A short overview of an experimental evaluation of lower teacher-pupil ratios is presented in this report. The research under review is the Student Teacher Achievement Ratio (STAR) project. This longitudinal project was conducted in Tennessee and was designed to prove to state legislators the efficacy of smaller class sizes. For STAR, evaluation data included standardized measures of student outcomes and progress, teacher logs, observations in classrooms, student data (attendance, behavior, age, race, sex, free lunch, etc.), and data about teachers and administrators. Some of the positive results that emerged were: (1) small classes benefit all students, but minority and traditionally hard-to-teach students received approximately twice the benefit from the same investment and treatment; (2) small classes benefited teachers and parents and improved instruction; (3) students in small classes were less likely to be held back than students in large classes; (4) STAR students performed better on all measures; and (5) benefits obtained in K-3 remained with students up through at least grade 9. Other findings, such as the ineffectiveness of teacher aides, are also reported. Contains 27 references. (RJM)