Race and Sex Variations in the Causes of the Expected Attainments of High School Seniors

Causal models of adolescents expectations have been based on the assumptions of path analysis: Additivy and recursivity. We test these assumptions on a body of 1973 data gathered in Louisville, Kentucky. A second-order interaction of race and sex with all relationships in a model which integrated models proposed by Duncan, Haller, and Portes (1968) and Sewell, Haller, and Ohlendorf (1970) was found, and important similarities between white males and black females were observed. White males and black females showed the strongest reciprocal peer effects on educational plans and the strongest parental-encouragements effects on grades received. There was no feedback from students' expectations to the amount of parental encouragement, although for whites only there was feedback from students' grades. The model best fits the data for white males. This difference in fit, Commonly interpreted as a function of the disadvantaged positions of blacks and females in the stratification system, is futher specified with reference to the particular varibles in the model.