A Causal Inference Seminar with Magdalena Bennett.
High-stake testing plays a crucial role in many educational systems, guiding policies of accountability, resource allocation, and even school choice. However, non-representative patterns of attendance can skew how useful these measures are for accomplishing their main objective. Are we really measuring the quality or performance of a school if a non-representative sample of their students are taking the test? In this paper, we study the effect of high-stake testing on student composition on the day of the test using rich administrative data from Chile and daily attendance. By combining an event-study framework and a machine learning prediction approach, we find that, overall, attendance patterns are highly heterogeneous. While in lower grades, students that perform worse in comparison to their peers are more likely to be absent the day of the test, the opposite is true for higher grades, where high-performers show an increase in attendance. Additionally, we find that individual schools present different attendance patterns according to their characteristics, which is particularly important in the presence of general policies that aim to impute missing information.