The current status of state testing programs is assessed drawing primarily on information provided by the Educational Testing Service publication, "State Testing Programs, 1973 Revision." Increases in state operated programs are indicated and are probably due to an increase in federal money for testing purposes. Because of possible confusion over the differences between a state testing program, a state assessment program, and a state testing service, some explanation is given as to the properties of each. A history of state testing programs is outlined, and new directions for such programs are proposed. Criterion-referenced and norm-referenced testing is contrasted, and the advantages and limitations of criterion-referenced tests are indicated. The problem of evaluating affective educational outcomes is explored and may be explained by the very limited role of noncognitive tests in state testing programs. The relation between the purposes of testing and the time of year the tests are given is discussed, and this timing is seen to affect the extent to which a particular purpose is served well or poorly. As to the type of test that should be given, standardized tests and tailor-made tests are compared, and their advantages and limitations are discussed. (RC)