One major reason for the problems of test validation is an overemphasis on the need for empirical validity data, and a failure to recognize the primary importance of explicit verbal definitions of what the test is intended to measure and rational arguments in support of the means chosen for obtaining the measurement. (Author/LC)