The impact of in-service teacher training on student performance. Evidence from a developing country

About this event

There is limited research about the impacts of in-service teacher training on students' performance; more importantly, we don’t know how to incentivize the take-up of effective teacher training and continuing professional development (Burgess, 2016). The main problem in identifying the causal impact of training is the selection of teachers to training and the selection of students to schools. One way to overcome the first problem is to randomly assign teachers, who are otherwise very similar in their characteristics, to training. In this paper, we use a regression discontinuity design that takes advantage of the fact that in Ecuador teachers need to get a minimum score at a knowledge test to be promoted to the next earning category. Teachers whose scores are above the threshold have very similar characteristics to teachers whose scores are below the threshold; however, the former are motivated to undergo substantial training to get a promotion, while the latter are ineligible for promotion. In line with the previous literature, we find that students of teachers who are marginally eligible for a promotion perform better at standardized math tests than students of teachers who are ineligible for promotion. This is not the case for verbal tests which suggests that probably the content of the courses does not match the content of the tests that students take.