Research on the effectiveness of educational inputs, particularly research on teacher effectiveness, typically overlooks teachers' potential impact on behavioral outcomes, such as student attendance. Using longitudinal data on teachers and students in North Carolina I estimate teacher effects on primary school student absences in a value-added framework. The analysis yields two main findings: First, teachers have arguably causal, statistically significant effects on student absences that persist over time. Second, teachers who improve test scores do not necessarily improve student attendance, suggesting that effective teaching is multidimensional and teachers who are effective in one domain are not necessarily effective in others.