or instructional program (tracking, discovery learning, social promotion, heterogeneous classrooms, traditional math programs, constructivist math programs, spiral curricula, et al.) and compare it to DI and see how many of the big concepts the program violates. When you do this it is clear why so much that is done in education today is so ineffective.

Follow-on Posts
So Why is Scripting Used in DI?

-- KDeRosa - 16 Nov 2005

Comments

After entering a comment, users can login anonymously as KtmGuest (password: guest) when prompted.
Please consider registering as a regular user.
Look here for syntax help.

It makes excellent sense to me to give the poorer students more practice on the areas they are having problems with.

I do wonder though, if the high performers are cut back to 1 period of ELA a day once they're a year ahead, what do they do for the other period that day?

-- TracyW - 16 Nov 2005

I believe they were cut back because in low SES schools there is a 25% mobility rate, so if they advanced too far the class would be very small. So instead of concentrating on pure acceleration, either the pace was slowed or the students got a free period. I have no idea what the kids did during that free time. I can think of many good uses for that free period -- more math problem solving, foreign language study (like Latin which will help with vocabulary, a course in logic, etc).

-- KDeRosa - 17 Nov 2005

Wow!

Fantastic!

-- CatherineJohnson - 17 Nov 2005

Thanks so much for doing this!

-- CatherineJohnson - 17 Nov 2005

I'll put a notice up front as soon as Carolyn's birthday 'card' has been there for awhile....

-- CatherineJohnson - 17 Nov 2005

low-performers need to be explicitly taught skills, like how to behave and follow directions, that we take for granted in their middle class peers

Actually, middle class kids have lousy skills in this area, too.

Apparently in Japanese schools they spend quite a bit of time directly instructing the kids on how to handle the classroom—how & where to store their belongings & supplies, and especially how to transition quickly and smoothly between activities.

They spend a lot of time learning & practicing transitions.

I just looked up Stevenson & Stigler's figure: U.S. 5th graders spend only 65% of their time in academic classes actually doing academics.

-- CatherineJohnson - 17 Nov 2005

Students are placed where they already know 70% of the material introduced for the first time.

Wickelgren's rule of thumb for accelerating your child was to get him through the entire grade-level curriculum plus 30% of the next-year's curriculum (i.e. 30% of the curriculum of the accelerated class he would be entering).

Christopher and I slaved away trying to do this, and didn't make it.

But we did get a BIG leg up.

-- CatherineJohnson - 17 Nov 2005

Of course, that makes it even more horrifying that he bombed his Chapter 2 test, given that he was the only kid in the class who'd had a lot of experience with the properties and with solving simple equations.

sigh

-- CatherineJohnson - 17 Nov 2005

Gifted students need 8-12 repetitions for mastery. Average performers need 25-35 repetitions.

Interesting.

-- CatherineJohnson - 17 Nov 2005

Why is this a foreign concept?

-- CatherineJohnson - 17 Nov 2005

This is cool, Ken! I've been waiting for this page!

-- CarolynJohnston - 17 Nov 2005

In my opinion, the ultimate reason why DI works is because it recreates the learning environment of a high–performing student (the only students who really learn much in other programs) for average and low performers.

That's not what you said the other day!

Remember, you were talking about 'rough mastery.'

I do think this idea is interesting and, 'roughly,' true. (Engelmann talks frequently about slower learners making exactly the number of mistakes they should make given where they were placed & the amount of practice they were given.

But I don't think you can turn it around; not exactly.

The brainy kids are skimming the surface somehow.....it's not as if they're getting an efficient & effective education, either.

Actually, I'll stick with 'rough mastery' for those kids.

It's true they don't fall behind.

But they aren't learning to mastery, either.

Somehow, they manage to hop, skip, and jump through K-12 until they crash and burn when they finally reach a course they can't hop, skip & jump through.

-- CatherineJohnson - 17 Nov 2005

This is great!

-- CatherineJohnson - 17 Nov 2005

Does he do any work on middle school & above?

