School Closures and Effective In-Person Learning during COVID-19

Social scientists have developed several schooling mode trackers to measure in-person, hybrid, and remote learning of students during the COVID-19 pandemic. In this paper, we compare eight of the most popular trackers for the U.S. and uncover substantial temporal and geographical differences, due in large part to how the trackers define the three schooling modes. We then estimate a new measure of effective in-person learning (EIPL) that combines information on school learning mode with cell phone data on school visits. The new measure provides a single number of the fraction of time that students spent learning in person and is made publicly available for a large, representative sample of both public and private schools. Consistent with other studies, we find that a school’s share of non-white students and a school’s prepandemic grades and size is associated with less in-person learning during the 2020-21 school year. Notably, we also find that schools in more affluent localities with higher pre-pandemic spending and schools receiving more federal emergency funding provided lower EIPL. These results are explained in large part by regional differences, reflecting political preferences, vaccination rates, teacher unionization rates, and local labor conditions.