Flawed Diagnoses and Inappropriate Cures in Education

I don’t mean to pick on Joel Klein, the former New York City schools chancellor, but he has made himself such a caricature of self-styled school reformers who are undermining American public education that it would be a mistake not to respond to the claims on which he bases his efforts.

Last year, I addressed Mr. Klein’s conclusion that public education must be failing because he himself grew up in public housing as a “kid of the streets,” yet owed his success to great public schools; and if only children from public housing projects today had schools as good as his, they too would be successful.

The analysis, it turned out, was misleading. The New York City public housing in which the Klein family lived in the 1950s was segregated, constructed for white middle class two-parent households where the husband had a stable employment history and where market rents were charged with no public subsidy. Such housing projects no longer exist, and the conditions in which Joel Klein grew up bear no resemblance to those from which minority children in impoverished families come to school today.

Now, Joel Klein heads a division of Rupert Murdoch’s News Corporation, whose purpose is not only to sell internet-connected tablets to schools but to sell an entire tablet-based approach to curriculum and instruction. This week’s New York Times Magazine describes this effort, including extensive quotations from Joel Klein interviews conducted by the author, Carlo Rotella.

I have no opinion about the merits of tablet-based instruction, but I’m writing here because Mr. Klein’s justification for his product is based on gross misrepresentations of the state of public education in America. We should always be skeptical of treatments based on faulty diagnoses of the problems they are designed to solve. The treatments may work, entirely by accident, but this is unlikely.

Here is how Joel Klein, in his interviews for the magazine article, described why it is necessary to revolutionize American education with tablet-based instruction:

“K-12 isn’t working …and we have to change the way we do it… Between 1970 and 2010 we doubled the amount of money we spent on education and the number of adults in the schools, but the results are just not there. Any system that poured in as much money as we did and made as little progress has a real problem. We keep trying to fix it by doing the same thing, only a little different and better. This [tablet-based instruction] is about a lot different and better… We’ve spent so much on things that haven’t worked,” [he said, making a list that included underused computers as well as obsolete textbooks, useless layers of bureaucracy and smaller class sizes].”

Unfortunately, Mr. Rotella did not press Joel Klein for the basis of these assertions, so central to a belief that public education needs to be transformed by the technology he is selling.

In truth, the assertions are based on little fact, and turn out only to be the recitation of modern myth. This is what research actually shows:

  • It is true that money spent on education has doubled since 1970, but only about half of this increase has been devoted to improving the academic education of regular students. The other half has mostly gone to special education for children with disabilities who were not entitled to a free public education in 1970. We now spend a lot of money on a lot of adults—special education teachers with very low pupil-teacher ratios—who are dealing not only with learning disabilities but with children who have severe emotional, cognitive, and behavioral problems. It is foolish, as Mr. Klein in effect does, to claim that because we are now spending so much money on children with disabilities, schools must be failing because the spending has not caused the achievement of regular students to improve.
  • Yet the achievement has improved, and dramatically. Nobody knows why it has improved—perhaps the other half of the spending growth devoted to regular education has played some role. Our only sources of information about trends in academic achievement are two sampled tests sponsored by the federal government, the National Assessment of Educational Progress. One, a multiple-choice test of more basic skills, shows that academic achievement for black children has improved so much that black fourth graders nationwide now have average basic skills proficiency in math that is greater than that of white fourth graders in 1970. The other, a test requiring original computations and written answers, shows the average academic achievement of black fourth graders to be greater than that of white fourth graders in 1990. Improvements have also been substantial in reading, and for eighth graders. White students have improved as well, so the black-white test score gap has not changed very much, narrowing only to the extent that black achievement has been rising faster than white achievement.
  • Because the assumption that schools are failing is so frequently repeated and accepted, without evidence to support it, there has been almost no effort by scholars to understand the causes of the dramatic improvements that have, in fact, occurred. For example, Mr. Klein asserts that smaller class sizes for regular students haven’t worked. This, too, is only the incantation of conventional wisdom, but is not what the research shows. The only scientifically credible study of class size reduction, an experiment conducted in Tennessee 20 years ago, found that smaller classes were of particular benefit to disadvantaged children in the early grades, but without similar benefits for middle class children. There have been no comparable studies of class size reduction for older children.1

The assertion by school reformers—that their treatments are necessary because the patient is dying—echoes in the proclamations of many others, not Joel Klein alone. Influential education policymakers tell this to each other so often that they apparently never wonder if it is well-founded. Another leader of the reform crusade is Bill Gates, whose foundation has funded many of the initiatives advocated by Joel Klein and other reform leaders. In a 2011 Washington Post op-ed, the Microsoft founder said, “over the past four decades, the per-student cost of running our K-12 schools has more than doubled, while our student achievement has remained virtually flat.” When shown evidence that the statement was false, Mr. Gates’ office could provide no contrary evidence and simply replied that he was talking only about high school students. Yet even for high school students, there has been improvement. For example, in these four decades, the share of young adults who graduate from both high school and college has doubled.

Of course, like any institution, public education should be improved. We should be able to do much better. But some, perhaps many of the things American schools have been doing have turned out to be quite successful. By making a blanket charge of failure and proposing to overturn the entire enterprise, whether in favor of tablet-based instruction, charter schools, short-term teachers, or private school vouchers, the reformers may well be destroying much of what has worked in favor of untested fads.



1.  An earlier posted version of this blog incorrectly asserted that “The only scientifically credible study of class size reduction, an experiment conducted in Tennessee 20 years ago, found that smaller classes were of particular benefit to disadvantaged children in the early grades, but without similar benefits for middle class and older children.” As Leonie Haimson, an advocate of class size reduction, reminded me, the Tennessee experiment did not include children older than grade 3, so did not find that there were not similar benefits for older children. As the corrected version of the blog states, there have been no comparable studies of class size reduction for older children, and it is possible that older children also benefit from smaller classes. Joel Klein’s statement that class size reduction has not worked remains without evidentiary support.