Predicting early grade retention: A longitudinal investigation of primary school progress in a sample of rural South African children

Background. One hundred and fifty rural South African children, newly enrolled in Grade 2 in 1994, were retraced in 1998 when they were scheduled to have entered Grade 7. Only 39% of the cohort had progressed smoothly to Grade 7; more than a third (36%) had left their original primary school, and 25% had been retained at least once. Aims. The present study investigated factors that were measurable at the start of Grade 2 which proved useful in predicting subsequent retention. Method. Details of children's academic progress from Grade I in 1993 through all subsequent years including 1998 were collected. Predictor variables included age at school entry, sex of child, nutritional status, academic achievement in Grade 1, cognitive test status at Grade 2, teacher assessments of children's behaviour, and biographical variables such as caregiver education and household size. Results. Rural children's experience of primary school was relatively disrupted. For those who remained in the same school, a relatively good predictive model for retention was built, with Grade I academic achievement as well as caregiver education, and cognitive test scores being important predictors. Conclusions. Results corroborate those of developed world studies, in showing that early academic achievement is a strong predictor of retention. In addition, they highlight the importance of early curriculum mastery - rather than broader cognitive skills - for smooth progression through school.