Two approaches to interest inventory validation are considered. The choice between the two depends on the use being validated. The first approach assumes that interest inventories are to be used in predicting which occupations counselees will enter or prefer. The second assumes that interest inventories are to used in suggesting occupations for counselees to consider on the basis of compatibility of interests. Validation of these two uses of interest inventories requires different treatments of criterion group base rates. As illustrated by data drawn from a published study, the two approaches to validation can produce substantial differences in reported hit rates. Such differences may be found in any study validating group membership predictions if criterion group sizes vary greatly. (Author)