-- CatherineJohnson - 17 Nov 2005

Ok, here's what I said:

My take on this is that the reason why high-performers achieve so well is that because they are so smart they are able to stay on top of the material no matter how poorly or quickly it is presented. In effect they achieve a rough mastery of the material, as their high grades attest. These kids get mastery learning from the traditional curriculum.
In constrast, the average and low performers never master the material. Knowledge stays at least party to mostly inflexible and the accumulation of imperfectly learned material eventually impedes further learning in a reasonable period of time.

I think this is fairly consistent with my new grand theory of learning. Let's see if I can better explain and/or harmonize the two.

By rough mastery I meant that high performing students are able to stay on top of what is being taught without getting lost. They keep pace. They do master at lesat some of the curricula just by solving problems in the classroom and in homework and through the shear repitious nature of the slow spiral. Plus, their higher cognitive ability can compensate somewhat. A fair number of high perfomers are able to successfully complete AP level calculus by the end of high school even without mastery learning (though they probably haven't reached their full potential). many of these kids have the ability to succeed in a rigorous college curricula too.

I suppose this might still be possible in the better constructivist curricula and through supplementation. It's just a very inefficient use of time.

Under DI, low and average performers get the same type of learning environment in that they are now also able to stay on top of the material like the high performers. Arguablly, they are receiving a superior education (at least a much more efficient one), though their lower cognitive ability reduces some of the benefits. (After all, I believe they tend to decrease in performance somewhat once they re-enter a more traditional curricula).

-- KDeRosa - 17 Nov 2005

I think there is a middle school and higher DI curricula for some non-remediaol subjects, but I haven't located any commercial versions.

Doesn't KUMON go all the way into calculus? SO it is possible to design such a curricula.

I was reading earlier today that Engelman actually designs his curricula by starting at the end and working backwards so hew knows exactly what he needs to achieve his goal.

-- KDeRosa - 17 Nov 2005

Ken, it seems Kumon goes past calculus and on into probability and stats.

The part where you describe the low performers learning to touch the letters, and to sit quietly and pay attention, sounds very much like the earliest lessons in Applied Behavioral Analysis, although I don't recall the condition that the kids learn to 90% mastery before moving on to the next lesson. Still, there are strong echoes of ABA here, up to and including the scripted aspect of the lessons.

That's a good thing. ABA is a godsend.

I really look forward to hearing more specifics about DI (as told by you). Frankly, your descriptions go down much more pleasantly than Engelmann's! ;-)

-- CarolynJohnston - 17 Nov 2005

I just told Christopher's math teacher about your page.

-- CatherineJohnson - 17 Nov 2005

Arguablly, they are receiving a superior education (at least a much more efficient one), though their lower cognitive ability reduces some of the benefits.

Yes, I think this is what I'm focusing on.

I think average & low-ability kids get a better education with DI than high-ability kids who can keep pace get with constructivist & traditional instruction.

-- CatherineJohnson - 17 Nov 2005

I was reading earlier today that Engelman actually designs his curricula by starting at the end and working backwards so hew knows exactly what he needs to achieve his goal.

Cool!

-- CatherineJohnson - 17 Nov 2005

The part where you describe the low performers learning to touch the letters, and to sit quietly and pay attention, sounds very much like the earliest lessons in Applied Behavioral Analysis,

Touch Blue!!!!!!!!!

I spent many, many years of my life watching Jimmy fail to Touch Blue.

-- CatherineJohnson - 17 Nov 2005

Actually, kids flat-line in ABA, which is my question with DI.

(More on this later.....)

-- CatherineJohnson - 17 Nov 2005

I think average & low-ability kids get a better education with DI than high-ability kids who can keep pace get with constructivist & traditional instruction.

No doubt about that. I'm also thinking that the high performers will do better in DI too.

BTW, Engelmann's Low Performer's Manual has finally been posted.

-- KDeRosa - 17 Nov 2005

"Actually, kids flat-line in ABA, which is my question with DI."

Yep.

-- CarolynJohnston - 17 Nov 2005

Ken—do you mind if I post links to two of Willingham's articles here? (I know Engelmann gets seriously bent out of shape dealing with cog sci!)

I'd like to post links to 'inflexible knowledge' & to 'overlearning.'

This is an incredibly helpful page. I'm going to be sending the link to everyone.

