replication of the results in learning about heterogeneity in returns to schooling (replication data)

A recent article (Koop and Tobias, 2004) proposes a direct way to characterize the extent of heterogeneity in returns to education. They investigate the adequacy of several competing models and conclude that returns to schooling are heterogeneous and are best modelled as a bivariate normal distribution. The results of this replication paper basically agree with the authors.