More On Jay Greene And The United Cherry Pickers

Jay Greene and his ideological camp followers may not want to take a vacation from “cherry picking” education research, but I am on vacation with my family, and they could care less about Greene’s toiling in the fields of research in search of cherries, so this reply will be quick and to the point. His latest contribution involves textbook examples of the research “cherry picking” we identified in our original post.

Case One: Hank Levin’s research

Greene quotes a couple of Levin’s lines about education competition in general and suggests that this means he approves of vouchers. There is, of course, a considerable difference between a discussion of competition in a general way [which may include such phenomenon as charter schools and public school choice within a district] and a specific discussion of vouchers. Hank Levin has stated his assessment of vouchers directly in numerous forums, many of which are readily available on the Internet. Perhaps none is more definitive than an article in the Journal of Policy Analysis and Management entitled “Educational Vouchers: Effectiveness, Choice, and Costs.” Levin’s précis:

Most of the policy discussion on the effects of educational vouchers has been premised on theoretical or ideological positions rather than evidence. This article analyzes a substantial body of recent empirical evidence on achievement differences between public and private schools; on who chooses and its probable impact on educational equity; and on the comparative costs of public and private schools and an overall voucher system. The findings indicate that: (1) results among numerous studies suggest no difference or only a slight advantage for private schools over public schools in student achievement for a given student, but evidence of substantially higher rates of graduation, college attendance, and college graduation for Catholic high school students; (2) evidence is consistent that educational choice leads to greater socioeconomic (SES) and racial segregation of students; and (3) evidence does not support the contention that costs of private schools are considerably lower than those of public schools, but the costs of an overall voucher infrastructure appear to exceed those of the present system.

Bottom line: Levin’s view of vouchers is quite critical, rather different than what Greene suggests.

Case Two: Cecilia Rouse’s research on the Milwaukee voucher experiment

Cecilia Rouse’s research on the Milwaukee voucher experiment is critical of a publication making a claim of strong achievement gains for students using vouchers. You would never know by reading his account, but the publication making these claims was written by Jay Greene. Rouse argues that his account of improvements in reading could not be sustained, as the test score differentials were as often negative as positive when students using vouchers were compared with students who did not use vouchers. In math, Rouse found a modest advantage for students using the vouchers. She notes that the database was deficient, with a significant number of missing students, but believes the results are still valid, based on two statistical and econometric methods she uses to verify them. In her conclusions, she offers three caveats concerning the reliability and robustness of her findings in math, and specifically cautions against using them to conclude that vouchers work.

Bottom Line: Greene’s claim that Rouse’s analysis is supportive of his pro-vouchers position is strongly misleading.

 Print