Like in many other countries, few investigations have been carried out in Chile to measure the long-term effects of nutritional status at an early age on scholastic achievement in a multicausal approach. The objectives of the present study were to describe the impact of nutritional, intellectual, family, educational and socio-economic variables at the onset of elementary school in 1987 that may affect achievement on the academic aptitude test (AAT) taken in 1998 at the end of high school, and to quantify the impact of these independent variables on the AAT. The present study comprises two cross-sectional stages: in 1987, a representative sample of 813 elementary school first-grader Chilean children from the Metropolitan Region was randomly chosen; in 1998, 12 years later, 632 school-age children were located and only 351 of them graduated from high school and, from these, 260 students took the AAT. In 1987 nutritional status was assessed through anthropometric parameters, intellectual ability by the Raven's Progressive Matrices Test, scholastic achievement through Spanish language and mathematics tests, and socio-economic status using Graffar's modified scale; family variables were also recorded. Maternal schooling, scholastic achievement, intellectual ability and head circumference-for-age z-score (anthropometric indicator of both nutritional background and brain development) all in 1987 were the independent variables with the greatest explanatory power for AAT variance in 1998 (r2 0·402). These results provide a foundation to identify the risk factors at an early age that affect AAT scores and should be useful to improve nutritional and educational policies.