This paper examines the issues arising when a set of federally funded model demonstration projects, the secondary curricula/student transition projects of the Office of Special Education and Rehabilitative Services, were obligated by federal initiative to improve the amount, quality, and use of evaluation. These issues included: (1) the nature of federal expectations and needs for evaluation data, including emphasis on outcomes and quantitative data in evaluation, the overpromising of program goals at project initiation, and funding cycles; and (2) the extent of local capabilities to meet those expectations and provide evaluative data, including problems of lack of evaluative expertise and objectivity at the local level and the uncertainty of rewards associated with project response to federal evaluative needs. Regulatory and methodological implications of the discrepancy between federal expectations and local capabilities along with implications for evaluation technical assistance are discussed. (13 references) (JDD)