Investing in Education
Are we putting our money where our myths are?
Is it money that matters? When it comes to improving student performance, maybe not, according to a recent study by an economist at the University of Rochester.
“There is no strong or systematic relationship between school expenditures and student performance,” writes Eric Hanushek in his report, The Impact of Differential Expenditures on School Performance. “School reform discussions that begin with the premise that constraints on expenditures are the most serious roadblock to improved student performance are, at best, misguided.”
Hanushek reviewed 187 previous studies that had attempted to show a correlation between school expenditures and student performance. After all, it would seem that spending more money on education—increasing teachers' salaries, for example, or improving school facilities—would lead to higher achievement levels among students. But that isn't necessarily the case, Hanushek's study concludes.
Among his major findings:
- Higher per-pupil expenditures don't guarantee an improvement in student performance—at least as measured by standardized tests (70 percent of the studies Hanushek looked at used such tests for their data). Spending more money on students made no difference at all in three-fourths of the studies, while only 20 percent noted a change for the better. In addition, about 5 percent found a negative impact.
- Smaller class sizes don't automatically lead to better student performance. Of the 152 studies that looked at the effects of smaller classes on student achievement, 82 percent found no measurable impact, 9.2 percent saw positive results, and 8.6 percent found negative results.
- There is little correlation between higher teacher pay and improved student performance. Of the 69 studies that dealt with teacher salary, 78 percent said higher pay made no difference in the classroom. Sixteen percent said performance improved along with increases in teacher pay, but nearly 6 percent said indicators of student progress actually went down when the pay went up.
- Increasing administrative expenditures also apparently has little effect on academic performance. Out of 61 studies that dealt with that subject, 86.9 percent found no difference, although 11.5 percent noted an improvement.
About the only significant impact Hanushek found was in the 140 studies that asked whether teacher experience affects student performance. Of those, 28.5 percent said yes, while 64 percent could find no difference and 7 percent found a negative impact. “But these results are hardly overwhelming,” Hanushek writes.
So, money doesn't matter? “That's much too strong a conclusion to draw from this review because there are lots of questions these studies don't answer,” says Albert Shanker, president of the American Federation of Teachers and a leading proponent of school reform. In his regularly published column, Shanker asked, “What does per-pupil expenditure tell us? Not how the money was spent. How many high-spending districts spent lots of money on administration and on activities that don't connect directly to teachers and kids? If that's what the money was used for, then per-pupil expenditure might well have no impact on student performance. Did the studies break out districts or schools this way? We don't know.”
Shanker also criticizes the means by which “student performance” was measured in most of the studies: standardized tests. “Those certainly tell us something basic about performance,” he wrote, “but they don't tell us whether students can write a good essay or have an intelligent discussion or create something.”
Hanushek himself admits that “there are several obvious reasons for caution” in interpreting the evidence from the 187 studies he examined. “For any individual study, incomplete information, poor quality data, or faulty research could distort statistical results,” he says. But because a majority of the studies point in the same direction, Hanushek believes his findings are basically sound.
Hanushek admits that money could make a difference—under the right circumstances. “For example,” he says, “almost every economist would support the argument that increasing teacher salaries would expand and improve the pool of potential teachers. Whether or not this would improve the quality of teaching, however, would depend on whether.schools systematically choose and retain the best teachers from the pool.”
Hanushek, who has conducted similar research on school expenditures and student performance, says that people frequently misinterpret his studies. “One interpretation is that money doesn't matter at all, which isn't what I said. What people usually miss is that deep down, the theme of the work is that if you're interested in improving student performance, you have to develop management techniques and policies that are geared directly toward performance.”
Vol. 01, Issue 01, Page 36