Comparing Two Age-Period-Cohort Methodologies to Model Grade Progression Probabilities

Raquel Guimarães, Centro de Desenvolvimento e Planejamento Regional (CEDEPLAR), UFMG
Eduardo L. G. Rios-Neto, Centro de Desenvolvimento e Planejamento Regional (CEDEPLAR), UFMG

We realized a comparative methodological exercise of two APC estimators: the conventional restrict estimator obtained by the generalized linear models (REGLM) and the so-called intrinsec estimator (EI). The object of our interest are the progression probabilities to the first grade of Elementary School for the brazilian women. The APC modelling of grade progression probabilities is justified since age effects reflect both the mininal age of school entry as well as the trade-off between study and work. Period effects are associated with economical and political conjuctures, as well as with the state of the educational policies. Cohort effects reflect social attributes proper of some group of individuals. Both instruments were contraposed in terms of efficiency and significance of the parameters. The results reveal the potentiality of the APC model based on intrinsec estimator, which presents excelent statistical properties, namely: small variance and a large number of significative parameters.