How can quantitative research and common sense be so at odds? A natural experiment involving 16 elementary schools serving minority, high-poverty populations in East Austin, Texas, provides an answer. In the late 1980s, student absenteeism in these schools was high and student scores on state-mandated achievement tests were extremely low--low vis-a-vis statewide averages and low vis-a-vis the rest of Austin. In 1989, as part of the resolution to a desegregation court case, the 16 elementary schools were designated “priority schools” and each school was given $300,000 per year for five years. The $300,000 payments were additions to normal school spending. The settlements were made on the theory that money mattered. If the theory was right, this money should have been enough to make a difference.
Five years later, at the end of the 1994 school year, the money had produced mixed results and an important management lesson. In 14 of the 16 schools, student attendance and student achievement had remained very low. In the two other schools, Zavala Elementary School and Ortega Elementary School, the record was very different. By the end of 1994, student attendance rates at those two schools were among the city’s highest and student test scores had risen to the city’s average. There was no “creaming” here: Zavala and Ortega had continued to draw students from the same poor neighborhoods where family incomes averaged $12,000. But somehow, the two elementary schools had managed something extraordinary.
In quantitative terms, the data from East Austin reinforce existing research: In 14 out of 16 schools, more money made no difference. But consistent with the intuition of administrators and parents, money made a big difference in two schools. The crucial policy question is why. The answer lies in how the money was spent.
Like all organizations, schools are places of vested interests and fixed routines. If new resources arrive, the path of least resistance is to spend the resources in ways that change as little as possible. In 14 schools--all but Zavala and Ortega--this is what happened. New money was spent to hire extra teachers to reduce class size. But little was done to change what happened inside the classes. In the words of Terry Bishop, Austin’s school superintendent in 1994:
"They didn't change the way they were doing things. ... All they did was take that support, lower pupil-teacher ratios, still use the same curriculum, still use the same instructional methods. You'd go over there and they'd actually have ... 10 students in a class, but guess what they would be doing? You'd have two rows of five students, and the teacher would still be sitting up there in the front of the room, and still using ditto sheets like they were before. You can do that with 30 students as easily as you can with 10. Politically, I think they were afraid to use the money in any other way."
Zavala and Ortega also used much of the extra money to lower class size. But in these two schools, smaller classes were part of a larger program. At Zavala Elementary that program began in an evening PTA meeting when Alejandro Melton, the newly appointed principal, asked a parent to read aloud the scores of Zavala’s students on the statewide tests. Before this recitation, Zavala’s teachers and parents had been locked in mutual misunderstanding. The parents had seen report cards with A’s and B’s and assumed their children were doing well. The Zavala teachers had thought Zavala parents had little interest in education and so saw no sense in raising issues that would meet with parental indifference or worse. In the short run, the evening produced intense parental anger and a few student transfers. In the longer run, it led to strong parental support to raise student achievement.
The Zavala principal and teachers adopted for all students the reading and math curriculum that the rest of the district used only with gifted and talented children, and they used the new curriculum as a lever to change instructional methods. The adoption required that the principal get a waiver from the school board, and with the support of the school’s teachers and parents, he did. Zavala put all special-needs children into regular classes--an action that smaller class sizes made possible and that freed up a considerable amount of money. Professional development helped Zavala teachers learn to teach the new curriculum effectively and to manage classes containing special-needs children. Despite initial opposition from the school board, the faculty and parents brought health services to Zavala, which turned out to be important in raising attendance in a school population in which 40 percent of the children come to school without state-mandated immunizations and the alternative to school-based health services is waiting all day in the emergency room at the nearest hospital. Zavala invested heavily in getting parents involved in their children’s schooling, including having parents participate in the governance structure of the schools, right down to sitting on hiring and budget committees.
The details at Ortega were different, but the story to this point provides the first part of the management lesson. Dollars made a difference in Zavala and Ortega because both schools specifically adopted the goal of raising student achievement and they found ways to engage teachers, parents, and students in pursuing these goals. Without this consensus, there would have been no support for the changes needed to raise student achievement.
Indeed, many changes were needed. Before a school like Zavala could report success, it had to provide teachers with the incentives and the opportunities to improve their teaching. It had to put in place a method of measuring progress. It had to make clearer to students and parents why achievement was so important. Each of these activities had setbacks, and the staff had to persevere as they learned from their mistakes.
The list of pieces--establishing clear goals, structuring incentives, providing high-quality training, developing good measures of progress--sounds daunting, and yet a moment’s reflection suggests that all the pieces are needed. It is hard to structure serious incentives for teachers and students if there is no agreement on organizational goals. It is hard to sustain effort toward reaching goals without measures of progress. But a list this long in a world of limited time and energy leads to an obvious question: Where, exactly, is a school supposed to start?
The answer, which Zavala and Ortega discovered on their own, and which political economist Albert O. Hirschman described 40 years ago, is the second part of the management lesson. Writing about strategies for economic development in the 1950s, Mr. Hirschman developed a key insight about “getting from here to there.” In those years, the prevailing wisdom was that balanced growth was the most effective economic-development strategy for a poor country. Balanced growth meant moving on all fronts at once: raising financing for new factories, building roads so the new factories could easily move raw materials and finished products, establishing schools to train labor for the new factories, and so on. The idea of balance--avoiding bottlenecks--sounded wonderful, but it required a herculean coordination effort well beyond the capability of most governments.
Mr. Hirschman’s advice was to attack the development problem quite differently. Instead of trying to make progress on all fronts simultaneously, start with the initiative that creates the most pressure for other constructive changes. For example, if the government concentrated on raising funds for new factories, the owners of the new factories would create pressures to build the most needed roads and to develop the most essential training programs.
Mr. Hirschman’s logic applies equally well to a strategy for school improvement. Start with the change that is most likely to create pressure for other needed changes. In the Zavala school, this was getting parents and teachers to agree that low student achievement was the problem.
At both Zavala and Ortega, extra money made a big difference because it was used to support significant changes in the work teachers and students did together each day. The two-part management lesson is about what it takes for money to make a difference in other schools as well.
Richard J. Murnane, an economist, is a professor at Harvard University’s graduate school of education. Frank Levy is the Daniel Rose professor of urban economics in the department of urban studies and planning at the Massachusetts Institute of Technology. This article is based on their book to be released this week, Teaching the New Basic Skills (Free Press, 1996).
