In the present study the estimation of a Production Function is discussed
Educational with information from the PISA database (Program for International
Student Assessment) 2006 from Uruguay. First a method is established
to control the eventual selection bias due to the unobservability of the
dropout students of the education system. Secondly the problem is discussed
of endogeneity of some variables whose effect is of interest from the point
of view of educational policy. In particular, evidence on the
inconsistency of the OLS estimator of the effect of “having attended a school
Full-time during primary education ”, concluding that from the
PISA information can not infer a negative causal effect between the
Attendance at a full-time school and the skills assessed, such as
suggested by the OLS estimate of an Educational Production Function with
Said information base.
The educational production function: the possible bias in estimating institutional effects with PISA data. The case of full-time schools
