In response to growing frustration over the lack of information about the national effectiveness of the Chapter 1 program, Congress enacted the Education Amendments of 1974. Section 151 of the Amendments directed the U.S. Office of Education to develop evaluation models that would allow school district data to be aggregated to provide national estimates of program effectiveness. The norm-referenced model was the most easily applied of the alternatives developed. This model substitutes test norms for a traditional comparison group. Posttest standing relative to the norm group is compared with pretest standing relative to the norm group. The 1975 document, "A Practical Guide to Measuring Project Impact on Student Achievement," specified the conditions in which the norm-referenced model could be used. Several difficulties have arisen in implementing the models, but school districts today are still required to evaluate their Chapter 1 projects. Requirements enacted in 1988 mean that districts essentially must use nationally normed tests or tests equated to nationally normed tests to measure student achievement in both basic and more advanced skills. Test norms and norm-referenced tests are reviewed, with attention to measurement error, the effects of high-stakes testing, and the relevance of national norms as a comparison group. Ways in which information on program effectiveness could be better provided are discussed. Five figures and one table illustrate the discussion. (SLD)