PURPOSE: The purpose of this paper was to describe the characteristics of smoking prevention programs in Korea, to estimate overall effect size of Korean smoking prevention programs, and to investigate effect size variations by program modality and instruction method. METHOD: Meta-analysis was performed on21 programs in 20 studies. RESULT: The estimation of overall effect size for knowledge and attitude was not possible because effect sizes were not homogeneous in this meta-analysis. However, effect sizes of studies that were socially influential programs or active/interactive methods were larger than information-oriented programs or passive/non-interactive methods in the pictures. The effects for behavioral outcomes were generally not as positive and not statistically significant. Q statistics showed that variations among effect sizes within program modality and instruction method classifications were heterogeneous. CONCLUSION: The results from this meta-analysis support the continued use of socially influential programs and active/interactive methods for smoking prevention programs. Because behavioral effect might be the fundamental objective of smoking prevention programs, the present results indicate that smoking prevention programs should consider adopting more effective programs.