Research in education and developmental psychology indicates that behavioral engagement in learning is a critical predictor of children's academic success. In an effort to improve academic achievement, school administrators are continually in search of methods to increase behavioral engagement. Previous research has indicated that classroom-based physical activity (PA) lessons have a positive impact on academic achievement. However, little research has been done in assessing the impact of such interventions on the behavioral engagement of students with learning behavior difficulties. This study assesses the impact of classroom-based PA on teacher-rated classroom behaviors of students with identified learning behavior difficulties. Two schools (one intervention, one control) participating in a larger, cluster-randomized trial provided scores on a teacher-administered classroom behavior scale. This scale was used to collect information on 15 characteristics identified as being essential to behavioral engagement. Participants included male and female students in second and third grade classrooms who were identified by their classroom teacher and school counselor as having difficulties with learning behaviors. Mixed linear modeling for repeated measures was used to examine the changes over time in the classroom behavior scores. The intervention group showed significant improvement over time in classroom behavior while the control group showed no change or a slight degradation over time (i.e., group × time interaction, F[2132] = 4.52, p = 0.01). Schools must meet the diverse needs of students today, including those who exhibit less than optimal learning behaviors. Combined with the evidence that PA is linked to several health and cognitive-behavior benefits, providing classroom-based PA that is incorporated within the curriculum provides common ground for all students to participate. It is a potential solution to increasing behavioral engagement, and in turn stimulating and enhancing learning.