Through a Glass Dully
Alex Tabarrok writes about the demonstrated merit of Direct Instruction as a teaching system that doesn’t rely on the sort of star teacher Hollywood makes heartwarming movies about. But he doesn’t seem to extend his analysis of the star teacher problem far enough:
We are supposed to be uplifted by these stories [of heroic teachers] but they depress me. If it takes a hero to save an inner city school then there is no hope. Heroes are not replicable.
What we need to save inner-city schools, and poor schools everywhere, is a method that works when the teachers aren’t heroes. Even better if the method works when teachers are ordinary people, poorly paid and ill-motivated – i.e. the system we have today.
Fair enough! And . . .
. . . large experimental studies have shown that the teaching method which works best is Direct Instruction (here and here are two non-academic discussions which summarizes much of the same academic evidence discussed in Ayres). In Direct Instruction the teacher follows a script, a carefully designed and evaluated script.
But . . .
Contrary to what you might think, the data also show that DI does not impede creativity or self-esteem. The education establishment, however, hates DI because it is a threat to the power and prestige of teaching, they prefer the model of teacher as hero.As Ayres says “The education establishment is wedded to its pet theories regardless of what the evidence says.” As a result they have fought it tooth and nail so that “Direct Instruction, the oldest and most validated program, has captured only a little more than 1 percent of the grade-school market.”
Isn’t this just a way of saying that Direct Instruction itself is hardly more practical as a universal solution than “teacher as hero?” It’s simply that Direct Instruction fails for different sociological reasons. Star teachers don’t solve the problem of education because not enough teachers are capable of stardom. Direct Instruction doesn’t solve the problem because not enough teachers are willing to do that work. Direct Instruction sounds like utter drudgery from the teacher’s perspective. I’m inclined to think the resistance to it in the existing education establishment indicates not that the class of teachers are full of themselves, but that you could no more fill America’s schools with people willing to speak the same script day in and year out than you could fill them with Jaime Escalante clones.
See also, Megan McArdle on a similar, arguably parallel, issue in the medical profession.
Comment by Sam_S —
September 27, 2007 @ 8:18 pm
No opinion on the grand theory of it all, but I would guess that direct instruction works much better in some disciplines than others.
I DID teach a DI math course to middle schoolers in an inner-city school. It was extremely well-designed to push them forward inch-by-inch, and I gotta tell you, it worked wonderfully! It wasn’t such boring drudgery (well, no more than ANY math course): there’s plenty of interaction with students, coaching, etc. going on, which is the fun part of teaching anyway.
Comment by Anonymo —
September 27, 2007 @ 9:47 pm
I suppose the obvious problem is that U.S. primary and secondary education, being largely divorced from market forces, earns more through failure than success.
Comment by Leonard —
September 27, 2007 @ 9:48 pm
Yes, and employees also hate making fast food the same way, over and over again. That’s why all fast food restaurants rely on “star” burger flippers and fry chefs. But there just aren’t enough! They never get results! My fries are always different, occasionally Harvard-quality, but usually limp and greasy. Year in and out, and in the inner city, watch out!
If only these evil corporations would implement Direct Food Preparation, things would be better. Imagine it: McDonald’s with consistent (although mediocre) food! But obviously they can’t do that.
Comment by tc —
September 27, 2007 @ 9:51 pm
If we were surrounded by movies and TV shows glorifying the power of drudgery, maybe.
Comment by Jim Henley —
September 27, 2007 @ 10:28 pm
Leonard, you’re being obtuse and perilously close to an ass. However, if you believe the same labor pool that provides McDonald’s employees can also provide the nation’s teachers, then you’re clear on the ass charge. It would be a tough case to make, though.
Sam: Thanks much for the first-hand report. I’d certainly like to hear from any other teachers who have actually used DI.
Comment by Glen Raphael —
September 27, 2007 @ 11:30 pm
Is teaching grade school kids really inherently much harder than feeding them?
As I understand it, Jaime Escallante’s success mostly came from the fact that he used good self-paced homeschooler materials. The kids taught themselves at their own pace. And that part is replicable.
