A counterintuitive take on school reform
Before we can fix what's wrong with public education, we have to be absolutely clear on what exactly is wrong, says Jay Greene, author of the new book Education Myths: What Special-Interest Groups Want You to Believe About Our Schools—and Why It Isn't So. Greene, head of the brand-new department of education reform at the University of Arkansas, examines 18 widely held beliefs—that schools are desperately short of money, that they can't do much to overcome the effects of poverty, that smaller classes are automatically better, that teachers are underpaid, and more—and explains why he thinks they're misguided ideas getting in the way of change.
Not everyone agrees with Greene's conclusions, for sure. Critics says the outspoken Harvard-educated scholar, who has been affiliated since 2000 with the free-market-oriented Manhattan Institute, is hardly a disinterested observer; one reviewer writes that he "enjoys punching left but generally avoids criticizing the right." But Greene has become influential beyond conservative circles: His critique of how public schools miscalculate graduation rates has played an important role in the high school reform debate, and he notes that he doesn't hew to the right's line on issues like bilingual education (his study found that students benefit when they study for part of the day in their native language). He is also a Democrat who says he briefly registered as a Republican only in order to vote for John McCain against George W. Bush in the 2000 GOP primary. Recently, he shared his thinking with U.S.News & World Report Senior Writer Anne McGrath.
Why a book about education myths?
I've found it frustrating when I speak about education to an audience or to reporters that so much of what they see as "fact" is plausible but inconsistent with the bulk of the evidence.
Believing in myths hurts children because if we're not aware of the magnitude of the problem, we're not properly addressing it. The myth that nearly everybody graduates from high school today, for example, means that we're not addressing the fact that 30 percent of our students don't graduate. The most common myth about education is that our schools are desperately in need of more money before they can do better. It's not that more money would be unhelpful, if it were used wisely—it's just that we haven't figured out how to get schools to use money wisely. Just giving schools more money and allowing them to continue to spend it as always blinds us to other things we may need to be doing, like figuring out how to spend money more effectively.
As evidence of ineffective spending, you point out that per-pupil expenditures have doubled over the past 30 years and that achievement scores have remained flat. But doesn't that comparison ignore the fact that schools are dealing with greater demands? Tougher social problems, for instance?
There is no doubt that family background has a profound effect on student achievement, in the same way that obesity has a profound effect on heart disease. But there's a line that people cross all the time between saying "we know background characteristics affect achievement" and saying "background characteristics make success impossible." When people cross that line, they're promoting the "myth of helplessness." We can and should expect schools to make a difference with students, no matter where they begin, in the same way that no matter how overweight or underweight you are, when you go to the doctor we rightly expect the doctor to make a difference. If the doctor can't make a difference, then it's a waste of your money.
There's a great variety of successful programs out there that are able to produce gains in learning for disadvantaged students no matter what their initial level of achievement is: KIPP [Knowledge Is Power Program] and Success for All, for example. They tend to establish high expectations—they don't say to themselves or to the students [that] "these kids have such great difficulties that we can't really be expected to do anything with them"—and they tend to have highly structured curricula.
So are you a proponent of giving schools credit under No Child Left Behind for their students' growth as well as absolute achievement?
Yes, I am. Schools really need to be judged on the value they add, the student learning that takes place. We can measure that—imperfectly, but we can measure it—and we need to reward schools and educators who are able to contribute to greater learning.
What do you say to those who point to the exorbitant cost of yearly testing under NCLB as proof that schools are underfunded?
In fact, the cost of developing, administering, and reporting test results is estimated by Caroline Hoxby at Harvard to cost around $20 per pupil—while total per-pupil spending now averages close to $10,000 nationwide. Testing is relatively cheap.
How do we get schools to use their resources better and improve results?
In general, we need to attach performance to revenues, as we do in every other field. So schools that perform better ought to be able to earn more revenue. School choice allows schools to try to improve revenue by doing a better job. Accountability systems provide incentives by attaching rewards and sanctions to the performance of students on standardized tests. Right now, schools, administrators, and teachers have no—or very few—rewards or sanctions attached to whether they make wise or unwise choices. They can choose bad curricula, bad teaching approaches, bad compensation systems, and there are no consequences. Accountability and choice provide schools with incentives to use their money more effectively in ways that contribute to greater student learning: things like finding pedagogical techniques that are more effective, and identifying and rewarding excellent teachers.
