To study the effects over time of teacher practices concerning parent involvement on student achievement test scores, longitudinal data from 293 students (in Baltimore, Maryland, Grades 3 and 5) who took the California Achievement Test in the fall and spring of the 1980-81 school year, were analyzed. The students were in the classrooms of 14 teachers who ranged in their emphasis from confirmed leaders to infrequent users to non-users of parent involvement in learning activities at home. Results show that, from the fall to the spring, students whose teachers were leaders in the use of parent involvement made greater gains in reading achievement than did other students. However, there were no effects on math achievement for students whose teachers involved parents. These results are the first to link particular teaching practices concerning parent involvement to the actual responses of the parents and the change in achievement of their children. Apparently, important consequences can occur when teachers make parent involvement in learning activities at home part of their regular teaching practice. (KH)