This paper presents the research design for a longitudinal study of elementary school students, which will examine patterns of cognitive, affective, and social growth and identify the school and home variables which influence such growth. Shifts in dependent variables during a single school year will be examined during the initial year for all grades at the elementary school level. During this year, considerable attention will be paid to the actual classroom behavior of the teacher. During the subsequent 2 years, continuing data will be collected for a first and fourth grade cohort. Thus, in a 3-year period a picture of growth during the initial 6 years of schooling will be obtained. Various aspects of the longitudinal study which are discussed in some detail in this research design are: assessment of variables, testing instruments, sampling plan, testing sequence, testing schedule, migrating children, and plans for data analysis. (Author/CS)