Reflections on the Registered Report Process for “Academic Harms of Missing High School and the Accuracy of Current Policy Thresholds”

I am grateful to have had the opportunity to preregister data with AERA Open, as this gave me an opportunity to replicate a long-established line of research examining the relationship between missing school and academic outcomes but with much greater attention to analytic details. To my knowledge, there are few (if any) studies that have pre-registered hypotheses, variables, and regression models prior to analyzing administrative data provided by school districts. As a researcher who almost exclusively uses secondary data, this was certainly a unique opportunity.

Registering my study prior to conducting analyses forced me to be more intentional in my conversations with the school district providing the data. Knowing that I would be expected to follow through using the empirical models I specified a priori, I thought more deeply about what data I would have access to and what coding, construction of variables, and inclusion of covariates would be most ideal for someone who was unfamiliar with the data (e.g., reviewers). Apart from the logistics involved with preregistration, this is a process that I think is undervalued by many current researchers, especially those with access to immense amounts of data or those who conduct numerous studies using the same data source.

This special topic also gave me pause to understand more about the evolving processes of preregistering data, registering a report, and general perspectives in making a priori hypotheses. Regardless of whether these extra steps in the research process become the norm for most researchers, understanding their principles and why these processes came about should give any data analyst pause about the implications that come from using secondary data.