The second condition for establishing an indirect effect is the independent variables, church attendance and importance of religion, must have significant effects on the dependent variable, lying to parents, when the mediating variables are not included. Table 3 depicts the result for the multivariate analysis of adolescent religiosity and lying to parents. Model 1 includes the control variables, church attendance, and importance of religion. As expected, both church attendance and importance of religion have significant effect on lying to parents. Surprisingly, the effect of church attendance on lying to parents is positive, while the effect of importance of religion on lying to parents is negative. That is adolescents who believe that religion is important are less likely to lie to their parents than are adolescents who do not believe that religion is important. However, adolescents who attend church frequently are more likely to lie to their parents than are adolescents who attend church less frequently.