There has been little research as to the development of self-rated health among adolescents. In this paper, we present a replication of a model of student self-rated health initially proposed by Vingilis, Wade and Adlaf (1998) using data from 1993 Ontario Student Drug Use Survey (n=840). The data for this replication come from the first wave of the US National Longitudinal Study of Adolescent Health (public release data, n=5673). The original model is tested by a series of nested regression equations and path analysis. The results add support for the original model and are notable given that many items used to build the underlying constructs differ across the two datasets.