system. We therefore used as a yardstick a simple measure: whether a state's decision would help improve student learning and reduce the achievement gap. Simple compliance with the law was not enough.
Second, the committee agreed that the education improvement system, of which assessment and accountability are key components, should be conceived of and implemented as just that—a system. That is, the system should consist of a number of components, at various levels (classroom, school, school district, and state), each of which plays a role in measuring and contributing to student learning, yet which are interrelated, not separate from one another. Applying this principle, one might view one component—for example, a state testing program—that by itself falls short, as appropriate in a system that also includes components that complement it.
Third, the committee agreed that a hallmark of the state and district systems should be continuous improvement at all levels—for students, for teachers and administrators, and for the system itself. Improvement for students is obvious; as noted above, it is the reason for the system. Improvement for teachers and administrators is also a necessary part of the system. Although the role of professional development was not formally part of the committee's charge, the committee found it impossible to discuss assessment without addressing the role of enhancing the knowledge and skills of teachers. High student performance depends on high-quality instruction, and building the capacity of teachers and administrators is at least as much a necessary condition of educational improvement as establishing standards or putting in place accountability mechanisms.
Improvement for the system itself is also essential. Assessments and accountability schemes change constantly, despite the best-laid plans of educators and blue-ribbon panels. Legislators, responding to teachers, parents, and other constituents, frequently mandate adjustments, from adding multiple-choice components to requiring individual student reports. And administrators change programs as data come in that show the effects of the system on students and schools. Although the Title I statute calls the assessments that are to be put in place by 2000–2001 “final,” these assessments are likely to undergo numerous rounds of revision, even if the structure of the education improvement system remains in place.
Moreover, as noted above, the effects of the components of a standards-based system are not completely certain. Many of the systems the committee studied are relatively new or still in some cases under development, and their full impact on students, particularly disadvantaged students, remains to be seen. Therefore, states and districts need to monitor their improvement systems continually, to ensure that they are achieving their desired goals.
For these and other reasons, the committee developed criteria not for the one best system—which does not exist—but for systems that are continually changing and adapting to new knowledge and new circumstances. In this way, the committee's framework is appropriate for states and districts just starting out