This study was designed to determine the relative effectiveness of a nongraded approach to elementary mathematics instruction and a traditional graded approach by comparing an experimental group of pupils who had had six years of nongraded mathematics instruction with three control groups who had had six years of traditional graded mathematics instruction.