Assume the knowledge being taught is primarily conveyed via books and prerecorded videos and the occasional pre-scripted lecture that introduces the other materials. Further assume older kids can be enlisted to tutor younger ones who fall behind and even kids at the same level can tutor each other for a small amount of pay or extra credit or gold stars. What’s left for the teacher is essentially a babysitting role.
The McDonald’s approach to teaching shouldn’t require much in the way of specialized knowledge or skills beyond those needed by a babysitter, a nanny, or a fry cook. Should it?
Comment by Thomas Nephew —
September 27, 2007 @ 11:50 pm
The McDonald’s approach to teaching shouldn’t require much in the way of specialized knowledge or skills beyond those needed by a babysitter, a nanny, or a fry cook. Should it?
Yes, it should, I think. Even with a DI approach, you’d like a teacher to know substantially more than his or her students, and know it more confidently. Not that that’s impossible for babysitters, nannies, or fry cooks, but you’re essentially arguing that training, substantive knowledge and aptitude don’t matter at all in being a good teacher, and I don’t think that’s true.
Comment by Sam_S —
September 28, 2007 @ 2:27 am
“Is teaching grade school kids really inherently much harder than feeding them?”
Bwa-ha-ha! Try both for an hour! (hint: they LIKE food).
Yes, you gotta know more than the students, and I doubt the DI proponents would argue that teaching aptitude doesn’t matter (I’m not going to argue either way, I’m just fascinated that the subject came up). I suspect that the higher-ups and other pontificators (maybe even some of the teachers) might say the method is some kind of “cheating”, but in private the teachers at my school said “Hallelujah, something that works!”
But this was ‘rithmatic, mind you. I would bet putting Science or History into DI format is much more challenging. Adding and subtracting can be learned in tiny increments along a line of progress, but more verbal, lateral-thinking subjects…. I don’t know how I’d do it, but I bet there are wizards (heros?) at work on it now.
Comment by Leonard —
September 28, 2007 @ 8:15 am
Jim that was pointed satire. It seems silly to me to have discussions like this ignoring the elephant in the room. The state ownership of the means of production is not something that we should just accept without comment.
McDonald’s does, in fact, achieve “DI”, and has since time immemorial, with a labor force far less tractable than teachers. (And no, they are nowhere near the same labor force: McDonalds has it much, much harder than a school. Tell me who you’d rather have working for you: a crew of surly mildly educated inner-city teenagers, or a crew of 30-40ish educated women?) How does McDonalds do it? It’s simple: they train new employees their way, and they fire anyone who will not implement their system. This has a strong filter effect, as well as “encouraging the others”.
Well, the schools “could” do that, but they don’t, for institutional reasons. This is what you’re saying. What are those institutional reasons why schools fail to do what they should? Neither you or McArdle make that jump, even though as libertarians (you more than her) I should think the answer would spring naturally to mind.
The managers at McDonalds are motivated to get things right: they are running the company for profit, and therefore try to align it with the interests of the customers in spite of how much grief this causes the labor force.
The managers at our school system are motivated by election. So long as they don’t get something obviously wrong (so obvious that the sleepy masses wake up and vote them out), they generally keep their jobs. So, they have no motive to change anything. Rather the opposite: in democratic socialism there is a strong motive for stasis. And they have every motivation to give in to the workers and create a workplace almost totally controlled by the teacher’s unions. This is what has happened.
This gordian knot sits before us, and that knot holds shut a prison that is failing millions of children, every single year, and nobody does a goddam thing. Yes, it makes me mad. Libertarians have a knife: privatization. It bears mentioning.
Comment by Barry —
September 28, 2007 @ 8:50 am
Aside from the many, many differences which you are glossing over, Leonard, how well have charter schools been working out? One of the things which has actually surprised me, as a liberal, is just how poorly they’ve performed in various school district experiments.
Comment by Barry —
September 28, 2007 @ 8:51 am
Sorry, but I shouldn’t have hit ‘post’:
To me, the obvious aspect of privatization is that a privatized system would have absolutely no trouble simply dropping a large number of students. After all, it wasn’t until well after WWII that the working-class expectation was high school graduation.
