That's how Joanne Jacobs characterizes the Obama administration's refusal to consider reauthorizing the DC voucher program. She's building off University of Arkansas education professor Jay P. Green's new City Journal piece, "So Much for the Evidence," which is closely aligned in tone--and far more detailed and hence damning--than the Washington Post staff editorial I linked to yesterday.

Here's Green:

In a major education address last March, President Obama declared that his administration would "use only one test when deciding what ideas to support with your precious tax dollars: it's not whether an idea is liberal or conservative, but whether it works." Unfortunately, the test that seems to guide the Obama administration's education priorities is not whether a policy works, but whether it serves a political constituency. Nothing illustrates this disregard for evidence better than the administration's treatment of two federally funded programs: the D.C. voucher program, which it is helping to kill, and Head Start, on which it has bestowed billions more dollars. If the administration actually made its funding decisions based on results, its positions would be just the opposite.

How do we know that the D.C. voucher program works? Take a look at the rigorously designed studies released by the Obama administration itself. Last April, the Department of Education put out its official evaluation of the voucher program. The evaluation, which used a gold-standard, random-assignment research design, found that after three years, D.C. students who won the lottery to attend a private school with a voucher significantly outperformed students who lost the lottery. The gap between voucher and control students was the equivalent of about five months of extra instruction in reading. Rather than embracing what manifestly worked, however, the administration stood by as Congress worked to phase out the D.C. voucher program. "Big picture, I don't see vouchers as being the answer," Secretary of Education Arne Duncan told the Washington Post. They're certainly not the answer that the pathologically anti-voucher teachers' unions wanted him to embrace.

Meanwhile, the administration fully supports the government-operated Head Start preschool program, despite excellent evidence that the program doesn't work. Obama has said that Head Start is "the first pillar of reforming our schools . . . [and] that's why the American Recovery and Reinvestment Act that I signed into law invests $5 billion in growing Early Head Start and Head Start." He might have added that this would come on top of the more than $100 billion that taxpayers have spent on Head Start since 1965. But the Department of Health and Human Services' official evaluation of Head Start, released last week, confirms what several earlier studies have found: kids get no lasting benefits from participating in the program. By the end of kindergarten and first grade, students who had been in Head Start are no further ahead academically or behaviorally than students who lost the lottery to enter the program.

The way the administration released the two reports also spoke volumes. The D.C. voucher study was released after a key congressional vote that declined to reauthorize the program--and the study came out on a Friday, without an official press release to draw attention to it. The Head Start findings, on the other hand, were not released on a Friday and came with a press release--but the release contained false claims from administration officials about the program's effectiveness. It quoted Assistant Secretary for Children and Families Carmen Nazario saying that "Head Start has been changing lives for the better since its inception" and Health and Human Services Secretary Kathleen Sebelius declaring that "research clearly shows that Head Start positively impacts the school readiness of low-income children"--even as the study showed that Head Start had done no such things. Again, the ideological priority to expand union-backed federal programs trumped an official evaluation, conducted, as with the D.C. voucher study, using a gold-standard, random-assignment research design.

If the administration really wants to show that it's guided by evidence and not ideology, it might consider changing its policy positions when solid evidence contradicts them. Empirical evidence shows that D.C. vouchers work; that program should be expanded, not killed. The evidence also shows that Head Start is a long-running failure; that program should be wound down, not funded with new billions. Even diverting a few hundred million from Head Start into a reauthorized D.C. voucher program would go some way toward restoring the administration's credibility.

On the subject of Obama's words coming back to bite him: I am reminded of a recent Jon Stewart sketch on the how the President needs to learn to make promises in such a way that no one can ever say he's not keeping them.