In evaluating the deleterious effects of missing in-school time, research has almost exclusively focused on absences, and almost no attention has been paid to tardiness. Hence, this study contributes a new dimension to the field by examining the effects of student tardiness on academic achievement. Employing an empirical model on a large-scale, longitudinal, multilevel dataset of urban elementary school children over 6 years of observations, there are 2 significant findings. First, students with greater tardiness perform worse on both standardized reading and math tests. Second, holding constant an individual's own record of tardiness, students whose classmates are tardy more frequently also have lower test scores. Hence, the achievement gap widens for students in classrooms whose peers have higher rates of tardiness and widens even further for students who also have greater individual levels of tardiness. Policy implications are discussed.