This paper draws upon data from a longitudinal study of students in four Wisconsin elementary schools to probe for the linkages between school- and home-related variables and students' academic achievement. Data were collected from fall 1979 through spring 1982. Subjects, enrolled in the third grade during the 1979-80 school year, were followed through their fourth and fifth grade years. The sample also included children who entered school in the fall of 1980, at the beginning of their fourth grade year. Additional data were collected from parents, teachers, other professional staff, and school and district administrators. Findings indicated that the variables included in the analyses were not particularly helpful in understanding students' gains in reading and mathematics. Findings further emphasize the complexity of human learning and the uniqueness of individual learners. Neither time-on-task in reading or mathematics nor time spent on homework was a potent predictor of student gain. Students of teachers who held a graduate degree did less well than students whose teachers did not have a graduate degree. Teachers' attitudes and beliefs were at least as important as other teacher characteristics. Finally, students' academic aptitude was not as potent a predictor of student gain in reading and mathematics as might be expected. (RH)