This study extends previous research on labor market effects of vocational education by estimating the relationship between vocational education in high school and the intervening factors in its relationship to labor market outcomes. The strategy is to use data from a stratified national longitudinal survey to estimate a simplified, reduced-form model of outcomes for individuals that can contribute to understanding why positive earnings effects have been hard to find for men and why the effects vary between men and women. The estimated model shows that (1) vocational education may have both direct and indirect effects on earnings, income, and unemployment; (2) the indirect effects operate through such intervening factors as unionization, industry, occupation, labor market experience, and postsecondary education; and (3) the indirect effects differ between men and women and between whites and nonwhites. The relatively small total effects on males' earnings are more likely attributable to imprecise specification of curricula and neglect of the importance of finding training-related work than to tendencies for conflicting indirect effects to offset each other. Training-related placement is a significant distinction in estimating earnings differentials. Benefits are attributable to occupationally specific skills rather than general work habits or attitudes. Four policy implications are suggested. (YLB)