Recent molecular genetic studies provide suggestive evidence for sexual dimorphism in genetics of obsessive-compulsive disorder. However, only a few twin studies have addressed the question of sex differences in genetic and environmental contributions to variation of obsessive–compulsive symptoms. The aim of the present study was to estimate genetic and environmental influences on obsessive–compulsive symptoms in South Korean twins, with a special emphasis on sex difference. In total, 751 adolescent and young adult twin pairs (ages: 13–23 years) completed a Korean version of the 30 items of the Maudsley Obsessional — Compulsive Inventory (MOCI) through a mail survey. A sum of the answers for the 30 items was calculated to represent a total score for obsessive–compulsive symptoms (hereafter, the MOCIT). Males had significantly higher variance of the MOCIT than did females. In males, monozygotic (MZ) twin correlation was significantly higher than dizygotic (DZ) twin correlations (.56 vs. .24), whereas in females, MZ and DZ twin correlations were not significantly different from each other (.39 vs. .36). The general sex-limitation model was applied to the twin data. The results of model-fitting analyses indicated that the unstandardized genetic variance as well as heritability estimate (53% vs. 41%) for the MOCIT was higher in males than in females. However, shared environmental influences did not attain statistical significance perhaps due to insufficient statistical power.