The thing is: even if you're not crazy about the idea of 100% 'scripting' of classes, which most people, off the bat, will not be, the principles are all sound and powerful whether you have scripting or not.

You've done fantastic 'rhetorical' work here; the structure sings.

THANK YOU

-- CatherineJohnson - 17 Nov 2005

Yes. Definitely.

I've become interested in this aspect of education....because of course I was one of these kids.

I think I mentioned that when I was home over the summer I discovered that my sister, who wasn't as speedy as I was at math, has more conceptual knowledge than I do. We both think it's due to the fact that she couldn't just memorize things.

I had, as a kid, and frankly as an adult, too, an amazing memory. I just.....remembered stuff.

Ros, my sister, told me that memorizing the math facts had been hard for her, and she'd had to find ways to remember them, which meant finding ways to understand them. (Conceptual knowledge does help memory; it seems that's a fairly solid & non-controversial idea.)

On her own, she figured out mathematical meanings & concepts, and used them as a memory aid!

Meanwhile, I just remembered stuff.

Ros went on to work in business, where she did lots of budgeting. She's much more facile at everyday financial math than I am.

I'll probably write a real post about this soon-ish, but I had an interesting conversation about this with Ms. LaBella, who was Christopher's 4th grade teacher.

I was asking her about the history of math tracking in Irvington.

She said that what would happen, in 4th grade, is that you'd get all these kids with fantastic memories, who'd just remembered every single thing any teacher had ever said to them.

They all thought they were great at math, and they'd come into her class and have to do something conceptual (this was before TRAILBLAZERS) and they'd hit the wall.

(I love it! Hitting the wall in 4th grade! Which, of course, is when our kids do seem to hit the wall—that's assuming you see it as a wall rather than a slow, steady, & cumulative decline....)

I found that fascinating.

Working memory is closely connected to general fluid intelligence, so closely connected that for awhile there cog scientists were (I believe) thinking it might be the same thing.

Looks like it's not the same thing, but the two are extremely close. If you have a good working memory, you have a high IQ & vice versa. (I believe I'm stating this accurately.)

Working memory isn't the same thing as a 'good memory' in the sense Ms. LaBella was talking about, but still.....I think there's something there.

There's obviously a group of brainy kids who have these sterling memories, and probably no one knows exactly what they're actually 'getting.'

I'm thinking out loud here.....

Expertise emerges from 'remembering stuff'; at some point growing domain knowledge produces expertise, which means the ability to solve novel problems in the domain.

But I wonder whether kids like me, although we acquire lots of domain knowledge, don't make that jump to expertise.

I say all this because, re-learning math, I've been stunned at how much knowledge I have (it's really quite a lot) and how little expertise.

Maybe that would be true for lots of adults, I don't know.

-- CatherineJohnson - 17 Nov 2005

Carolyn

When did you see kids flat-lining?

My experience is two years.

Interestingly, Tristram Smith, Ivar's protege, spoke at our conference about all the follow-up studies, and that's what he said, too.

He said ABA is useless after age 4 except possibly for the highest functioning kids, and it was iffy with them.

I also interviewed Cathy Lord at length on this subject.

She said she started as a fervent Lovaas person (one of 'the Lovaas girls' I think was her exact phrase) and then finally had to shift to TEACCH.

IIRC, it was the same thing: she saw the technique simply lose effectiveness.

Do you have much experience with TEACCH?

It's pretty terrific, and of course uses most of these principles, but not in the highly scripted 'Touch Blue' way Ivar classic ABA uses them.

TEACCH also introduced the idea of visual aids, visual communication, visual schedules, etc, which all the ABA folks now use, too.

-- CatherineJohnson - 17 Nov 2005

Catherine, I have no experience with TEACCH, although they tried introducing some TEACCH ideas in the early grades with Ben, like having visual cues around the room (he just ignored them, because a. he didn't need the special cuing and b. he wasn't paying attention to anything anyway).

I knew of some kids who just leveled out after a while on ABA. I don't think it was even 2 years. In Ben's case we did real ABA for less than a year, and after that we had to do other things to keep him learning.

