Background. Based on the research that has been conducted by the author in October 2011, the number of junior youth in Bantul who had tried smoking by 39.5%. The results also showed that one of the main factors that cause teen smoking is a friend who smokes. Based on these problems, and use one of WHO’s strategy in tackling the problem of tobacco that is through • policy waniing the dangers of tobacco, the author has conducted peer education to increase • youth involvement in prevention SMP smoking behavior in Bantul, Yogyakarta. Objective. Increasing the involvement of teenager through peer education in preventing smoking behavior at Bantul, Yogyakarta Mcthodc Approach. The design of the peer education program using a quasi experimental design approach (quasi-experimental) design with non-equivalent (pretest and posttest) control group (Creswell, 2003). Pretest conducted before training on peer educators and peer. Postes I done 3 months after the training and posttest 2 performed 4 months after the training. Monitoring performed 1 and 2 months after training. Total sample size of 89 people. While the num’er of students who selected 22 people as peer educators. Selection of peer educators method of “snowbaLling” peer educators the choice given to members of the group Conclusion. The conclusion of the program indicate that: 1) there isa strong commitment from the school (intervention group) to form a group of peer educators, 2) There was a significant increase knowledge and peer groups and peer education about smoking and its impact on health (repeated ANOVA test, p <0.05), 3) An increase in intensity group peer educators in prevention peer smoking behavior (Friedman test p <0.05, median pretest (61.5), posttest 1 (67.0), posttest 2 (70 , 0). Recommedation. Monitoring and mentoring of program makers or schools need to do more intensive that the role or involvement of peer educators in the prevention of smoking behavior can be more effective.