Does The Extra Spending Do Any Good

April 09, 1995|By M. William Salganik | M. William Salganik,Sun Staff Writer Sun staff writer Anne Haddad contributed to this article.

It's school budget season. Teacher unions and school administrators, PTAs and school boards, county councils and county executives are haggling over how much money is needed to "maintain quality education."

Meanwhile, in a Baltimore courtroom, there are arguments over whether the state is providing enough aid for Baltimore's public school system to offer an adequate education.

There is debate about class size and teacher salaries and overall spending.

In Carroll County, for example, school officials are seeking $301,352 to reduce average class size in middle schools from 27 to 26.5. County Commissioner Benjamin Brown, at a recent budget hearing, suggested 28 might be good enough. Superintendent Brian Lockard replied quickly, "We'd like it to be 25. Sorry."

But education research questions whether any of it matters, whether spending -- or the major factors which contribute to it, such as class size -- really improves achievement.

"The nation is spending more and more to achieve results that are no better, and perhaps worse," wrote Eric A. Hanushek, a University of Rochester economist, in a report last fall published by the Brookings Institution.

In his research over a number of years, Dr. Hanushek has reviewed hundreds of studies measuring the effect on achievement of factors including overall per-pupil spending, teacher experience, teacher education, teacher salary, facilities and class size and pupil-teacher ratio. None of these factors, he concluded, shows a significant, consistent, positive effect on student achievement.

If money doesn't matter, why do people fight so hard over school budgets?

"It's now taken as conventional wisdom [by researchers] that money is not related to student performance," Dr. Hanushek said in a recent interview. "This hasn't filtered down to school boards because people don't want to hear that message."

While he says his conclusions are "conventional wisdom," Dr. Hanushek hardly stands unchallenged. Larry V. Hedges, a statistician who is a professor of education at the University of Chicago, and two co-authors challenged Dr. Hanushek last year in an exchange published in Educational Researcher, the journal of the American Educational Research Association.

The debate quickly hones in on fine points of statistical methods, and the articles are full of statements such as this one, by Dr. Hedges and his co-authors: "Because output variables were not measured on the same scale in all studies, the partial regression coefficients for the resource input variables could not be combined directly. Consequently, all of our indices of relations predicted standardized output, that is, output in standard deviation units."

Beyond the technical talk, here is what the debate boils down to:

* In addition to his summing-up of the hundreds of studies, Dr. Hanushek points out that education spending has increased consistently in real, inflation-adjusted 1990 dollars: from $164 per pupil in 1890 to $772 in 1940 to $4,622 in 1990. The growth in spending, he continues, cannot be accounted for by rising teacher salaries (teacher salaries are declining relative to those of other college graduates) or by rising special education enrollment (a factor in the cost increase, but a relatively small one).

While spending has been increasing, he continues, the best measure of achievement, the National Assessment of Educational Progress, has shown little gain since the program began in the early 1970s.

* Dr. Hedges and his colleagues question the statistical methods used by Dr. Hanushek. Using different techniques, they reviewed the same studies he did and concluded that an increase in overall per-pupil spending did lead to improved student achievement. However, they concede, many of the variables that contribute to spending, such as class size, show little connection to improved student work.

As for the lack of achievement gains nationally while spending has increased, Dr. Hedges said in an interview that the problem is that other factors are not constant. In particular, he said, an increasing proportion of the population is staying in school into high school, and "the contributions that families are able to make have changed," with such trends such as more single-parent homes.

In some ways, however, the antagonists are staking out positions that are not so far apart.

While class size "in the aggregate" makes no difference, Dr. Hanushek said, "there are clearly some classes and some subjects and some teachers for whom smaller classes can be productive."

However, he said, there has been little research into whether, say, class size is more important in math or reading, elementary school or high school, with gifted students or in remedial classes. "Nobody has incentive to figure out where it makes sense and where it doesn't," he said. And "what parents don't figure out" when supporting smaller classes at budget hearings is that "it really is a trade-off -- if we didn't have smaller classes, maybe we could do something else."

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.