Cross-validation in relation to choosing the best tests and selecting the best items in tests is discussed. Cross-validation demonstrated whether a decision derived from one set of data is truly effective when this decision is applied to another independent, but relevant, sample of people. Cross-validation is particularly important after statistical data have been used to choose the best tests to make up a battery for use with the next group of people. A cross-validation experiment on a new group will tell how good the choice of test really is. Item analysis is a means of improving tests. The data from the analysis are used to eliminate doubtful items and to determine the best scoring weights. By applying the revised test to a new independent group, the inventory is refined. When the test is, without change, administered to an entirely new and independent set of criterion groups, cross validation data are obtained. (For related document, see TM 002 947.) (DB)