Comment by No Nym —
September 28, 2007 @ 9:33 am
The literature on DI isn’t quite as rosy as its proponents paint it to be. The curriculum did moderately well in Project Follow Through, and its sellers have continued to massage those data and the early cohorts to sell their wares. It helps when you can drop the comparative data and reconstruct your control groups using multiple linear regression.
So far, I am aware of no evaluations of the curriculum that were conducted in an even quasi-experimental manner (though I’d be happy to learn of one that has been done). The appeal of the curriculum is the back to basics approach and the (apparent) ability of the curriculum to bring inner city schools near to suburban standards.
What’s glossed over, other than the data massage that you need to do to get DI to win consistently, is that the suburban national average is fucking piss poor. If DI could clearly and consistently improve the national average, I’d be more excited. As it stands, DI is probably a really nice remedial curriculum. It is not, however, Singapore Math (just to randomly name a rival curriculum that beats the pants off the US average on a regular basis).
Comment by Tracy W —
September 28, 2007 @ 10:44 am
Well, firstly, under DI teachers don’t say the same script day in and year out. Each day involves a different lesson. After all, if the teacher presented only one lesson every day for the whole year the kids would learn that one lesson very well but they wouldn’t learn much overall. There is some repetition in the material so a kid who has missed one or two days of school shouldn’t need special remediation, and of course there is continual practice of past concepts built in, but the lessons do change from day to day.
Secondly, DI lessons only cover reading, writing and maths, and only take half the school day, leaving the other half for the teacher to teach music, geography, history, art, PE, etc. Teachers may be encouraged to use relevant DI principles in those lessons (such as regularly asking questions of the class to check if they are understanding, and relating new knowledge back to old knowledge [this means that for example, if you have taught children where the US is on a globe, and then you decide to teach them about Switzerland, it makes sense to show them the US on the globe and then trace with your finger over to Switzerland]). But teachers will need to develop their own lessons for those classes. The things about music, history, art, geography, history is that learning them is not so dependent on past information as reading and maths. A child who does not recognise the letter “x” is going to have difficulty reading a word like “x-ray”, a child who cannot do addition is going to have problems with subtraction, but a child who has never heard of the dinosaurs can still learn something about the Incas. Obviously, any particular school or school district can schedule this half of the day down to the last second, or fill it up with pointless busywork, or what not. But that would and does happen reagardless of whether DI is adopted.
Thirdly, DI does not consist of mindlessly reading out scripts. Kids are expected to provide responses about 10 to 14 times a minute on average and the lesson is modified based on those responses. Therefore the teacher is constantly listening to the kids to discover what they are understanding and what is more difficult. If any kid is getting their answers wrong then the teacher must go back, discover the point of error and correct it. And then there are all the meta things teachers must think about, eg “I wonder why Ben keeps getting the answers wrong. Maybe it’s a problem with his hearing? I’ll get the school nurse to test him.”
Fourthly, teachers get the positive feedback of their kids successfully learning things and progressing. Don’t underestimate that as a reward. Which teacher do you think will be more inspired to carry on? One who gets to the end of their first year after floundering the whole way, with perhaps half of their students meeting grade standards and another half they just can’t seem to reach, or one who has brought a class from barely able to hold a pen to being able to write sentences for themselves, who has successfully identified that Ben’s problems were caused by some hearing loss, who has had a million and one other successes.
So to summarise:
– teachers present different scripts each day
– teachers have half the day to teach other topics, for which no scripts are provided
– teachers are dynamically adapting their lessons on the fly based on kids’ responses so the teachers’ brains are working even when using the script
– teachers are getting positive rewards themselves.
Comment by Joshua Holmes —
September 28, 2007 @ 12:00 pm
McDonald’s does, in fact, achieve “DIâ€, and has since time immemorial
No one remembers the 50s? My grandparents would beg to differ.
To me, the obvious aspect of privatization is that a privatized system would have absolutely no trouble simply dropping a large number of students. After all, it wasn’t until well after WWII that the working-class expectation was high school graduation.
Outside of the kids going to serious colleges – i.e. not party schools or community colleges – who gets anything out of the last 2 years of high school anyway? Talk to a kid taking standard 12th grade English, and ask him what he’s doing there. He has no idea. (She, too.) Two hundred years ago these kids weren’t in school – and neither were any other – they were out doing things, working, contributing to the world. We’ve gotten rid of all the other avenues for teens to learn and contribute besides “go to school”. Perhaps that trend is irreversible, but a society where it isn’t the norm isn’t a problem.
