Now we turn to the issue of measuring school effects. In particular, we will consider the work of Anita Summers and Amy Johnson on school-based management (chapter 5, this volume) and Eric Hanushek on a broader set of school reforms. The principal conclusion of Summers and Johnson's work is that the invention or implementation of any new management system or organizational structure in schools does not of itself necessarily lead to any improvement in education outcomes. Without organization around content-defined objectives for student performance, neither improved student achievement nor efficiency will follow willy-nilly from a governance change. In accord with the findings of the Panel on the Economics of Education Reform, an international study conducted through the Organization for Economic Cooperation and Development (OECD) reports that there are no clearly demonstrated links between increased school autonomy and student learning (OECD, 1994). School-based management may improve certain complementary aspects of schooling, not the least of which may be teacher morale and enthusiasm for the reform effort; however, little empirical evidence can be mustered to support the assertion that greater stakeholder participation directly improves student performance.
The OECD report goes on to suggest, however, that to the degree that a reorganization effort is conducted with a clarity of purpose to improve classroom teaching and learning, positive outcomes may accrue. In other words, to improve student learning, the content and instruction delivered to students must change as well as the organizational structure of the school. They complement each other.
This is not rocket science. Included in the administration's reform agenda discussed earlier are features intended to facilitate school responsibility in resource management by pushing resources down to schools, by granting waivers, and by other strategies. At the same time, incentives are provided so that schools will focus on challenging content standards and aligned performance assessments that together raise expectations for student learning, hopefully leading to increased efficiency and improved student achievement.
Eric Hanushek, too, finds no positive relationship between a variety of measures and school performance.5 How does one reconcile this with the eight different sets of findings we described earlier that, taken collectively, suggest we know a lot about how to change the opportunities for children to achieve to higher levels. How do we balance these different perspectives?
A close look at the survey/production function studies Hanushek examines shows that those studies by and large neglected to consider any of these eight research findings. There is no measure of whether the teachers and schools in the surveys were focusing their instruction on bringing all students to achieve to high