The Association of State Assessment Programs, an informal association of state assessment directors, began collecting information about large-scale assessment programs in 1977. This report is a continuation of that effort that is currently conducted by the North Central Regional Educational Laboratory and the Council of Chief State School Officers. The annual survey asked state test directors to comment on assessment programs, including nontraditional assessments and Title I assessment and evaluation. The results of the survey completed by the 50 states are presented. Statewide assessment programs are found in 45 states, and 2 others have temporarily suspended their assessment systems as they design new ones. Thirty-two states have at least two components to their programs. An approximately equal number of states report the use of multiple-choice and nonmultiple-choice assessment types, and an approximately equal number of states use norm-referenced and criterion-referenced tests and writing samples. Performance testing is used more often than constructed, open-response testing, and portfolio assessment is used in only a few states. Most states use their assessment results for two to four purposes, with improving instruction, school accountability, and program evaluation the most common. The tensions that exist when assessment is used for accountability and instructional improvement cause difficulty for those who design and implement these programs, and these tensions are exacerbated by placing negative consequences on poor performance, thus increasing the stakes for schools and students. An appendix presents a survey summary table. (Contains 26 charts, 3 tables, 4 figures, and 1 appendix table.) (SLD)