Reviews hierarchical linear models that have been developed to address problems of the measurement of change and the unit of analysis in educational research. Introduces a three-level hierarchical model that should constitute the basic paradigm for future quantitative research on student learning. (Author/BJV)

Argues that the personal commitments of teachers, common sense, and normative requirements can be valid bases for action. Asserts that overreliance on research knowledge is unwarranted: its value lies primarily in the scientific ethos and in processes of inquiry and only secondarily in the facts researchers lay claim to. (Author/KH)

Examines patterns of enrollment, choice, and degree attainment of undergraduate students attending college between 1975 and 1986 in five metropolitan areas with private institutions as well as public systems enlarged during the postwar period. Finds that minority access as well as achievement have diminished considerably since the mid-1970s. (DM)

Reviews "School Matters" by Peter Mortimore, a rigorous longitudinal study of the effects of elementary schools in London. Concludes that school differences influence student development, and explores the connections between school leadership, classroom life, and student cognitive and noncognitive development. (FMW)