Choice and Markets: The Reprise
So, let's start with Adam Emerson's critique. Here's what he says:
Advocates for choice and competition in American education have for years encountered the straw-man argument that charters, vouchers and the like are ineffective because standardized-test performance in these sectors is mostly indistinguishable from that in the public schools (the reality is, of course, more nuanced, but more on that later). But [Tucker] has taken a more extreme path: he has made the bogus claim that no evidence supports the theory that school choice has any merit at all.
Emerson then goes on to cite a series of studies which, in his judgment, refute what he says is my claim that there is no evidence that school choice is effective.
But I never made any such blanket statement. What I said was, that choice and charters "have neither raised student performance nor lowered costs at the scale of a state, province or nation." That is a very different statement and there is nothing in the references provided by Emerson that would cause me to change what I said.
The most casual reader of Education Week knows that there are research studies on charters and vouchers that purport to show that, at the locations and in the periods in which the studies were done, the charter or voucher schools produced superior results. And they also know that that there is an equally voluminous series of studies that purport to show that, once the appropriate adjustments are made to make sure that similar students are being compared, there are no gains for charters or vouchers vis-a-vis regular public schools.
Diane Ravitch, in her book, The Death and Life of the Great American School System: How Testing and Choice are Undermining Education (revised and expanded edition), describes the most prominent examples of the research studies done on both sides of this divide. Most were of a small number of students in a particular city. But she also cites a major study conducted by a research team led by Margaret E. Raymond at Stanford University that looked at the data from 2,403 charter schools serving 70 percent of all the students at the time in charter schools. Her team found that 37 percent had learning gains significantly below those of public schools, 46 percent had gains no different from those in regular public schools and only 17 percent had gains exceeding those in public schools. "In the aggregate," they concluded, "charter students are not faring as well as their traditional public school counterparts." Another study, by Tom Loveless of the Brookings Institution in the same year, looked across all the studies done on charters by both sides in the controversy and concluded that some researchers found positive effects, some found negative effects, but on the whole 'none of the studies detects huge effects—either positive or negative." The Stanford study by Raymond and her team was funded by two foundations with a history of strong support for charter schools. In the same year that the Stanford study was released, another study was released based on the latest NAEP data. Ravitch tells us that, "after controlling for demographic and other variables, the study found that the advantages of private schools and charter schools disappeared, and, in some instances, demonstrated the superiority of regular public schools."
When the choice and charters movement was launched, its advocates confidently proclaimed that choice and charters would inevitably lead to greatly improved student achievement and lowered costs. After more than 20 years of charter schools and more than that of choice programs, what we actually have is Tom Loveless' finding that any gains attributable at scale to charters and choice are very small.
Now permit me to go back to what Adam Emerson actually said in his withering blast. It actually sounds like a concession speech. He does not deny my assertion that charters, vouchers and the like are ineffective because standardized test performance in these sectors is mostly indistinguishable from that in the public schools. He calls it a "straw-man argument" and says that the reality is more "nuanced." What is so "straw-man" about this argument? John Chubb and Terry Moe, in Politics, Markets and America's Schools, virtually launched the modern movement for choice and competition with the claim that choice and competition would by themselves greatly improve student performance and lower costs. It simply has not happened. And that was the point I was making in my blog. This is hardly a straw man argument. When Emerson says the truth is more nuanced, he is, I presume, referring to his argument that some of the research he cites claims to show evidence that charters may have a modest positive effect on the performance of regular public schools. Whether that is true or not, a question I will leave to others, it is a very long way from a strong claim that competition, choice and charters are the only interventions required to produce major improvements in student performance and reductions in the cost of education at scale. Indeed, I will stand by the claim I made in my piece that there is no evidence for a much more modest assertion, that charters, competition and choice will produce major improvements in student performance at the scale of a state, province or nation.
But that statement, according to Andrew Coulson, is not true. Coulson tells his readers that "the Netherlands, Belgium, Australia, and Canada all significantly outperform the United States in every subject tested [on the PISA assessments]....Could the author of the Education Week commentary possibly be ignorant of the Dutch and other examples that flatly contradict his claim?" I am then charged with "an oversimplified and flawed understanding of how to draw lessons from foreign educational experiences." He asserts that the research shows that "the most market-like, least regulated school systems have the biggest advantage over state school monopolies such as are the norm in the United States."
Really? Coulson refers to the literature he has reviewed in his Journal of School Choice. I wonder if he has actually visited the countries whose education strategies he so confidently cites.
After the second round of TIMSS data came out, showing the Netherlands and Flemish Belgium leading Europe in mathematics performance, I put a team together to go visit in both countries, to see if we could figure out how they did it.
Flemish Belgium is the Dutch-speaking part of Belgium. Its students strongly outperform the students in the German-speaking and French-speaking parts of Belgium. The organization of the Flemish education system and the strategies they use to get their results are very similar to the organization and strategies to be found in the Netherlands.
