Objectives: This study was conducted to examine the factors related to the perceived peer smoking prevalence for adolescents. Methods: A self-administrated questionnaire was administered to 352 students in a technical high school in Kangwon Province in May, 2002. The questions included in the questionnaire were concerned with the estimated number of smokers among ten students of the same grade in their school, the sociodemographic characteristics, the smoking-related behaviors and attitudes, and the smoking-related environments. All the students had their expiratory carbon monoxide level measured with EC50 Micro-Smokerlyzer? to verify their smoking status. Multiple regression analysis was applied for data analysis using Windows SPSS 11.5. Results: The former and current smokers overestimated the peer smoking prevalence. Multiple regression analysis for estimating the peer smoking prevalence for the male showed that the perceived smoking prevalence was higher in the female students than in the male students, higher in those students whose mothers had a higher educational level, who had smoked more frequently, who had more best friends smoking, and who had a higher actual smoking rate of the class. For estimating the peer smoking prevalence for the female, the perceived smoking prevalence was higher in the female students than in the male students, higher in those who smoked more frequently, whose five best friends smoked, who had higher actual smoking rate of the class, and who had smoking siblings. Conclusions: This study showed that a higher perceived peer smoking rate is related with their own smoking history and smoking frequency, the smoking related environment, and gender. Smoking prevention and smoking cessation programs need to focus on correcting the falsely perceived smoking prevalence.