You say that it's a myth that teachers are underpaid. But aren't notoriously low salaries a big part of the problem in finding and keeping the excellent people?
I have nothing against paying teachers a lot more, but we have to start that discussion with an understanding that the average teacher is making a salary that is similar on an hourly basis to what other professionals earn. They make about $31 an hour, more than civil and mechanical engineers, chemists, biologists, and accountants.
Part of our difficulty in attracting and retaining the best people in teaching is that we have flat pay scales that pay good teachers and bad teachers alike. Because we do not distinguish, we end up with an average, and this is very frustrating to successful people, who like to go into fields where their excellence is recognized and rewarded. Right now, teachers work harder only out of their goodwill. Luckily, we have a lot of teachers of goodwill out there, but you can't run an entire system on goodwill. You have to reward people for excellence. The most important thing in improving educational achievement is improving teacher quality.
One big difference between public and private schools—and a reason private schools get better results—is their incentive and ability to identify teachers who improve learning and to get rid of less effective teachers.
Can't the better results also be explained by the fact that private schools have more money and can choose their students?
That's the "Exeter myth." Every city has one or two highly selective elite academies, but the vast majority are not very selective at all and really can't afford to be. And the expulsion rate in public and private schools is about the same, about 1 percent. Most private schools are not in the business of turning away customers or expelling them. In particular, a lot of Catholic schools manage to perform well with remarkably little money.
There's lots of disagreement about whether voucher programs, which allow families to choose private schools using public funds, serve kids well. Why do you favor vouchers?
The "inconclusive research myth" says that we don't really have any idea how vouchers affect the academic achievement of students who receive them—that there are just as many negative effects as positive effects, and it's just a mess, and nobody really knows the answers.
While it's true that researchers disagree about various things in their studies—often forcefully and in print—there are some common results across all random-assignment studies of vouchers. Whenever you hold a lottery to allocate spots in a program, you have a random-assignment experiment: Everyone applied, everyone had the same level of motivation, and, on average, the two groups will be exactly alike. All eight studies show positive academic effects for voucher recipients. Seven of the eight have statistically significant positive results for at least some subgroups of students on some tests. Results may be mixed in terms of the scope and magnitude of benefits, but none of the studies found evidence that students' test scores were harmed in voucher programs.
President Bush has been criticized for his proposal to grant funding to Katrina victims that they can use at a school of their choice. What do you think?
I think that it's important that we figure out how to provide a quality education to all of the children who were displaced by the storm and however we can do that most effectively, I think we need to consider. We can't let prior practices or old political arguments stand in the way of trying to deliver the best services as quickly and effectively as possible. There was a very large Catholic school population in New Orleans that is all displaced, and there are public and private school systems in Texas and elsewhere that have been overwhelmed with new students. I think there is a logic to the cost-effectiveness of helping displaced families by providing them with the resources to find the best schools for their children, public or private, religious or secular.
Creating small high schools has become a popular reform strategy. But you say it's a myth that small classes produce big improvements. Why?
My work doesn't speak to the small schools movement but to the very expensive movement to require class-size reductions statewide. This has occurred in California, Florida, and elsewhere. All else being equal, it's better to have fewer other kids in the class. But the problem with these across-the-board class-size reduction mandates is that they essentially force school systems to go on teacher-hiring binges. So the benefit of fewer children in the class is offset on average by a reduction in teacher quality as school systems dig deeper into the labor pool. In California, there were no benefits observed, and yet it's an extremely expensive intervention. A one-third reduction in class size requires roughly a one-third increase in per-pupil expenditures.
Frankly, we don't have a whole lot of high-quality evidence on small schools yet.
Do you see any reason to be optimistic about where we're headed?
I actually see a lot of reason for optimism. It's important to note that student achievement has not gotten worse. There's a "myth of decline" that it's important to get out of the way as well—people pine for a golden age when schools were better and kids were better. It just didn't exist.
The ideas of accountability and school choice are becoming more acceptable. Because the evidence shows these reforms have promise, we have reason to expect that they will be yielding better results in the future as they are adopted on a broader scale. Gradually, over time, school systems will experiment with merit pay, greater choice, different kinds of accountability systems. Through trial and error and careful examination of the evidence, they will gravitate toward more-effective policies and discard less effective ones.