This study was done to construct and test a structural model to explain health risk behavior of late adolescents.

Data for this study were secondary data from the 2010 Korea Adolescent Health Survey based and 3,675 high school students who participated. Data were analyzed using SPSS 18.0 and AMOS 19.0 programs.

After 7 lines were removed, fitness statistics for the hypothetical model were appropriate (χ²=559.13, p<.001, GFI=.98, SRMR=.03, RMSEA=.04, NFI=.88, IFI=.90, CFI=.90, TLI=.86, AIC=671.13). The result showed that drinking-smoking is directly affected by 5 variables (32.5%), obesity is directly affected by 2 variables (0.7%), lack of physical activity is directly affected by 5 variables (22.2%), skipping of breakfast is directly affected by 3 variables (11.9%), improper sleep is directly affected by 3 variables (7.5%), and psychological adaptation is directly affected by 4 variables (26.8%).

The results of this study, indicate that late adolescents' health risk behavior is affected by many factors with complicate correlations suggesting further study compare youth health risk behaviors in a variety of environments.