Methodology

Most of the previous attempts to measure educational attainment have treated education as a categorical variable, whose mean is computed as a weighted average of the official duration of each cycle and attainment rates, thus omitting differences in educational achievement within levels of education. This aggregation into different groups may result in a loss of information introducing, therefore, a potential source of measurement error.

classroom-2

With the aim of improving educational attainment estimates, we explore a more nuanced alternative, which considers the continuous nature of education. This continuous approach allows us to impose more plausible assumptions about the distribution of years of schooling within each level of education, and to take into account the censoring issues characteristic of these data in the estimation.

The main assumption of our methodology is that the same functional form is used to model the distribution of years of schooling in all countries over the whole period. However, the distribution of years of schooling in a developed country may be different from the shape exhibited in a developing economy. At the same time, the shape of the distribution might suffer important transformations over time. We argue, therefore, that a flexible parametric specification must be considered. More specifically, we employ the generalized gamma distribution to model the time that individuals attend the school until they complete the educational cycle or decide to drop out. It should be noted that, even when we assume the same model in all cases, the parameter estimates vary across countries and years, thus allowing us to model particular features of the distribution of schooling.

For a more detailed description of the methodology see: Jorda, V. & Alonso, J.M.