Neglected Evidence - educational research

Washington Monthly, Nov, 1998 by Richard Whitmire

While the Sanders method is used mostly to judge the performance of schools, the bigger breakthrough popped up when looking at teacher quality. In studying two Tennessee districts, Sanders was able to rank teachers into five categories of effectiveness and then track students through their classrooms. The result: Students of equal ability in second grade had greatly different test scores by fifth grade, depending on teacher quality. The most powerful predictor of how a child will fare academically, it turns out, is not race, not poverty, but the effectiveness of individual teachers. The "value added" technique was used in Dallas schools to rate teacher effectiveness, and the results were the same: a 35-point test gap difference depending on the effectiveness of teachers. "What surprised us the most was the size of the effect," Dallas researcher Richard Mendro told The Education Trust.

This should be good news, right? Finally, school districts know where to put their resources: teacher quality. Fat chance. First, the teacher pipeline out of education colleges is brimming with bottom-of-the-barrel recruits, as Massachusetts just discovered when 59 percent of its prospective teachers failed its basic literacy tests. For too many years universities have treated education departments as cash cows, collecting the tuition of education majors and shifting the money to the more cash-demanding engineering and sciences departments. Second, teacher shortages in rural and inner city schools leave them desperate for a teacher. Any teacher.

Finally, there's a conflict between improving teacher quality and lowering class sizes, an education initiative with unlimited popularity but very limited proof-in-research. Everyone from Republican governors to a liberal President wants to throw money at reducing classroom size. Clinton wants to spend $12 billion over seven years to hire 100,000 teachers to reduce class sizes. And the research behind all this enthusiasm? Only a single experiment rises above the bar of scientific respectability, says Harvard's Christopher Jencks. That study is Tennessee's Project STAR, conducted in 1984. Project STAR found smaller class sizes a promising reform, especially for minority students. But the experiment seemed to raise as many questions as it answered, and not all researchers agree Project STAR proved the value of reducing class sizes. "It's a very expensive solution that is unlikely to have much, if any, effect," predicts University of Rochester researcher Eric Hanushek.

With education spending mostly a zero-sum funding game, we're left with a choice: smaller class sizes or higher quality teachers. So far classroom size is winning out over teacher quality. Of course, that draws even more teachers into the profession, and not always the best teachers, as California is discovering. When the Los Angeles Times recently looked at the impact of Gov. Pete Wilson's campaign to reduce class sizes in kindergarten through third grade, they found schools were hiring uncredentialed teachers on an "emergency" basis at three times the rate before the class reduction effort started. California now has 31,000 teachers who lack certificates--teachers who are learning on the job. Many of these newcomers soon drop out; in Los Angeles, the attrition rate of unlicensed teachers runs as high as 70 percent, the Times reported.