Once Critical, Researcher Says Spending, Performance Not Tied
WASHINGTON--In 1966, the sociologist James S. Coleman and his colleagues shook the education world with their landmark study, "Equality of Educational Opportunity.''
Much-criticized and often misinterpreted, the study nonetheless attracted enormous attention, and helped steer education policy, with its conclusion that family backgrounds, not schools, are the primary determinant of student performance.
Today, a quarter-century later, one of the critics of the Coleman study now says that, despite flaws, the earlier report was partly right.
Analyzing 187 studies that have been conducted since 1966, Eric A. Hanushek, a professor of economics and political science at the University of Rochester, found what he says is an unmistakable conclusion: There is no "strong or systematic'' relationship between school expenditures and student performance.
The Coleman report "was right in one sense and wrong in another,'' Mr. Hanushek says.
"The interpretation that comes out that says schools don't matter is entirely wrong,'' he says. "What is the case is that all measurable attributes of schools don't matter. That's confirmed.''
Mr. Hanushek adds that his findings have clear implications for two popular, but controversial, policy remedies.
School boards should stop seeking reductions in class sizes, he says, because such efforts would not have much effect on performance.
And, he suggests, school-finance lawsuits aimed at equalizing spending among school districts are equally fruitless.
Not surprisingly, Mr. Hanushek's research has attracted sharp criticism, much as Mr. Coleman's did.
Jonathan Kozol, the author of Savage Inequalities, a 1991 book that documented the squalid conditions in many inner-city schools, scoffs at the idea that money makes little difference in schooling outcomes.
"If money is good for the son of Prescott Bush, it is good for the child of the poorest woman in Chicago,'' Mr. Kozol says. "Why doesn't [Mr. Hanushek] tell President Bush his father wasted money sending him to Andover?''
"Anybody who has spent a lot of time in inner-city, real schools,'' he continues, "knows very well that money is not everything that is needed, but it is a precondition for everything else.''
Weak and Ambiguous
Mr. Hanushek's analysis, first published in 1989, has attracted a good deal of attention. In June, for example, he spoke at a seminar here sponsored by the U.S. Education Department's office of educational research and improvement.
Diane S. Ravitch, the head of the O.E.R.I., maintains that the department was not conferring its blessing on his conclusions.
"We are interested in hearing people who have interesting things to say,'' she says. "We don't have a censorship process. We choose people who are politically correct, and people who are politically incorrect.''
In his study, Mr. Hanushek examines distinct studies, from 38 separately published articles and books, that related school expenditures to student performance. The studies include all regions of the country; half examine elementary grades (grades 1-6), and half look at secondary grades.
More than two-thirds of the studies use standardized-test scores as a measure of student performance; the rest use dropout rates, college enrollment, and performance after school, among other factors.
The vast majority--152--of the studies look at teacher-pupil ratio as a measure of instructional expenditure; other factors include teacher experience (140 studies), teacher education (113), teacher salary (69), facilities (74), expenditures per pupil (65), and administrative inputs (61).
If school spending makes a difference in performance, Mr. Hanushek argues, the expenditure factors should have a positive effect on student achievement. In fact, though, the relationships are weak and ambiguous.
Of the 152 estimates of the effect of class size, for example, only 27 are statistically significant, and, of these, half show a negative relation between small classes and student performance. Likewise, the studies that look at teacher education also provide little evidence of a positive relation.
The only possible exceptions are the studies that examine teacher experience. Some 40 of the 140 studies show a positive relation between years of experience and student performance (10 show a negative relation). But Mr. Hanushek points out that the causation might run the other way: Experienced teachers may gravitate to schools with high-performing students.
Counter to Common Sense
Mr. Hanushek concedes that his conclusion appears to run counter to common sense.
"How could it be I claim no relationship between expenditures and performance when we see it all the time?'' he asks. "Kids perform well in Montgomery County, Md., and poorly in low-spending communities. Isn't there an obvious relation?''
"What I am trying to give,'' he continues, "is a measure of how expenditures add to performance of kids over and above families and other sources of learning. It's a value-added notion of schools. It's the right answer if you want to think of policies schools can adopt.''
Mr. Hanushek also suggests that much of the spending on affluent schools--such as New Trier High School in suburban Chicago--goes to non-educational amenities.
"I would bet dollars to doughnuts they have the best violin instruction and Olympic swimming pools,'' he says. "They're buying consumption goods. They are trying to throw money at something they think is important.''
And, he acknowledges, the examples in Mr. Kozol's book are "tragic.''
But, he says, the answer is not to redistribute funds from wealthier districts to poorer ones.
"The simple answer, readjust funds, could be damaging to some schools that got their act together,'' he contends.
Mr. Kozol strongly disagrees, and argues that the inequalities in physical conditions have a powerful effect on students.
"A school that is physically repellent has an unquestioned power of obliteration on a child's sense of self-respect and motivation,'' Mr. Kozol says. "They know the difference'' between their schools and those in affluent suburbs.
"If money is not the issue,'' he argues, "why do rich districts panic at the fear of losing it?''
Differences in Teachers
Mr. Hanushek concedes that there are inequities among schools, but he says that looking at per-pupil spending is the wrong way to look at equity.
Rather, he says, the important differences among schools are the differences in the quality of teachers.
"It makes a difference what classroom you are in,'' he says. "If we didn't find that, we'd be suspicious--parents know. In fact, research confirms it.''
To address that issue, Mr. Hanushek argues, policymakers should find a way to attract better teachers.
But, he says, "there is little hope of changing schools importantly by doing such things as raising salaries and letting the system rip. Bad teachers like higher salaries just as much as good teachers.''
The best way to reward good teaching, the economist suggests, is to reward teachers after the fact--pay more to those who produce the greatest growth in student achievement.
But such a solution, he acknowledges, might be impossible politically.
"One thing we know is merit-pay systems never survive in schools,''