U.S. education

[hudebnik]

I just read the book Let's Kill Dick and Jane, which chronicles the rise and fall of the Open Court textbook publishing company, which started in 1962 with a mission to produce grade-school textbooks on a European model, with higher-caliber literature and more-challenging language than was available in the U.S. at the time. Later the company added a series of grade-school math textbooks, dabbled unsuccessfullly in science and music, and after decades on the perpetual brink of bankruptcy, was sold to McGraw-Hill in 1996. The book is part hagiography of the company and some of its evangelists, part indictment of the U.S. educational bureaucracy and culture for their inability to adopt anything truly new, no matter how well it works.

For example, a chapter on the state adoption process points out that most state textbook adoptions cover several years of school at once, e.g. K-8. Now suppose (hypothetically) a curriculum and textbook series were to become available that produced twice the learning rates of anything currently on the market. The 8th-grade curriculum would assume, as a prerequisite, the 7th-grade curriculum, which would be equivalent to 14 years of anything currently on the market, and thus be far beyond the capability of students entering 8th grade next year; it would be sheer lunacy to adopt such a curriculum for use next year. The only place a truly improved curriculum could possibly be adopted is at the Kindergarten or 1st-grade level, and it would produce students who were hopelessly bored in their next year or two of school; again, it wouldn't make much sense to adopt such a curriculum without its sequels. In other words, even if there were a dramatically better way to teach the material, and everybody agreed that it was dramatically better, it couldn't possibly be adopted; an adoptable curriculum must closely resemble what's already in use.

Of course, publishing companies know very well that nothing will be adopted on a large scale unless it closely resembles what's already out there. The article The Muddle Machine describes how the sausage factory of textbook publishing works: in brief, a publisher

1. lists everything that all the existing textbooks do


2. throws in a few mentions of the latest educational buzzwords


3. uses staff writers to put all this together into something resembling a book,


4. tweaks things to explicitly match the textbook adoption guidelines of Texas, California, and Florida (the largest-population states that have statewide adoption)


5. censors anything that might offend political groups powerful in any of these three states (contraception, relativism, characters conforming to racial or gender stereotypes, etc.), and finally


6. starts looking for "authors" with academic credentials to put their names on the product.

Not exactly a recipe for innovation.

So there are problems at the state-bureaucracy and publisher levels. At the grass-roots, individual-classroom level is a different problem: most teachers teach the way they were taught, and even if they wanted to try something new, their workloads don't leave a lot of time for re-training and development of new course plans. Frankly, doing something truly new in the classroom is hard work; it's a lot easier to just do the same thing you did last year. So every time a new educational fad comes along, thousands of teachers across the country (voluntarily or otherwise) adopt its terminology and superficial features, while not actually doing anything fundamentally different. (In computer programming, "object-oriented" became the hot phrase about twenty years ago, and lots of people said they were doing and teaching object-oriented programming when they were really writing the same old programs, with the same old techniques, in C++ that used to be written in C or Pascal.)

Then there's the whole "professionalism" question. Teaching is not an attractive or respected profession. To quote from the 1985 government report Becoming a Nation of Readers,

College students who choose education as a major have lower average scores on a number of indices of ability than students who select other majors. Among students who begin an education program, those who complete the program have less ability than those who switch to other programs. Among college graduates who get teaching certificates, those who seek teaching jobs are less talented than those who do not. Most alarming of all, among people who take jobs as teachers, those who remain in teaching after five years are less able than those who leave to enter other fields.

In other words, at every stage, the "best and brightest" are somehow driven away from being teachers -- whether by the pay, the hours, the safety and working conditions, the social respect, the lawsuits, the bureaucracy, or something else -- and we're left with a lot of mediocre teachers (I say, as a college teacher whose brother is a middle-school teacher and who has a number of high-school-teacher friends).

A lot of educational decisions seem to be made by politicians on a political basis, rather than by educators on a scientific basis. For some reason, the "phonics" approach to reading instruction became associated with the political right, and the "look-say" and later "whole language" approaches with the political left; as a result, which approach a school, district, or state takes is often more determined by the politics of its leaders than by what actually works. (Open Court's approach apparently involved a lot of phonics in the first year, and a lot of high-caliber literature thereafter, so it was objectionable to both sides.) The "Ebonics" debate of a few years ago is another example: it should have been decided by educators, on the basis of what works, when, for which students, but it became a political cause celebre, and mundane questions like "what works?" became irrelevant.

There's actually a government-sponsored database of educational techniques that have scientific evidence that they work, called the What Works Clearinghouse. As one of its directors told us during a campus visit a year or two ago, "it's a very small database": apparently there are so many random variables in education that almost nothing -- no textbook, no classroom technique, no syllabus, no curriculum, no student grouping -- can be proven, with statistical significance, to make any difference whatsoever in how much students learn. But there are some such interventions, and a little bit of halfway-decent research. Almost everything in the database is at the grade-school level. I've been trying for several years to come up with statistically-valid results on what works in teaching computer programming at the beginning college level, and it really is amazingly difficult to tease signal out of the random noise.

Comments

[siderea]

Welcome to my world. :/

Have you read Lies My Teacher Told Me by Lowen(sp?)?

For some reason, the "phonics" approach to reading instruction became associated with the political right, and the "look-say" and later "whole language" approaches with the political left

Really? I thought it the other way around. Huh.

Because that book concerns curricula Before Internet it probably doesn't mention the latest wrinkle: as schools and curricula start involving internet tech, they adopt platforms which embody certain presumptions about teaching and learning... very, very narrow presumptions. Online instruction is thus often less flexible than classroom. If it won't run on the publisher's or the schools' LMS, they won't adopt it. So products are developed within the very narrow constraints of educational behavior the LMS supports: lecture, discussion, drill, quizes. *sigh*