He never responded well to traditional ABA. I remember seeing a video of a kid who had Williams Syndrome AND autism (sheesh) working with an ABA therapist. This kid was smiling, responding, learning. doing REALLY well. It was clear ABA really did something for him that it had never done for Ben, who just hated it and struggled all the way through. And yet TEACCH, with its visual methods, wasn't the thing either.

Ben has an attention deficit that's really godawful. That gets much more in the way of his learning than any LDs he might have.

-- CarolynJohnston - 17 Nov 2005

ABA was fantastic for both Jimmy (who didn't start til he was maybe 7) and Andrew (who did it when he was around 3).

But two years was the limit for both of them.

I always felt—though obviously it didn't work this way for Ben—that ABA was good for autistic kids when they were really little, because they were SO unbelievably out of it ABA oriented them if nothing else.

The way I thought of Andrew's interior was:

I'M HERE

I'M AT THIS TABLE

I'M TOUCHING BLUE

Very strict ABA seemed like the only way to situate him in time and space....

-- CatherineJohnson - 17 Nov 2005

Ken—do you mind if I post links to two of Willingham's articles here? (I know Engelmann gets seriously bent out of shape dealing with cog sci!)

Go right ahead. I think Engelmann's beef with cog sci is that cog sci principles alone don't guarantee successful teaching. There could be both good and bad implementations based on the same cog sci principles.

The thing is: even if you're not crazy about the idea of 100% 'scripting' of classes, which most people, off the bat, will not be, the principles are all sound and powerful whether you have scripting or not.

I think there are some good points to make with respect to the scripting and its benefits. I'll add this to my list of DI topics to discuss.

-- KDeRosa - 17 Nov 2005

But this is exactly what it did for Ben. It showed everyone, including Ben, that he could come far enough out of his fog to learn a little. But it moved fast. He wrote his brother's name on the first day he did ABA, for example; it was the first time he'd ever used a pencil or marker purposefully, either to scribble or draw or write. What he worked on hardest was counting items; he would pick up some number of checkers and dump them into a pail, counting them one by one. It wasn't that he couldn't count; the problem was that he'd literally space out in the middle of a count. He still spaces out and loses his place in the middle of tasks.

Ben is a bright kid with minimal cognitive problems and maximal attention problems. He'll be struggling with this his whole life.

-- CarolynJohnston - 17 Nov 2005

Doesn't Liping Ma say that Chinese teachers regard the curriculum as a script?

I think so....

-- CatherineJohnson - 17 Nov 2005

btw, I'm not myself (necessarily) opposed to scripting.

My point is that this is highly persuasive even if you are.

-- CatherineJohnson - 17 Nov 2005

wow!

So it did the same thing for him.....

Well, that's sure the way I came to see it, especially with the little guys.

I always had the strongest sense Andrew (and Jimmy) had to FIGHT to find out their coordinates.

-- CatherineJohnson - 17 Nov 2005

btw, I'm not myself (necessarily) opposed to scripting.

My point is that this is highly persuasive even if you are.

But the scripting issue is one of the frequently raised criticisms (spun with a negative connotation) that get brought up. This misconception needs to be diffused and a ready response is needed.

-- KDeRosa - 17 Nov 2005

But the scripting issue is one of the frequently raised criticisms (spun with a negative connotation) that get brought up. This misconception needs to be diffused and a ready response is needed.

Absolutely.

The word itself—scripting—is death.

-- CatherineJohnson - 17 Nov 2005

One of the answers, for me, was the discovery that 'scripting' does not mean 'lecturing.' (I realize that not everyone equates scripting with lecturing, but that connotation is always there, lurking.)

When I read Carol Gambill's description of her method of teaching algebra, a fully-scripted, Direct Instruction approach, the scales fell from my eyes.

She scripts questions.

Not just statements.

The woman in the direct instruction video I posted awhile back says exactly the same thing.

She describes a looser version of scripted instruction, in which a teacher scripts, and rehearses, more questions than she's going to need.

She has a lot of arrows in her quiver going in, so to speak.

-- CatherineJohnson - 17 Nov 2005

I suspect we need 'second-generation' Engelmann....

DI is really ABA for normal people.

