More than three years ago, schools in the small central Ohio city of Coshocton launched an experiment to pay elementary students for passing or scoring high on state exams. The results from that experiment suggest that in some respects, little Coshocton’s big gamble paid off.
The findings, unveiled at a school board meeting this month, show that the prospect of getting as much as $100 at the end of the year was enough to motivate 3rd through 6th graders to improve their mathematics scores on state exams. The cash incentives had no impact on reading scores, though, and led to small—but not significant—gains in science and social studies.
On the other hand, the motivational boost in math may not have had much staying power: When students were eligible to vie for cash awards one year but not the next, the learning gains were smaller after the dollars disappeared.
To the benefactor of the experiment, nonetheless, the results sounded encouraging. “I’m hoping this final year that the results will be even more meaningful,” said Robert E. Simpson, the local manufacturer whose family foundation is footing the $30,000- to $50,000-a-year bill for the program. “It’s a good thing for this district.”
Counter to Research?
Outside of Coshocton, though, the results were not strong enough, or widespread enough, to persuade critics to jump on the money-for-achievement bandwagon.
In an era when the federal No Child Left Behind Act is putting unprecedented pressure on schools to take bold steps to improve test scores, a growing number of districts, including Baltimore, New York City, and Fulton County, Ga., are turning to cash incentives to motivate students to do their best. (“Ohio District Tests Performance Pay—for Students,” Jan. 17, 2007, and “Promises of Money Meant to Heighten Student Motivation,” Feb. 13, 2008.)
Critics continue to note, however, that the trend runs counter to decades of research suggesting that such practices, in the long run, could squelch students’ inner drive to learn for the sheer enjoyment of it.
“I don’t think the results beg for continuation, given the potential risk to students’ motivation,” said Ronald E. Chennault, an associate professor of education policy studies at DePaul University in Chicago. “Will students have to continue to be paid forever?”
Picked by Lottery
The experiment in the 1,800-student Coshocton district began at the start of the 2004-05 school year. That September and each year since, educators bused 3rd through 6th graders to the high school for a lottery to determine which grades in which schools would be eligible to take part in the rewards program.
The high school cheerleaders and the marching band were recruited to perform. And pupils whose grades were selected were told that they could earn $15 for every “proficient” score and $20 for each “accelerated” or “advanced” score they earned on state exams.
Students who ace all their exams could earn up to $100. Students who qualify for rewards get them in June in the form of “Coshocton Kid Bucks,” colorful coupons that can be redeemed at stores in the economically depressed city—but only by children.
According to Eric P. Bettinger, the researcher from Case Western Reserve University in Cleveland who is analyzing the results, the “robust” test-score gains that students made in math as a result of the program compare in size with those generated in the Project STAR (Student/Teacher Achievement Ratio) study, a well-known experiment in Tennessee during the 1990s to boost learning in the early grades by shrinking class sizes.
The effect of the cash incentives also held up, he added, regardless of students’ socioeconomic differences, racial or ethnic groups, or gender.
“Where we really saw the effect was happening,” Mr. Bettinger said, “was with kids who were already going to pass and were moving up to a higher level.”
Also, while the test-score improvements grew over time for students whose grades were selected more than once, the bump up was much smaller in the second year.
“At this point, I can’t say if there is a cumulative effect, and I can’t say there isn’t, either,” said Mr. Bettinger, an associate professor of economics.
In reading, he added, the lack of an impact could reflect a quirk in the district’s accountability calculations for 3rd grade. Third graders take reading tests twice a year, but only the higher score “counts.”
“Kids who scored high in the fall realized, lo and behold, ‘I’ve already earned as much money as I can, so I’m going to expend my efforts elsewhere,’ ” he said. “These 3rd graders actually responded to the incentives in a pretty sophisticated way.”
To remove that temptation, the district this year—its fourth in the program—will consider only spring test scores in reading.
What the findings don’t definitively show is whether the good effects were due to students’ working harder or teachers’ changing their instructional practices. To study that issue further, the school board agreed to expand the program to 7th and 8th graders this year and randomly assign each individual student to either the experimental group or a comparison group so that teachers will not necessarily know which students are participating.
The program will also focus only on math and reading this year, with student award levels rising to $20 to $25 per subject.
“We don’t know the long-term impact of this yet—whether it’s a quick fix or something that’s going to help students in the long run,” said David L. Hire, Coshocton’s new superintendent. “We’ll just see how it continues to pan out over the next year.”
A version of this article appeared in the February 27, 2008 edition of Education Week as Students in Cash-Incentives Study Score Higher in Math