黄色电影

Last month, The74 ran a  on a newly launched education funding non-profit, creatively named EdFund. EdFund is heir to EdBuild, which The74 described as a “one-of-a-kind … groundbreaking non-profit dedicated to advancing equity in education funding.” On the likelihood that these organizations are two-of-a-kind, policymakers should be prepared for the education finance conversation to be predicated on even more statistical deceit.

EdBuild’s reports consistently made national headlines in the education press, because it twisted data to create education reporter catnip: the opportunity to write stories on the theme that “schools are racist and need more money.”

For example, EdBuild published a report in 2019 titled “Fractured: The Accelerating Breakdown of America’s School Districts.” EdBuild wanted to raise the alarm that communities creating new school districts were “using secession to segregate students [and] to undermine equity and efficiency.” In the past 20 years, about seventy new school districts were created. Half of them were in Maine (where, you can be sure, racial segregation was not a driving motive). In the middle of the last century, over 100,000 school districts were consolidated out of existence. Does the creation of about 30 new ones outside of Maine in two decades reflect a crisis? Maybe… if you’re a reporter at a publication funded by the same philanthropies that funded EdBuild.

EdBuild also published a paper claiming that there was a $23 billion gap between school districts that were largely white and those largely non-white. That sure sounded like systemic racism to reporters. It was well-known that states don’t fund Black students at a lower level than white students. States like Massachusetts and Arizona are not systematically spending less on black students. But the largely white Newton, MA does spend more than the largely minority Tuscon, AZ.  You can’t really look straight at that and call it racist. But if you hide the fact that you're comparing districts in high-spending states to districts in low-spending states, you sure can spin it.

EdBuild and now EdFund CEO Rebecca Sibilia once took to the pages of the New York Times to decry America’s alleged unfair system of local school finance. Of out approximately 13,000 school districts in America, Sibilia juxtaposed two nearby districts in New Jersey. In Gloucester City, the median property is worth about $120,000, “but four miles away in Haddonfield Borough a median home sells for $500,000. From this wealthy tax base, Haddonfield can raise $13,500 per student, four times higher than what can be collected in Gloucester City.” You could read her entire piece and assume that the richer school district spent more money per pupil. But the exact opposite was the truth. The more-affluent Haddonfield Borough spent  per student annually, while the less-affluent Gloucester City spent . 

Rather than produce their own research and op-eds, EdFund will now fund and promote such research and public affairs work. The last thing that the study of school finance needed was another pressure to produce pro-spending, narrative-friendly work. In the past decade, academics have applied more complicated econometric methodologies and found that, contrary to received conventional wisdom, money really does matter a lot.

For example, Northwestern’s Kirabo Jackson studied the financial effects of the Great Recession and concluded that they proved that money mattered greatly. He also reviewed all of the academic literature in a meta-analysis and concluded that virtually all the rigorous research proves that this is true. When academics have looked under the hood, however, they’ve found severe flaws in his work. The University of Arkansas’s Josh McGee (who served on the board of EdBuild and serves on the board of EdFund) documented that Jackson could have organized their data in many reasonable ways. Most of those ways would have shown no effects. The arbitrary way that showed an effect was published. Similarly, the Heritage Foundation’s Jay Greene wrote an expert witness testimony report credibly alleging to expose real-time p-hacking in Jackson’s meta-analysis.

Neither work seemed to make much difference amongst education academics. Perhaps because they are, as a group, driven less by a desire for the truth than a desire to produce socially desirable and politically congenial results. This social and ideological pressure to produce a bogus consensus pre-dates EdFund, but it’s sure not to be helped by the new money it will throw at such studies.

Cannily, EdFund plans to “white label” what it produces so that its fingerprints will not be directly on the various graphs and factoid graphics it produces. But, fortunately, aside from financially reinforcing an intellectual bubble, it’s far from clear that its work will have any effect.

Democratic lawmakers don’t particularly need to point to an argument that spending money helps no matter how it’s spent, because they want to spend money no matter whether it helps. And Republican lawmakers are, broadly, hip to the intellectual and ideological corruption of education researchers and advocates. Still, if they need more reason to distrust the experts on school finance, EdFund will provide it.