State education agencies are in increasing agreement regarding the basic principles that should underlie state educational assessment programs, though some areas of divergent opinion remain. The agencies generally accept the ideas that comparisons between states will be made, that assessment programs should serve multiple purposes, that meaningful comparisons cannot be made without knowledge of the contextual factors affecting the provision of education, that a large number of indicators must be employed to fully display an educational system's status, and that assessment programs are bound to be costly. The agencies disagree over the uses to which assessments should be put, the educational outcomes that should be measured, the extent to which various contextual factors affect education, the specific indicators that should be used, and how the costs of assessment are to be controlled and allocated. Agencies need to develop policies for assessment that will resolve all of these questions in ways that are appropriate both to the individual characteristics of the state and to the demand for data that will allow meaningful assessment and comparison on a national level. (PGD)