The Wall Street Journal, August 21, 2008, Thursday

The Wall Street Journal

August 21, 2008, Thursday

The Wall Street Jouranl

When Schools Offer Money As a Motivator
More Districts Use Incentives To Reward Top Test Scores; So Far, Results Are Mixed

By JEREMY SINGER-VINE

More and more school districts are banking on improving student performance using cash incentives -- a $1,000 payout for high test scores, for example. But whether they work is hard to say.

In the latest study of student-incentive programs, researchers examining a 12-year-old program in Texas found that rewarding pupils for achieving high scores on tough tests can work. A handful of earlier studies of programs in Ohio, Israel and Canada have had mixed conclusions; results of a New York City initiative are expected in October. Comparing results is further complicated by the fact that districts across the country have implemented the programs differently.

Still, school administrators and philanthropists have pushed to launch pay-for-performance programs at hundreds of schools in the past two years. Advocates say incentives are an effective way to motivate learning -- especially among poor and minority students -- and reward teaching skills. Critics argue that the programs don't fix underlying problems, such as crowded classrooms or subpar schools.

In Texas, high-school students enrolled in Advanced Placement classes who got top scores on math, science and English tests were paid up to $500. (AP classes are considered more difficult than traditional high school curricula, and some colleges award credit for AP coursework.) The research, by C. Kirabo Jackson, an economics professor at Cornell University, found that over time, more students took Advanced Placement courses and tests, and that more graduating seniors attended college. Most of the gains came from minority students in the 40 high schools studied, accounting for about 70,000 students in all. The study, set for release on Thursday, will appear in the fall issue of Education Next, a journal published by Stanford University's Hoover Institution.

ACADEMIC AWARDS

Ways schools are rewarding students:
• Schools in Texas and six other states reward high scores on AP tests with up to $500. New York City schools award up to $1,000.
• Students in Baltimore, New York City and Coshocton, Ohio, get cash for scoring well on standardized exams.
• Perfect attendance comes with cash for some students in Chelsea, Mass., and Tucson, Ariz.

"There's a lot of buzz about pay-for-performance, but we still only have a small amount of studies on these programs, and a lot of them don't come from the U.S.," says Jonah Rockoff, a professor of finance and economics at Columbia University who is familiar with Mr. Jackson's research. "I think he's done a very careful job of doing the evaluation," he adds, noting that the study was not a true randomized experiment.

Previous data collected by the nonprofit Advanced Placement Strategies Inc., which runs the Texas program, found that in the 10 schools where it was initially launched, passing AP test scores doubled in the first year, quadrupled in the second year and have continued to increase. The program is now used in 61 schools statewide.

But exactly how much the cash incentives contributed to the improvements remains unclear. Teachers in these districts received additional training and bonuses of up to $10,000 when their students scored well. So it's inconclusive whether paying the students, rewarding the teachers or a combination of these led to the improved test scores.

In New York City, 31 high schools with large populations of poor and minority students last school year offered rewards of up to $1,000 for passing AP tests. In some subjects, such as chemistry, the number of passing scores leapt by as much as 82%. But overall, the number of students who passed AP tests slightly decreased from year ago. On Wednesday, nearly $1 million in private funds was awarded to 1,161 students. Starting this fall, additional teacher training will be offered.

"If we are going to invest, why don't we invest in something that we know does work, like reducing class size or extended learning time?" asks Pedro Noguera, a New York University sociology professor, who is critical of cash-incentive programs. Many students have trouble learning because they "are just not going to good schools, and no incentive is going to fix that," he says.

This school year, six states -- Arkansas, Alabama, Connecticut, Kentucky, Massachusetts and Virginia -- will begin replicating the Texas program, each with five-year grants of $13 million from the National Math and Science Initiative, a nonprofit organization launched last year with funding from Exxon Mobil Corp. and other private sources. About a dozen schools in each state are participating this school year, with plans to add more in following years.

The cash helped persuade Christopher Means, a senior at Marion County High School in Lebanon, Ky., to take his first AP classes. "It's definitely piqued more interest [in AP classes] than it has in the past," he says.

Overall, 60% more students will take AP courses at the participating schools in the six states this year over last year, says Tom Luce, chief executive of the initiative and a former official with the Department of Education. The organization plans to expand to 20 states within five years.

The initiative hopes eventually to expand nationwide. "I'm not Pollyannish -- it is not going to happen overnight, but it's certainly our goal," Mr. Luce says.

Previous studies of cash-incentive student programs have shown mixed results.

"It's harder than we thought it was going to be," says Joshua Angrist, a Massachusetts Institute of Technology economics professor who has co-authored two studies on cash-incentive programs, one at Israeli high schools and another at a Canadian university. He found that females respond better to cash incentives than do males. Researchers in Texas and Ohio found no significant gender difference in scores.

Stanford professor Eric Bettinger, one of Dr. Angrist's former students, is studying elementary schools at Coshocton, Ohio, where students are offered up to $20 for high test scores. Students there did significantly better on standardized math tests, but there was no effect on science, reading or social-science tests.

Many researchers and policymakers are looking to Roland G. Fryer, an economics professor at Harvard and "chief equality officer" of the New York City public schools. He oversees a privately funded program in New York City separate from the AP rewards program. In the Fryer initiative, about 10,000 elementary and middle school students earn cash and prepaid cell phones for high state test scores and good grades. He recently launched a study of the program and expects the initial results to be complete by October.

One question is whether gains attributed to cash incentives will continue if students no longer are offered rewards.

"You pay a price in motivation," says Barry Schwartz, a cognitive psychology professor at Swarthmore College. Cash incentives could ultimately diminish students' desire to learn for non-financial reasons, he says.

Dr. Fryer says he's just looking for anything that will improve student achievement, particularly for low-income and minority students: "If [incentives] don't work, I'm going to be the first person to call a press conference and tell everyone to stop."

Write to Jeremy Singer-Vine at jeremy.singer-vine@wsj.com