Abstract
The purpose of this study was to develop and test two alternate causal models of positive health practices in adolescents, which built on a modification of the original theoretical formulation previously tested in adolescents by Yarcheski and Mahon (1989). The sample consisted of 202 adolescents, aged 15 to 21. In classroom settings, they responded to instruments and a demographic data sheet measuring variables in both models (age, gender, self-esteem, social support, and positive health practices), and to additional variables in Model 1 (future time perspective) and Model 2 (perceived health status). Both causal models were tested by the LISREL 7 program. By all indicators used in the study, there was a very good fit of both models with their data, there were no specification errors in either model, and neither model required modification. Visual inspection of the indicators suggested that Model 1 had a slightly better fit than did Model 2 with their respective data. These results suggested that causal models explaining positive health practices should include endogenous variables that share a common orientation, such as those with a psychosocial focus versus those with a health-related focus.