President Obama has repeatedly promised to use an “evidence-based approach” for social policy—and when it comes to education, he has been true to his word: He has systematically promoted programs such as universal pre-school with little evidence of success and panned ones such as school vouchers with lots.

In his recent State of the Union address, the president—not for the first time—hectored Congress to “make high-quality preschool available to every single child in America.” “Study after study shows that the sooner a child begins learning, the better he or she does down the road,” he insisted.

Actually, “study after study” has shown the exact opposite—that publicly funded preschool programs make no lasting difference in a child’s life.

Consider Head Start, the nearly half-a-century old early learning program targeted at low-income toddlers. About a million kids are enrolled in the program every year and Uncle Sam spends about $8,000 on each, not exactly chump change. Yet a majority of studies have found that while these kids show initial cognitive gains that make them more “school ready,” these gains disappear once they enter regular school. Pre-K believers have pooh-poohed these studies on methodological grounds arguing that they did not track the kids long enough and weren’t based on random assignment with a valid control group.

None of these objections apply to the Department of Health and Human Services’ December Head Start Impact study. It is the most ambitious and expensive evaluation of the program that the administration did its best to bury by releasing it on the Friday before Christmas.

The study compared kids who applied and got into Head Start through a random lottery with those who applied and didn’t get in, thereby controlling for the motivation level of parents and other intangibles. It also followed these kids up to third grade instead of just measuring school readiness for first grade. Yet it found “very few impacts” in any of the four domains of “cognitive, socio-emotional, health and parenting practices.”

Head Start is not the only preschool program that has failed to deliver on its promise. Oklahoma implemented universal preschool in 1998 when its fourth-grade reading score on the National Assessment Education Progress—the nation’s report card—was five points above the national average. Now it is five points below. Georgia’s fourth-grade NAEP reading score have improved—but 21 years after it embraced universal preschool it ranks 48th in terms of graduation rates.

How about minority kids in whose name these programs were justified? In both states, fourth-grade black math and reading NAEP scores were above the national average of black students in other states. Now they are at the national average.

So why does President Obama claim that every one dollar in early education saves $7 later? He has in mind Michigan’s 1962 Perry Preschool Program and North Carolina’s 1972 Abecedarian programs whose participants posted significant gains on language and math skills during the school years—and reduced crime, welfare use, and higher earnings later.

But these programs were the Lamborghinis of early education: They were super-expensive ($90,000 per child for Abecedarian), intense interventions where the best teachers and social workers targeted every aspect—parenting, schooling, nutrition—of 100 low IQ and neglected minority kids for several years, not just one year as is the case with regular preschool. By their very nature, they can’t be scaled up to a national program.

But if President Obama’s case for universal preschool is tendentious, his opposition to private school vouchers is mendacious.

At every opportunity he has tried to kill the Washington D.C. voucher program that serves about 1,600 poor, minority kids. For every kid admitted, there are four applicants trying to escape the violence-ridden school system that has among the highest dropout and lowest graduation rates in the entire country. Yet, within months of assuming office, the president signed a spending bill prohibiting scholarships to new students and requiring the program to be automatically scrapped unless Congress explicitly voted to renew it.  

Congress gave the program a five-year extension in 2011, but every year the president has tried to withhold its measly $20 million funding, half of what the government would spend on these kids if they stayed in public schools. His last two budgets initially included not a penny for the scholarship program even as his 2013 budget demanded $60 billion in additional education dollars as a stimulus measure.

What’s the administration’s defense of its step-motherly treatment of this program? “Private school vouchers are not an effective way to improve student achievement,” it insists. “Rigorous evaluations over several years demonstrate that the D.C. program has not yielded improved student achievement by its scholarship recipients compared to other students in D.C.”

Wrong again. These evaluations show the exact opposite. In fact, the DOE’s own 2010 study of the D.C. program found that 21 percent more voucher kids graduated than non-voucher kids and had more satisfied parents.

All of this is in line with studies of voucher programs elsewhere. Indeed, a meta-analysis of all the major voucher studies conducted by the Foundation for Educational Choice pointed out that nine out of 10 random-assignment studies—the most rigorous possible—found that vouchers improved reading and math outcomes of voucher kids. (The studies included in the Foundation's sample were conducted by such right-wing hacks as the New York Federal Reserve, Economic Policy Institute, and Department of Education!)

Voucher opponents allege that vouchers hurt public school kids by draining resources. The exact opposite seems to be the case. Indeed, out of 19 studies examining this impact, 18 found that competition by vouchers actually improved education outcomes in public schools too. “Every empirical study conducted in Milwaukee, Florida, Ohio, Texas, Maine and Vermont finds that voucher programs in those places improved public schools,” the Foundation’s analysis concluded.

Vouchers are cheap and effective whereas publicly funded preschool is expensive and ineffective. That’s what the evidence shows. And if President Obama wanted to be true to it rather than indulge his ideological fancy, he would push Universal Vouchers to improve student performance—not Universal Preschool. That might in fact free up public education dollars for more targeted interventions of genuinely at-risk kids while protecting taxpayers from yet another massive, new entitlement.