The usefulness of vocational interest measures has been questioned by those who have studied the predictive validity of expressed choice. The predictive validities of measured interest for decided and undecided students, expressed choice and measured interest, and expressed choice and measured interest when they are congruent and incongruent were investigated in a follow-up study of 408 university graduates 11 years after they had taken the Strong Vocational Interest Blank (SVIB) and the American College Test. Expressed choice predicted future occupation more accurately than measured interest. The accuracy of both variables was greatest when they were congruent. Expressed choice was found to be a more accurate predictor for women than for men. Predictive accuracy of the single highest SVIB scale was greater than that obtained from a grouping of SVIB scales. The SVIB was equally efficient a predictor for undecided students as it was for decided students. The predictions of future occupational choice from SVIB profiles yielded lower "hit rates" than the retrospective method used in most validity studies. (Author/NRB)