We were stunned by what we found there. It is certainly true that a small number of parents can get the state to put together a school for them, and it is also true that parents can freely choose among schools operated under the aegis of the state by private organizations like the Waldorf Schools and religious organizations like the Catholic Church. On this point, Coulson is absolutely right. But that is not what stunned us.
Consider the following. Our first stop was a visit with the mathematics inspectors in the Ministry of Education. There were five of them. When we talked with math teachers in the schools, they told us that the highest office to which a math teacher could aspire was inspector in the ministry. Wherever we turned, we observed that school professionals, university professors and independent observers all regarded the inspectors with enormous respect. It turned out that the inspectors were themselves responsible for setting the national student achievement standards for mathematics. In their "private" capacity, they were invited to chair the committees selected by the textbook publishers to write their mathematics textbooks. They set the national examinations in mathematics. And they were responsible for inspecting the schools. These inspections were not cursory. They took place over a period of days. The inspectors were specifically responsible for assuring that the curriculum as implemented in the school corresponded in detail to the curriculum as they had laid it out in very detailed and voluminous books of national standards for mathematics. They had the authority to shut down any school that failed to pass inspection. Though that rarely happened, the veiled threat was enough to assure the inspectors that their "recommendations," covering not just the curriculum narrowly defined, but other matters as well, were closely followed.
At the time of our visits, the mathematics curriculum in Flemish and Netherlandish schools followed closely the recommendations of Professor Freudenthal, a highly regarded professor of mathematics in the Netherlands, who, late in his career, had become interested in the teaching of school mathematics. In the 1970s, Freudenthal, through his Institute for Development of Mathematics Education, now the Freudenthal Institute, developed a powerful approach to the teaching of mathematics in the schools called Realistic Mathematics Education. The ideas, techniques and tools he developed rapidly spread through the research community, schools of education, schools and the Ministry. By the time we arrived on the scene, we observed that the implementation of these tools, techniques and ideas was more complete than any analogous development we had ever seen in any area of education in the United States.
It is absurd to attribute the position of the Netherlands and Flemish Belgium to choice, charters and competition in their schools. When we visited the Catholic schools in the Netherlands and Flemish Belgium, the only thing that distinguished them from the non-Catholic schools was the stained glass in their windows. What we found remarkable was the degree of consistency in the curriculum, the instructional methods being used, the texts and other materials. They were of a very high quality and much the same from school to school, no matter what organization, public or private, that the school was affiliated with.
But that did not surprise us, because by that time, we understood that the whole design of their education system was predicated on very strong control, from the Ministry, of the standards, curriculum, teaching methods, examinations and so on. No countries we have ever visited, anywhere, had tighter controls from the center on what schools did and how they did it.
The piece I wrote on choice, charters and competition was about both the theory and practice. The theory of choice in the United States was laid out by Chubb and Moe in the book I cited above. The intellectual foundations of the movement were set forth in Milton Friedman's 1955 essay on "The Role of Government in Education." Both the book and essay are founded on an aggressively libertarian premise. As Diane Ravitch put it, "Chubb and Moe wanted to sweep away 'the old institutions' and replace them with a new system in which almost all 'higher-level authority' outside the school was eliminated."
Let's be clear here. The charter school movement and the movement for more competition among schools was not about increasing the authority of the state over the schools. It has always been about decreasing the authority of the state over the schools. If Andrew Coulson had done his homework, if he had considered "the very different cultural, demographic, and economic conditions prevailing in different countries," then he could not possibly have come to the conclusion that charters, choice and competition in the Netherlands and Flemish Belgium are responsible for their superior performance. Coulson's conclusion that the research shows that "the most market-like, least regulated school systems have the biggest advantage over state school monopolies" simply flies in the face of the facts.
Yes, there is choice and something like a market in some of the top-performing countries. But those terms mean something very different in those countries than in the United States, and the larger policy context could not be more different. Their success is not a tribute to the libertarian ideal. It turns out to be a tribute to the opposite: strong central authority—exercised in systems that are thoughtfully conceived and well run.
I will repeat here what I have said before about school choice, charters and competition. I am not opposed to it. Nor am I for it. It is beside the point.
The lesson I take from the top-performing countries we have studied is that what matters most to the performance of those countries is whether they have equitable funding systems, the ability to recruit their teachers from the upper reaches of their graduating high school classes, teachers who have really mastered both the subjects they will teach and their craft, whether they have high academic standards for their students, a sound curriculum, high quality examinations that are well aligned with both standards and curriculum, good leadership and a system for educating their students all the parts and pieces of which hang together.
Almost all of these things are the result of what the state does. If the state does not play a strong role, it does not mean that all standards will be low, all teachers poor, all curriculum weak or all students funded at low levels. It means that these things will vary, often wildly, and is likely to mean that students who depend most on our schools in the race of life will get the least and those who start out with the greatest advantages will leave school even more advantaged.
That is the system we have now. There is no evidence that the libertarian ideal of schooling will improve it one whit.