Primary school achievement in english and mathematics in Zimbabwe: A multi-level analysis

Using a multilevel modeling procedure, the authors explore: the percentage of variance in primary school achievement in Zimbabwe that could be attributed to the types of schools and classes attended; the differences between schools in student achievement in mathematics and English; and the reasons for these differences. They compare five types of schools. Students in Former A (European) schools and high-fee schools outperform those in Former B (African) schools, low-fee schools, and district council schools in both subjects. In English, school-type differences persist after controlling for student intake variables. For mathematics, they disappear. School and class variables related to higher math achievement include the amount of teacher training and instructional time, and pupil-teacher ratio. Higher achievement in English is related to the pupil-teacher and textbook-to-pupil ratios, and to the amount of teacher training.