Elementary absenteeism over time: A latent class growth analysis predicting fifth and eighth grade outcomes

Abstract

Chronic absenteeism is deeply detrimental to students' academic and social-emotional development. Prior research on elementary school absenteeism has focused on variable rather than person-centered models, overlooking absenteeism as a continuous process over time. This study addresses this by using a latent class growth analysis with ECLS-K:1998 data to identify classes of student absenteeism throughout elementary school, and uses those to predict their fifth and eighth grade test scores and eighth grade school engagement. Results indicate students fall into four classes of absenteeism, and these are highly predictive of later school outcomes. Individual students tend to stabilize in their absenteeism rates after third grade, and the only truly poor outcomes are for students who have consistently high absenteeism over time, suggesting that as long as overall chronic absenteeism is low, a single year of high chronic absenteeism is not related to highly negative outcomes. Implications are discussed.