A total of 1,994 students (approximately 70% of those who participated in the American College Testing Program's Nationwide Study of Career Development) who were working more than half-time three years later responded to an extensive mailed questionnaire followup study seeking to describe the predictability of actual job choice from expressed vocational choice and confidence in choice. Both choices and occupations were categorized into Holland-type job clusters. Analysis involved the use of weighted hit rates and coefficient Kappa. Expressed vocational choices correctly predicted actual occupation two years out of high school for 38% of the total sample (males 40%, females 35%). When certainty level was added to expressed choice, hit rates were 43% for very sure choosers, 38% for fairly sure choosers, and 28% for not sure choosers. Each choice level proved significantly different from the others. Selected implications reported for counseling and research activities are as follows. For research: (1) Holland's typology may not be as powerful for explaining the early careers of semi- and unskilled workers as for managerial, skilled, and professional workers (further followup is suggested); (2) the expressed vocational choice variable demonstrated its viability and reliability via self-report; and (3) Vrooms force model may explain the dymanics of expressed vocational choice (since it was associated with predictability of choice). For counselors, a student's choice seems to indicate both a general direction and an implied prediction that exceeds change--at least within the six job clusters used in the study. Counselors can also infer from this study some employer requirements for entry into broad general groupings. (MEK)