Studying the Effects of Early Absences on Later Absences

• In Event: Excellence in Education Research: Early Career Scholars and Their Work

Abstract

In the last ten years there has been a renewed interest in the study of causes and consequences of school attendance. Part of this interest is motivated by the fact that school absences are at the center of the recently implemented school accountability systems that link school resources to improvements in student attendance patterns. Studying the effects of school absences is also driven by recent research findings highlighting the role of non-cognitive skills as determinants of fundamental school outcomes, as well as outcomes that extend well beyond the school cycle. Presently, both the educational policy and the academic community appear interested in a deeper understanding of the multiple causes and consequences for school outcomes associated to poor school attendance. Despite a rise in the academic literature on causes of school absences, no review has studied how early absences are causally related to later absences as its subject.
The present research aims at contributing to the overall understanding of the causes of school absences by inquiring how early school absences are causally related to later absences with reference to specific points through a child’s educational cycle. We propose an analytic model that reveals a key feature of the analytic task by emphasizing, as other researchers have noticed, that school absences are an age-dependent phenomenon. On the basis of fundamental insights we gain from that model, and with the objective answer as our core research question, the main task of the present research involves using “state of the art” statistical methods to estimate the causal effect of early absences on latter absences. Those methods (dynamic panel data models), we claim, allow dealing with complex threats of endogenous variables, when those variables are lagged dependent variables, and allow us to recover unbiased estimators of the causal effect of lagged absences.

Author

• Wladimir Zanoni, University of Chicago