To wit, a few years ago I worked with an intelligent young woman. She had some meaningless title, but at bottom she was a clerk; she did paperwork. She also had a bachelor’s in political science from a solid university. Why? My grandfather, just two generations ago, quit high school to fight in WW2. He came home, got his GED, and became a clerk for the local chemical company. Was her paperwork so much harder than his? I doubt it.
Government education and the cult of credentialism have started an arms race where you need a two-year college degree to type letters and answer the phone, and a four-year degree to prepare a report. What the hell happened?
Comment by TGGP —
September 28, 2007 @ 12:05 pm
I have not heard of any experiments with hero teachers. That is because it is beyond the power of the experimenter to cause such a thing. It was in the power of the makers of this study to put DI in place. In how many other fields are there such problems because employers cannot say to their employees “Do this”?
Regarding Escalante, it is often forgotten that he was not liked by other educators and did not stay on. So part of the sociological analysis is not just whether a given teacher will perform the hero/DI role, but how others feel about the teacher.
Regarding raising the national average, do you want to import people from Singapore? That would be one way to do it.
Comment by Leonard —
September 28, 2007 @ 12:28 pm
Barry:
Actually a privatized system would have strong incentive not to drop students. Students would be customers: paying customers. That is what the profit motive does in free enterprise: it induces people to serve others. You don’t make money off of customers you throw out of the store. If anything, the danger in a privatized system runs the other way: that the schools would attempt to keep people in the system that really don’t need it and should be moving on. (Graduate degree in English, anyone?)
What you are getting at, is that a privatized system would carefully look at the cost/benefit of serving students. If a student was a problem, then, yes, most schools would drop him.
It’s not clear what you think about this, but I believe it is a feature, not a bug. People who do not want education will not get it, in general, at least not with the level of violence we are willing to exert in the attempt to get them to comply. Yet we attempt anyway, to their detriment, as they become bored and angry about being confined in a place that they hate. And much to the detriment of all the other students who do want to learn, but cannot amidst all the disruptions.
The inability to meaningfully expel is one of the prime reasons for the failure of our schools.
As for the horror of having people around without highschool educations: that’s not such a bad thing. As you indicated, it used to be the case, and they seem to have survived. Further, while diplomas are nice and all, what we really want is not more paper but actual educations. On that score, we are already living in a society full of people who might as well have not been “educated”. According to the US government, some 22% or so of adults are minimally literate. (See http://ged.ed.sc.gov/NalsNarrative.htm ). Certainly the levels indicated for “level 1″ of literacy are way below high school.
Comment by Leonard —
September 28, 2007 @ 12:37 pm
As for charters, that is hardly privatization. But it is at least a baby step in the right direction, gaining some of the benefits.
I was not aware they had been failing; I thought they had been pretty successful. Perhaps we could start with the wiki entry on it:
http://en.wikipedia.org/wiki/Charter_school#Study_Results
Doesn’t look like failure to me.
Comment by Mark Z. —
September 28, 2007 @ 2:37 pm
Actually a privatized system would have strong incentive not to drop students. Students would be customers: paying customers.
They also have a strong incentive to drop students who are dragging down their test scores, or who require a disproportionate amount of teacher time.
Comment by Madeline F —
September 28, 2007 @ 2:52 pm
Sounds like DI might be equivalent to running a RPG from a published module. Helps the crappy GMs who have no idea how to set up a fun game, holds the good GMs back a bit (but usually they can still elaborate within the framework and throw out the crap aspects). If there are more crappy GMs in a population than good GMs, running entirely from modules wouldn’t be a terrible solution, because the module-writer would be a smarter GM than the average GM there.
Comment by Leonard —
September 28, 2007 @ 3:45 pm
Some might have the incentive to drop students for bad test scores. Sure. But test scores as a metric of collectives (as opposed to individuals) are an artifact of our socialistic schools. Because some fail, we are desperate for a means to measure them.
