Wednesday, March 24, 2010

Diane Ravitch on Mayoral Control

If Diane Ravitch insists that there's no evidence on a particular point, you can be almost certain that there is. Here's her piece on mayoral control, from a recent issue of Phi Delta Kappan:
Matt Miller of the Center for American Progress . . . argued that local control and local school boards are the basic cause of poor student performance. . . . In an ideal world, he wrote, we would scrap local boards and replace them with mayoral control, especially in urban districts. This one act of removing all democratic governance, he claimed, would lead to better education. . . . There is not a shred of evidence in Miller’s article or in the research literature that schools improve when democratic governance ends.
Once again, Ravitch misrepresents the literature. For example, there's Kenneth Wong's study of mayoral control, which found that "mayoral control has a statistically significant, positive effect on student achievement." Granted, Wong's study may be imperfect and it may be difficult to properly measure something as nebulous and potentially endogenous as mayoral control. But trying to refute Wong would be more defensible than claiming definitively that studies like his don't even exist.

P.S. If you're going to discuss scholarly literature with which you're not familiar, the wiser approach is to say, "I've never seen convincing evidence that such-and-such," which leaves you two easy outs: if anyone points out a study, all you have to do is note that you hadn't personally seen it, and/or that you don't find it convincing.

P.P.S. The Wong article above appeared in a book to which Ravitch herself contributed an article. So Ravitch had to know that her "not a shred of evidence" comment was false.



Blogger Malcolm Kirkpatrick said...

NCES does not make available to unaffiliated individuals NAEP scores for individual students, schools, or school districts. To assess the effect of district size on student performance, I use State-level NAEP 4th and 8th grade Reading and Math scores (composite, Numbers and Operations, and Algebra and Functions) and three indirect measures of district size: mean district size, the fraction of total government-school population enrolled in districts over 15,000 (or 20,000, depending on which year of the __Digest of Education Statistics__ you use) and percent of total government-school population enrolled in one or another of the nation's top 130 largest school districts. The correlation is negative. Scores fall as districts increase in size.

According to international evidence, countries which subsidize a parent's choice of school outperform countries which restrict parents' options for the use of the taxpayers' pre-college education subsidy to schools operated by State (government, generally) employees. Homeschoolers outperform conventionally-schooled children.

From these and other lines of evidence, I deduce the following generalizations:

1. As institutions take from individual parents the power to determine for their own children the choice of curriculum and the pace and method of instruction, overall student performance falls.
2. Political control of school harms most the children of the least politically-adept parents.

Homeschool. Extend daycare to age 18 if you cannot afford to sacrifice an income to homeschool.

4:12 PM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home