School-level predictors of student academic achievement in the kenya certificate of primary education examination using a two-level hierarchical linear model

Available literature on the predictors of student academic scores since the 60s continues to present mixed findings and interpretation. Utilizing the education production function models, researchers have sought to test whether school or teacher-level variables explain academic achievement variance to a greater extent than do student-level variables. Within this framework, we modelled school-level predictors of academic achievement in the Kenya Certificate of Primary Education (KCPE) examination in Mumias and Kuria East Sub-Counties in Kenya. Using a two-level hierarchical linear model (with 1824 students at Level-1 nested within 61 schools at Level-2), the explained variance in student scores by the two levels was 64.00% with 56.64% of that explained by school-level variables. At Level-2, School location in the two Sub-Counties; Boarding schools or a boarding component in a day school (often involving Class 6-8); and whether or not schools allowed their students to borrow school library textbooks and other learning materials for study and reference away from school were flagged as predictors of student academic achievement in KCPE after meeting the standardized regression coefficient cut-off of ≥0.10 to be flagged as such. Policy implications of these findings are discussed.