In this report, we have compared the performance of California students with the performance of students from states with populations as diverse as California's. While the data represent a snapshot rather than a long-term view, they provide important insights into how well public education is doing.

The data suggest a very divided student population in California. The higher-performing students-who often come from more advantaged neighborhoods-do about as well as similar students in comparison states. Low-performing students, on the other hand, fare considerably worse than low-performing students in comparison states.

This finding clearly demonstrates the need to focus particular effort on improving the achievement of low-performing students. The failure to adequately address the needs of these students not only carries serious consequences for the individuals, but for society as a whole. Education has long been considered the avenue to greater opportunity. The achievement disparities noted here raise the specter of a two-tiered society, where public education no longer provides to many the tools needed for a better life.

The achievement of California's students is also an important factor in the health of the state's economy. If businesses perceive that the typical student in California is less proficient than in other states or countries, companies may look elsewhere to locate or expand.

While there is no one strategy to improve educational performance, policymakers can contribute to the long-term success of the educational system. Improving the achievement of low-performing

students should be a high priority. Promoting local implementation of school-to-work programs is one avenue to improve academic and career opportunities for students who typically do not continue on to college after high school. In addition, policymakers can review whether programs targeted at low-performing students are sufficiently flexible, targeted, and funded to give local educators the tools needed to successfully address the needs of these students.

More generally, however, policymakers should be looking broadly at K-12 education to improve educational services for all students. The lesson of previous education reform is that improvements in education occur in the classroom. The state or federal government can encourage, but not accomplish, this goal. Therefore, policymakers should concentrate on supporting reform at the local level rather than mandating a new set of programs. A revision of the state's complex categorical program structure is one step that could encourage more successful local programs.

Improving the quality of educational data and training teachers and administrators to use that data to improve local programs is another important step. Outcome data are essential to understanding the success of local programs and of state efforts to improve schools. As we discussed in this report, very little comparable data are available and what exist are difficult to interpret. Educational data must improve to allow a deeper understanding of student performance and the factors that contribute to those outcomes.