Even if you accept that tests are fair measures of education, what matters is not what a school’s average is, but how well it does at increasing each particular individual’s test scores over time. But since as a collective we refuse to really even think about individual accomplishment very much, much less individual differences in intelligence, we are unable to use test score information in a rational way.
Or, put another way. Consider a school which accepts only borderline retarded students (those with IQs in the range 70-79). Another school accepts only geniuses with IQ 140+. On a standardized testing regime, each student in the first school improves by 10 points each year he/she is there. In the second school, the students do not improve at all. Each one gets the same score each year he or she is there. If the first school’s scores average 600 each year, and the second, 1500, which is the better school?
As for “disproportionate amount of teacher time” — yes, of course. Very generally, free markets charge for labor time proportionate to use. This is not a surprise: labor ought to be compensated, and people are resistant to working for free. However, I’d expect the way that this will be handled in free market education is not all schools setting one price and then dropping all students who requires more work than the price warrants. Rather, we’ll have a wide variety of firms, some offering more service at a higher price, others less at a lower price.
Comment by socratic_me —
September 28, 2007 @ 5:33 pm
So what you are saying Leonard, is that people who could afford the cost would get great educations and people who could not would get significantly “less great” ones. How exactly is that different from the current setup?
_
Also, I find it absolutely amazing that people think teachers/unions control the classroom and that their poor bosses, be it the administration or school boards are just powerless to actually lay down the law. I have yet to find a school that is unable to get rid of teachers for legitimate reasons (like failure to actually teach). What I have found is schools that are so desperate for teachers that they are unwilling to cut dead wheight for fear that they might be forced to hire someone even worse if they can find anyone at all. This is especially true of Mathematics and Science instructors.
_
If we actually doled out the money it took to make education a highly desirable field, I think you would find that Unions have a lot less say than everyone thinks when it comes to hiring and firing teachers. Unions don’t stop most personel issues, lack of high quality applicants does. This is why people keep getting back to quality teacher pay as an issue.
_
Many of us already teaching do it because we feel called to do it in spite of the incredibly long hours, terrible pay, etc. We also know that there aren’t 6.2 million people who feel this way and are actually qualified, which leaves us teaching alongside (and repairing the mistakes of) unqualified or unmotivated educators. What we would like to see is enough incentive that we could bring in more people who would like to teach and would be good at it, but just aren’t willing to make the sacrifices that are currently demanded of teachers.
Comment by Leonard —
September 28, 2007 @ 8:39 pm
Several ways.
For one, right now there are schools that are physically dangerous places. If you cannot expel students, but rather must slap their wrists (metaphorically: we won’t even do that), then that is what you can expect. The first thing I’d expect in a private system is real zero tolerance of assault or battery, whether by student on student or by student against teacher.
So, that’s a hint as to what I mean by “less great”. “Less great” means really bad. This sort of thing: http://www.city-journal.org/html/13_1_how_i_joined.html . I’m not talking about some poor white kid in Muskeegan having to make do with old textbooks. I’m talking about a black kid in an inner city, sitting in various classrooms all day long and not learning a thing. Nothing. No thing. Zero. Negative, really, considering what is actually being taught: that “the system” doesn’t give a shit about you; that people who break rules go unpunished and gain friends and attention.
In a private system, the disruptive kids making it impossible to teach and impossible to learn in that second scenario would not be there. They’d either be out of school entirely, which IMO is a better outcome than what we have now, or they’d be in schools that considerably more structured in some way. (Of course, many of them are perfectly capable of self-discipline if they see it as in their own interest, and might be right there learning away in a school with real expulsion.)
I don’t know that the set of people who are teachers would be the same, either. Certainly I think it would make some difference, just not as much as with the kids. I do think that the pedagogy would be much different, especially given time.
Second, IMO while it is normal to get a passable education in modern social democratic schools, getting a “great” education is something rare. I don’t see why that should be, for kids with the brains to handle it. I look back on my own high school, in a very good system. It was a small community system, controlled by the parents, many of whom were PhD scientists, and absolutely safe. I think that much of my public education was good. But there were many slow years, and many things never taught that should have been. Economics, just to take one thing. (I took the one course they had, not required, and it was poorly taught and full of turgid Keynesian macro bullshit. Stagflation was of great interest to the teacher because it had just happened, and yet, theoretically not possible!) Or, evolution, and not just of animals, but of people. I did get a very good math education, though.
