Cases in which parents deny their children modern medical treatment are increasingly rare. In medicine, the days of snake‐​oil selling quacks are mostly behind us. Sadly, the same isn’t true in education policy.


Medical researchers precisely define and test their proposed treatments. Compare that to a recent bit of education policy “analysis” in which the writer purports to assess Milton Friedman’s market‐​inspired proposal (minimally regulated school vouchers) by reviewing the outcomes of charter schooling. This is like testing insulin by administering Flintstones Chewables. Charter schools are opened and closed at the discretion of government authorities, lack market‐​determined prices, and cannot be operated for‐​profit or offer religious instruction. In many states, they cannot hire teachers who lack government credentials. Friedman’s voucher proposal shared none of these characteristics, and so to treat the two interchangeably is a sign of ignorance or intentional equivocation.


Even when relevant evidence is presented, the presentation is frequently inaccurate and unsystematic. To see just how serious this problem is, it helps to look at an example in detail. Consider a recent discussion of voucherizing U.S. federal education spending that drew lessons from Chile’s voucher program. Many of its facts are wrong, others are misrepresented, and key pieces of information are omitted.


The author claims that Chilean education spending as a share of GDP shrank between 1980 and today. But, according to the United Nations, it rose from 4.4 percent to 4.5 percent. And, due to the sustained growth of Chile’s economy since the mid‐​1980s, inflation‐​adjusted per pupil spending has more than doubled.


The author acknowledges that Chilean students are now the highest‐​performing in Latin America, but claims that his fabricated “budget cuts have led to overall decline in quality.” In fact, Chile is one of the fastest‐​improving nations in the entire world on international tests of academic achievement. He goes on to claim, without support, that vouchers have led to growing inequality, benefiting only upper‐​middle‐​income families, yet a Yale University study reports that the voucher program has reduced inequality in educational attainment and raised earnings equally for both the poor and the non‐​poor.


Finally, the author notes that lower‐​income students are more likely to attend public rather than private schools in Chile, but neglects to mention that public schools serving the poor receive a varying amount of additional funding that is not given to private schools serving similar students. Chilean economists Sapelli and Vial report that public schools receiving vastly higher funding per pupil outperform private schools (which explains their appeal), but in the rare cases in which the public sector’s funding advantage is 25 percent or less, it is private schools that perform better.


This is not an exhaustive list of the commentary’s errors, omissions, and misrepresentations, but it should suffice to show the level of quackery being doled out to the public by purportedly serious publications (it was published in the Washington Post’s education blog). We’re not exactly talking House or Doc Martin here.


Few parents would administer the medical equivalent of this claptrap to their children–they are generally protected from such errors by the health‐​care field’s comparatively careful, systematic research practices. But in education, they still suffer under the ministrations of charlatans. The result can be seen in the virtually unique productivity collapse that has beset American education for generations.


So what can we do about it? A first step would be for well‐​intentioned education policy analysts to make more systematic use of the high quality research that is available, and to add to that literature. But it is harder to conduct experiments on the impact of state or national policies than on the impact of drugs. Fortunately, there is a solution to this problem–one that we also owe, incidentally, to the medical field. I’ll be writing about that soon, and will update this post with a link when it’s available.


Update: My article in the Washington Post’s Answer Sheet blog.