The objective of this study was to identify the different features of New Zealand's National Certificate of Educational Achievement (NCEA) that can affect student outcomes. Despite the huge public debate surrounding its implementation, very little research has been carried out on the features and effectiveness of the NCEA. By analysing the national dataset of the first cohort of NCEA students this study suggests that the current implementation of the NCEA is less than optimal. New models for analysing NCEA data are introduced. Models based on quality of learning were more equitable and demonstrated higher predictability of university achievement than models based on mere accumulation of credits. Recommendations for improving the way the NCEA is conducted and implications for different groups of students are discussed. (Contains 9 tables and 1 figure.)