As part of an American Educational Research Association Symposium reporting the findings of the National Academy of Sciences (NAS) panel on placing children in special education, this paper discusses the implications of the panel's conclusions for exemplary assessment practices in special education and, indeed, for the very role of assessment in special education. When confronted with the central problem of over-representation of minority children and males in programs for educable mentally retarded students, the NAS panel responded not with solutions focussing on the direct reduction of disproportion per se—which tend to be simplistic in merely transferring the problem to other parts of the educational system—but rather with solutions focussing on deeper educational issues underlying disproportionate placement, such as the validity of assessment of student functional needs and the quality of services that should be provided. After first analyzing the three primary issues underlying the policy conflict between equal protection from the harm of separate placement and entitlement to needed services—namely, the quality of the regular instruction to which referred students were exposed, the validity of assessment of student functional needs, and the quality of the special education provided—the panel recommended two major components for special education assessment: (1) for entrance or placement in special education, a two-phase comprehensive assessment of student performance in relation to the quality of regular instruction received; (2) for exit from or retention in special education, the assessment of special education students' progress toward relevant academic goals, or the lack thereof, in relation to the quality of special instruction received. Thus, student educational progress emerges as the touchstone for equity in special education. (Author). (24pp.