In short, I’m envisioning moderate improvement in the middle, but great improvement for the low socioeconomic classes and for the highly intelligent kids. These are the groups that are the most ill-served (albeit in vastly different ways) in our current system, relative to their potential.
As for your implication that ability to pay would make a huge difference in quality of education, I’m not seeing it. Lots of expensive capital is not required to supply educations, and getting back to the subject of this whole thing, neither are “star” teachers. This is not the NFL. Normal people can, and do, teach. While it may take a “star” to teach in an out-of-control inner city school right now, it doesn’t take a star to teach in a normal school which has discipline. You may think that spending $13000/student-year is necessary, but I do not. (That’s what they spend in DC.)
I do worry about ability to pay cutting off education altogether. This is why I would advocate making long-term student loan plans available, voluntarily, to students. Parents would be allowed to “indenture” a kid to pay for her education. The child would have to repay, in return, 1/10 of her income until she paid off the loan. (If this kind of “indenture” appalls you, then consider that it’s milder by a good amount than our current system, which taxes at a higher rate than 1/10.) Vouchers would work as a compromise, being partway to privatization without scary problems of cutting off impoverished students.
You seem to think that teachers are underpaid. Perhaps they are, for what they must face. I don’t know, but then I am advocating a policy that means I don’t have to know. If you are right, and really pouring money in is the solution (although it has not worked yet, and we’ve been pouring for decades now) — then a free market will set teacher wages higher.
Comment by Jim Henley —
September 28, 2007 @ 8:49 pm
Carry on, folks. As to DI specifically, who can tell me the relative market penetration in private schools (religious, secular) vs. state schools. I’m curious how close they are.
Tracy W: Thanks for your substantial comment. Can you tell me your experience of DI? Are you a teacher who has used it on the job? A parent of a student who has been taught by it? An administrator from a school system that has adopted the program? Someone who works for the publishers of a DI curriculum? I’m interested in the broadest possible range of perspectives, and just want to understand where yours fits in.
Comment by dsquared —
September 30, 2007 @ 9:53 am
Is it me, or does that Tabarrok post start off by talking about problems with *high* schools, and then segue into talking about the market share of Direct Instruction in *grade* schools? Is that not just a teeny bit of a bait and switch? A bit of googling turns up no evidence of a DI equivalent for 12-16 education even existing.
Comment by Jon H —
September 30, 2007 @ 3:41 pm
then a free market will set teacher wages higher.
Maybe. That might happen. Or, maybe the market will end up focused on the end results of an education (the documents and certifications) rather than the process. If you can get the documents proving that you’re ‘educated’, then one needn’t pay the teachers much at all.
Or, perhaps an 800-lb Wal-Mart style company will be formed, which will provide those documents, very cheaply, with almost-acceptable-but-increasingly-low-quality instruction, and with market dominance become the only employer for educators, squeezing out better teachers due to the low wages.
The market will work, but nobody can predict how. Assuming it will work in the most favorable way is simply wishful thinking.
Comment by Jon H —
September 30, 2007 @ 3:49 pm
Tracy wrote: “- teachers present different scripts each day”
I think we all understood that, but the point is that those daily scripts are all part of one giant schoolyear script, and all teachers use the same script for a given day.
The way you present it, each teacher could have their own scripts, in their own order, with their own content. Which is pretty much what we have now with daily lesson plans.
Comment by Andromeda —
October 2, 2007 @ 7:29 am
Leonard: At present the free market sets teacher wages notably lower; private school teachers (of whom I am one) make less than their nearby public school counterparts. When our business guy was doing a presentation to the faculty on the compensation initiatives we have going on, he noted that private schools generally strive to pay above 90% of the local public wage, because otherwise attrition starts to be a problem. According to salary.com, the middle 50% of public school teachers in the zip code where I teach make $48K-$64K, whereas my school’s *entire salary range* is roughly $30K-$65K.
I very much doubt the free market will deliver higher teacher salaries, because it hasn’t, and because there’s something people tend to underestimate when they make that claim: running a school is extremely damn expensive.