Despite the importance of the National Senior Certificate (NSC) Examination in South Africa, the determinants of NSC achievement and the reasons for its changes over time remain largely unexplored. This gap in understanding is especially pertinent in light of the record-breaking NSC pass rates achieved in recent years, despite greatly accelerated learner flows into Grade 12 (due to pandemic-related leniency). This paper investigates the factors associated with NSC achievement and explores how such associations may have varied over time. It also investigates how much of the change in NSC achievement across time can be explained by observable changes in cohort characteristics. National learner-level NSC data is used, with specific focus on the years 2011, 2019, 2021 and 2023. Analysis is conducted separately for candidates in Quintile 1 to 3 schools and Quintile 5 schools, and methodologies include Ordinary Least Squares (OLS), school Fixed Effects (FE) estimation, and Oaxaca-Blinder decomposition. ‘NSC achievement’ is defined to be candidates’ average mark across their written subjects (excluding Life Orientation), while various learner and school characteristics—such as overage status, subject choice and candidates’ relative academic ability—are included as potential determinants of NSC achievement. Approximately 50% of NSC achievement in individual years can be explained by the study’s OLS models, driven by relative academic ability and, to a lesser extent, overage status and school quintile. A number of factors are shown to be associated with achievement, but associations vary too much to infer reliable relationships. Changes in age profiles and subject choice only partially account for changes in achievement over time, with a far larger share of changes being unexplained, potentially reflecting factors such as educational quality improvements. Counterintuitively, factors outside of the model improved for poorer schools during the pandemic, but deteriorated for Quintile 5 schools. The paper highlights the policy-relevance of determining the reasons for improved performance in the lower- repetition pandemic years, and its findings bring into question if South Africa’s high repetition rates in Grades 10 and 11 are justified—a particularly relevant question, in light of recent evidence
suggesting repetition rates are returning to their pre-pandemic levels.