Abstract: This paper examines the extent to which year-9 performance on the National Assessment Program—Language Arts and Numeracy (NAPLAN) predicts access to higher education as determined by subsequent achievement on year-12 Victoria Certificate of Education (VCE) exams. VCE performance is measured via three binary indicators: achieving an Australian tertiary admission rank (ATAR) above 50 ("ATAR50"), above 70 ("ATAR70"), and above 90 ("ATAR90"); and two continuous indicators: ATAR and the Tertiary Entrance Aggregate (TEA). We find that a four-way classification of year-9 NAPLAN results explains 35% of the variance in ATAR50, 37% in ATAR70 and 26% in ATAR90; and NAPLAN scores and basic demographic indicators explain 38% of the variance in ATAR and 42% of the variance in TEA values. Examining the joint effect of year-9 NAPLAN scores and socio-economic status in predicting VCE outcomes, we find that while both are significant, NAPLAN scores have a much stronger effect. At the school level, we find that predictions of success rates based on NAPLAN scores and basic demographic indicators.