Riley's Class Size Nostrum

U.S. Education Secretary Richard Riley came to Vermont on November 8, peddling what is probably the single worst idea for improving public education in America. Ironically, Riley delivered his prescription at a conference sponsored by an organization called the Foundation for Excellent Schools.

Riley's proposal is "smaller class sizes". It has three notable characteristics. First, there is no evidence whatsoever that this "reform" makes any contribution at all to educational outcomes above the kindergarten level. Second, it is the single most expensive "reform" available, competing only with "more school days" for that distinction. Third, the enthusiasm for "smaller class sizes" prevents real educational reforms which, though far more promising and less costly, thoroughly disturb the public educational establishment.

The smaller class sizes proposal has one more crowning feature: everybody believes it is good for the education of kids. Parents love it. Administrators love it. Teachers love it. The teachers union loves it. Poll respondents love it. With these credentials, the Clinton Administration obviously loves it, and even Congressional Republicans learned to love it once Clinton and Riley explained it to them, especially the part about the poll respondents.

To buy into the "smaller class sizes" nostrum, one has to believe that it produces educational benefits. Obviously, one on one with Horace Mann promises more benefit than a seat in a classroom with 100 other kids. But that isn't the choice. The choice is a class of 25 versus a class of 15. Does the smaller class produce better educated children?

Prof. Eric Hanushek of the University of Rochester is one of the top educational economists in the country. Last February he published an important monograph entitled "The Evidence on Class Size". After reviewing scores of achievement studies, Hanushek concluded that "we have had extensive experience with class size reduction and it has NOT worked... International experience suggests NO relationship between pupil teacher ratios and student performance... Extensive economic investigation shows NO relationship between class size and student performance."
(Emphasis in the original).

"Existing evidence indicates that achievement of the typical student will be unaffected by instituting the types of class size reductions that have been recently proposed or undertaken," Hanushek concluded. "The most noticeable feature of policies to reduce overall class sizes will be a dramatic increase in the costs of schooling, an increase unaccompanied by achievement gains."

Hanushek looked closely at the one study upon which smaller class size advocates base their case, the Tennessee Project STAR. The most that it shows, he observes, is that a one third reduction in kindergarten class size (23 down to 15) produces a small one time only gain in first grade.

David Kirkpatrick observes that Jaime Escalanate, the celebrated Los Angeles teacher who made contest-winning math students out of "disadvantaged" children, taught classes of 70, until forced out of his job by resentful union teachers who couldn't get his results in classes of 20. In South Korea, whose students ranked first in math in a 20-nation study, the average class size is 43.

Former Assistant Education Secretary Chester Finn explains why shrinking class size produces no educational results: "The teachers [in the smaller classes] don't do anything differently. The same lessons, textbooks and instructional methods are typically employed with 18 or 20 children as with 25 or 30. It's just that the teacher has fewer papers to grade and fewer parents toconfer with." No wonder teachers like it. And of course the teachers' unions love it, because smaller class sizes mean more dues-paying union members.

Finn says that instead of hiring 100,000 new teachers to teach smaller classes, the country would be far better off firing the 100,000 worst teachers and putting their kids into larger classes taught by better teachers.

Schools that go from twelve 25-child classes to twenty 15-child classes immediately need eight new teachers, requiring an additional compensation package of at least $250,000 (assuming first year teachers). They also need up to eight new classrooms, which typically means an addition to the school. This is a lot for the taxpayers to pay for, year after year, in return for no identifiable learning outcomes.

The hard truth is that public schools will not improve by hiring lots more teachers with the same questionable qualifications, using ineffective instructional methods in additional classrooms. Yes, the politicians will score points with voters concerned about education, and the unions will get more members from which they can raise more political action funds to elect more legislators who will mandate smaller class sizes. But the taxpayers will get larger tax bills, and the kids - whose education is the rationale for Vermont taxpayers spending $800 million a year - will come out with no identifiable gains. This is a reform?

A real reform agenda is much less expensive: create competition among providers for the customer's education dollars, and tell the public education establishment that the life of Riley is over.

# # #

November 1998