Causal path analysis was used to determine the direct and indirect influences of intellectual ability, quality of instruction, achievement motivation, quantity of academic coursework, and time spent on homework on high school students' learning, as measured by their grade point averages (GPAs). Other relevant influences on grades were controlled. The theoretical models proposed were recursive path models; paths were estimated by the beta weights from multiple regression analysis. The sample population consisted of 13,152 high school sophomores selected from the base year (1980), and first (1982) and second (1984) follow-ups of the High School and Beyond Study. The subjects were high school sophomores in 1980 and high school seniors in 1982. Although only the 1980 and 1982 date were used in this study, the second follow-up data tape was used to include GPAs and coursework in the "new basics" present in student transcripts rather than simply in student self-reports. The path model included ethnicity, family background, gender, and GPA as well as the variables detailed above. Results indicate that coursework in the new basics, intellectual ability, and academic motivation were the most influential variables affecting grades. Homework had a small, but meaningful, effect. Although quality of instruction did not have a meaningful direct effect on grades, it did affect other variables in the model, such as motivation and coursework. Three data tables and two figures are included. (TJH)