This study investigated the effectiveness of the KiVa Antibullying Program in two samples of students, one from Grades 1-3 (7-9 years old, N 6,927) and the other from Grades 7-9 (13-15 years old, N 16, 503). The Grades 1-3 students were located in 74 schools and Grades 7-9 students in 73 schools that were randomly assigned to intervention and control conditions. Multilevel regression analyses revealed that after 9 months of implementation, the intervention had beneficial effects in Grades 1-3 on self-reported victimization and bullying (odds ratios 1.5), with some differential effects by gender. In Grades 7-9, statistically significant positive results were obtained on 5 of 7 criterion variables, but results often depended on gender and sometimes age. The effects were largest for boys' peer reports: bullying, assisting the bully, and reinforcing the bully (Cohen's ds 0.11-0.19). Overall, the findings from the present study and from a previous study for Grades 4-6 (Kärnä, Voeten, Little, Poskiparta, Kaljonen, et al., 2011) indicate that the KiVa program is effective in reducing bullying and victimization in Grades 1-6, but the results are more mixed in Grades 7-9.