Empirical results and conventional wisdom : primary school effectiveness in Burundi

Data were obtained from a probability sample of 47 schools in 27 school directorships in rural areas, and from approximately 1,946 grade six students in 1989. School directors and teachers were surveyed and the students tested in language arts, mathematics, and science-agriculture. Subsamples of students took these tests in French while other subsamples were administered the same tests in their mother language, Kirundi. Language of assessment was found to affect the measurement of achievement, especially for students of high ability. The type of school a student attended (i.e., whether it was a core or a satellite school) had little influence on student test scores. Several school and instructional characteristics were shown to relate to student performance, and a model of school effectiveness is presented. Relationships between frequency of teacher supervision by the school director, instructional practices that increase opportunity to learn, and measures of teacher skills are investigated and shown to affect student performance in various ways. Among student characteristics, repetition in grade six was strongly and positively related to student performance, especially on those tests administered in French that were most comparable to the secondary school entrance examination. The findings of the paper are in some ways a vindication of conventional thinking about what makes for effective primary schooling in Burundi. They attest to the value of close teacher supervision by school directors, the positive effects of repetition, and the importance of such indicators of teacher engagement as punctuality. But in other respects the findings deviate from conventional wisdom about how language of assessment influences the measurement of individual differences or about how variations between core and satellite schools affect student achievement. The paper concludes with a call for focused experimentation to identify effective strategies for raising levels of student achievement, improving monitoring of student learning, and increasing internal effectiveness so that the progress already made in expanding access to primary schooling can be consolidated and reinforced. (Author abstract)