Link to Original Story

Bill Gates talks about teacher pay, class size

By Daniel de Vise

This afternoon I am turning College Inc. over to K-12 education to report on a morning interview with Bill Gates, chairman of Microsoft and, through the Bill & Melinda Gates Foundation, probably the most important philanthropist in education.

Gates, 55, famously dropped out of Harvard in 1975 to launch the company initially known as Micro-Soft, writing code that made the personal computer accessible to hobbyists and regular folks. He is a product of an upper-middle-class family and an exclusive Seattle preparatory school, whose women's club used proceeds from a rummage sale to purchase computer time for Bill and his friends.

Lately, Gates has been advocating paying teachers based on classroom performance instead of seniority and ending costly investments in class-size reduction, two of the more provocative topics in public education. We spoke in his foundation's Washington office shortly before a speech to the National Governors Association.

Gates's central point is that the nation is spending more than ever on public education and not getting better results in return. Per-student spending has more than doubled in the past four decades, yet math and reading scores haven't significantly improved. Performance is essentially flat on the National Assessment of Educational Progress, the only national test of reading and math with data that goes back to the 1970s.

Many education leaders would say that Gates's criticism is unfounded. While NAEP scores are flat, scores on many state tests have risen over the past decade, to great fanfare. Test scores in both Maryland and Virginia have risen substantially in that span.

Gates contends that those gains are probably largely a result of new-test phenomenon: Test scores almost always rise under a new test, as students and teachers familiarize themselves with the test and the material it measures.

"Whenever you have a new test, people learn what's in that test over the first three or four years," he said. "The fact that doesn't show up on NAEP at all is a bit damning."

Most of the new money pumped into public education has been spent on new teachers. For every 1,000 students, public schools employ 125 adults, an 8-to-1 ratio, comparable to the student-faculty ratio at Swarthmore College.

The number of instructors per 1,000 students has more than doubled since 1960, from 40 to 85.

Gates contends that the K-12 education industry has been steered for five decades by a misguided belief that the way to higher performance is smaller classes. Many states pursued class-size reduction initiatives in the 1990s. California, an example I covered as a reporter, reduced average class size from 30 to 20 in kindergarten through grade 3 in the mid-1990s, at a cost of over $1.5 billion a year.

Over the years, though, the research community has more or less confirmed that class-size reduction doesn't yield significant performance gains. The most expensive education reform is among the least effective.

Gates proposes ending class-size reduction experiments, lifting caps on class size and offering good teachers financial incentives to teach more students.

"If you look at something like class sizes going from 22 to 27, and paying that teacher a third of the savings, and you make sure it's the effective teachers you're retaining," he said, "by any measure, you're raising the quality of education as you do that."

Now, as any public school parent knows, class sizes across much of the nation aren't particularly small. In fact, average class size in self-contained classrooms has barely changed over the past decade, from 20.9 students in the 2000 academic year to 20.3 students in 2008, according to federal data.

Why doesn't more teachers translate to smaller classes? Many of the newly hired educators are not, strictly speaking, classroom teachers. They are math and reading specialists, mentor teachers and testing experts, hired primarily to raise test scores and as a direct result of No Child Left Behind, the federal law that requires public schools to demonstrate "adequate yearly progress" through reading and math scores.

One alternative to raising class sizes is to assign some of those teachers to classrooms.

"I'd bet you some of those specialized roles don't add value," he said. " ... We should have a general skepticism about those jobs."

Gates's other, equally provocative target is teacher pay. He is openly critical of the current teacher compensation system, which awards pay increases almost entirely according to seniority. Teachers are evaluated -- but "99 percent get the high rating," Gates said. Few teachers earn poor evaluations, and fewer still are rewarded or denied money according to those evaluations.

Gates advocates awarding pay raises according to effectiveness, not seniority. He said he is "aghast" that the present system offers essentially "no encouragement to improve." In our interview, he termed the current compensation model "a crazy thing."

In an op-ed piece published today in The Washington Post, Gates reasons that "we have to identify great teachers, find out what makes them so effective and transfer those skills to others".

The No Child law has spawned enough metrics to measure the performance of every teacher in the nation on the math and reading gains of students. But Gates contends that using test scores alone is too narrow.

"The test score is not as good as it should be in telling the teacher how to improve," he said in the interview.