Another Central Plan Fails
For at least two decades, the conservative wing of education experts has touted one magic bullet (apart from vouchers): high-stakes testing. The idea is to subject students (and teachers too) to a standardized test that would create incentives to learn the basics, compel curriculum committees to toss out the fluff, yield reliable data for assessing performance, and inspire students and teachers to keep their noses to the grindstone.
In state after state for the last ten years, these tests have become the leading avenue for education reform. The trend began in Texas and spread. Now millions of students begin their first day of classes with a profound awareness of the impending make-or-break tests, to be taken at regular intervals from the first to the twelfth grade. On the face of it, it seemed to work. Curriculum committees threw up their arms in despair, teachers started teaching math and stopped teaching so much fluff, and everyone had a sudden sense of accountability.
A conservative victory? The Bush administration thinks so, which is why it is working toward the supposed dream of a national testing system. New data, officials say, can be generated that allow for a comparison between states, build proof of success, and otherwise allow for a better national system of education. The "No Child Left Behind" legislation uses carrots and sticks to impose high-stakes testing on states that do not currently use the system.
While the plan seemed sound on the face of it, the reformers forgot one thing: we are dealing with public school, which lacks any real means of operating in a sound economic (which is to say resourceful and rational) manner. Think of it this way. Let's say that Soviet grain production had been down for 3 years straight and some clique came up with the idea that the workers and managers needed clearer rules for daily operations. The plan may look good on paper, but in the end it doesn't address the underlying problem: the fact of central planning itself.
Central planning has several universal features. It is coercive. It bypasses the needs of the consumers for the sake of politics. It relies on edicts which may or may not reflect reality. It does not take advantage of the price system, profit, or loss. It is impervious to change. It ignores local conditions. It does not permit flexibility according to circumstance. It robs those who know the most of the ability of make decisions and innovate. It creates incentives to obey the plan but diverts attention from the real goal, whatever it may be (and it may be the wrong goal). It ends up overutilizing material resources, underutilizing human ones, and not generating the intended results.
All of these features have doomed the testing movement, at least if you take seriously the results of a new study from Arizona State University, the first to examine the issue nationally. The researchers have found an inverse relationship between the ability to pass the tests and the scores on independent assessment tools like the SAT and ACT. The latter come up with a measure of the student's mastery of the ability to think and solve problems. The school exams, on the other hand, only measure whether students have mastered the material on the tests, which are not thought-based but curriculum based.
It turns out that even as students have shown consistent improvement on state tests, the opposite is true with regard to performance on outside tests. After adopting these exams, twice as many states slipped against the national average on the SAT and ACT as gained on it. This turns out to be true across the board, even on math scores (with the exception of middle-school math). And the trend on Advanced Placement tests were also worse in states that had adopted tests.
What's the story? Well, the tests themselves have become the curriculum. That's all that teachers focus on, and they do so at the expense of teaching valuable learning and thinking skills. The one goal of passing the tests has replaced the goal of producing good thinkers, students, learners. The students are being trained narrowly (the school tests measure that) but not broadly (as is shown by SAT/ACT data) and hence the whole point of education is being lost.
Just as strikingly, the study also found an increase in drop-out rates. In fact, it is likely that the study underestimates drop-out rates because it relied exclusively on reported data, while most everyone agrees that there are more drop outs than are typically reported. Now, the conservative response to this bit of news might be: good! If a student isn't there to learn, better that he leave the classroom and cease to hold back those who do want to learn. And, in some way, there's a valid point here.
Yet, I think back to the story once told to me by former Soviet economist Yuri Maltsev. When the Soviet government became alarmed at the high death rate in hospitals, an edict was issued from Moscow that gave a quota on the number of people who could die under official care. The result was hospitals hurling people on their deathbeds out the front door and down the steps to die. They complied with the plan but missed the larger point.
Something similar may be happening with the high dropout rates. Teachers and administrators are probably encouraging failing students to leave school rather than drag down the aggregate numbers. One public school teacher revealed to me her tactic for dealing with the mandatory 90 percent pass rate. When she enters a class of 30, she identifies the 3 students she can ignore and otherwise write off as obvious failures. Now, this is not a cruel woman, just a person who knows what's necessary to survive in the new environment. She said all her colleagues do the same.
So while the results seem at first counterintuitive—how can testing lead to lower scores and more failures?—when you think about it, the results make perfect sense. Students are being run through a cruel gauntlet of narrow examinations produced by the politicians, while teachers are robbed of the ability to deal with the students as individual learners. This system might appear fine for the average student but the data can be deceptive. High-end and low-end learners are being neglected and those who ostensibly benefit are only given the tools necessary to master exams.
How do conservatives respond? They first point out that a lead researcher in the study, David Berliner, is a critic of school vouchers, and that the study was underwritten by an affiliate of the National Education Association, which opposes do-or-die tests. In other words, they are saying the people opposing this central plan are partisans of another central plan.
Granted. But what about the substance of the results? Chester Finn, an education official under Reagan, had this to say to the New York Times: "You almost never have a pure cause-and-effect relationship. Yes, you're introducing high-stakes tests, but maybe you're also changing the way you license teachers, or extending the school day, or changing textbooks. There's always a lot of things going on concurrently, so you really cannot peg everything to the high-stakes tests."
Aside from observing that his critique applies to all social-science research, which always and everywhere involves human volition and infinite variables, Finn's argument misses the point. The case for the tests was in part driven by the desire to be able to measure results in precisely the way the Arizona study has done. If you live by the data sword, you have to be willing to die by it too, and it is hard to argue against the reality that the new data has produced some very deep cuts.
Most compelling about the study is how it comports with anecdotal evidence. Teachers and students these days are obsessed with the tests, way beyond anything that anyone over the age of 30 knows anything about. The basics—reading, writing, math, science—are hammered home like never before. Preparation for tests has become the sum total of all public-school education. Advanced students are bored out of their minds, while weak students are relentlessly frustrated. Teachers wonder why they spent so much time learning how to teach, when all they end up doing is drilling for exams.
This new system is unsustainable, especially now that it turns out that the results produce the opposite of what it intended. Now, that is not to say that the alternative of left-liberal education policies—with no tests and no focus on basics or accountability—is the answer. The problem with education is more fundamental: it is run according to a central plan, so it has all the classic failures of central planning, including vast expense, vast waste of material and human resources, and results that are always disappointing.
The whole subject of education and the institutions that support it needs to be rethought, away from the still-surviving Deweyite-Progressivist model and toward the ancient tradition of private tutoring now being revived in homeschools across America. All schools can learn from the experience of homeschools, with their attention to individual needs, the flexibility that allows students to develop in unique ways, their privately run and funded character, their employment of localized knowledge and resources. These are the elements that make for good institutions of all sorts, whether it is commercial businesses, charities, civic institutions, or schools.
In short, the answer is not to adopt yet another central plan. It is to disempower the planners altogether, and restore decision-making power back to the parents, the teachers they employ, and the students. Testing and better data will not save education in America. A wholesale repudiation of all educational centralized planning will.