The Koegels (Lynn Koegel was on the Nanny show) are second-generation Lovaas.

They created a brilliant child-centered version of ABA that is a wonder to behold.

Now, a teacher has to be very gifted, IMO, to do what the Koegels do. We had both our kids working with Koegel students. Jimmy was with a woman who was practically a 'Pivotal Response Therapy' genius; she could do anything with any situation.

Andrew was with a young graduate student who wasn't so naturally gifted at the method, and he broke her. (He really did. She was a beaten person by the time Andrew got done with her.)

The Koegels do conventional ABA, but it is never scripted. They always start from what the child is doing right that minute.

-- CatherineJohnson - 17 Nov 2005

I should add that the Koegel approach isn't as important for normal kids, I don't think.

One of the enormous problems with autistic kids is that they lack motivation.

The Koegels constantly 'build motivation'; motivation is one of their 'pivotal behaviors.' (Pivotal behaviors are a handful of behaviors that, when remediated, produce an enormous amount of corresponding change. It's a cascading change idea.)

To sit an autistic child at a table all day and follow scripted interaction with him almost by definition defeats the goal of building motivation.

Normal kids have tons of motivation, although they don't necessarily have tons of motivation to study and learn academic skills & subjects.

That's why I say my guess is a 'post-Engelmann' will pop up, someone who has incorporated all of his insights about timing, grouping, practice & so on with the insights of cognitive science concerning frontal lobe function.

-- CatherineJohnson - 17 Nov 2005

That's why I say my guess is a 'post-Engelmann' will pop up, someone who has incorporated all of his insights about timing, grouping, practice & so on with the insights of cognitive science concerning frontal lobe function.

We can dream. The ED community won't even admit DI is a superior teaching sytem, let alone adopt it. Much less try to improve it.

-- KDeRosa - 17 Nov 2005

We can dream. The ED community won't even admit DI is a superior teaching sytem, let alone adopt it. Much less try to improve it.

well, uh, yeah

-- CatherineJohnson - 17 Nov 2005

Ken, I'm posting these two graphs here for now, if you don't mind. (My computer's getting wonky, and I need to re-boot.)

Figure 1. Reading, math, spelling and language achievement for the nine models. Scores above the horizontal line at the 20th percentile indicate a positive effect of the program on achievement in that skill area relative to "control" children who did not participate in the project. Scores below the horizontal line indicate a negative effect, relative to controls. The Washington Times.

Figure 2. Basic skill, cognitive skill and self-esteem scores of children who participated in the nine models. Scores are relative to a baseline of zero that represents children who did not participate in Project Follow Through programs. Scores above the zero line indicate a positive change in performance in the target area, and scores below the zero line indicate a negative change in performance in the target area (See figure for legend).

Direct Instruction evidence

-- CatherineJohnson - 22 Nov 2005

Actually the DI results are even better than this. These scores included a large test site that stopped using DI half-way through the project and excluded results from certain sites because the control groups at those sites scored higher in the initial testing than the DI groups.

-- KDeRosa - 22 Nov 2005

Do you have charts anywhere?

-- CatherineJohnson - 22 Nov 2005

See Figs. 5-7. Actually, your graph looks like the fig 8 which also uses the corrected data.

Note that the Behavior Analysis model used DI as its curriculum in at least some of the test sites.

-- KDeRosa - 22 Nov 2005

Note that the Behavior Analysis model used DI as its curriculum in at least some of the test sites.

I was wondering what the difference was....

-- CatherineJohnson - 22 Nov 2005

I'm working on a summary of Project Follow Through and I have summaries of all the models tested. Here's a teaser: the models that used discovery learning, open classrooms, and whole language are the ones that performed the worst.

-- KDeRosa - 22 Nov 2005

This seems so obvious. It's always kind of miraculous that professional educators promote 'discovery.'

-- CatherineJohnson - 23 Nov 2005

Back to: Main Page.

Navigate KTM

Kitchen Table Math

KTM User Pages

Service Groups

Parent Groups

Personal Pages

Blogs

Special lists

Help

The Best Model on How to Teach Children (and Adults): Direct Instruction (DI)

Big Concepts of DI

Comments