We examined the equivalency of the relationship between alcohol use and antisocial behavior between sexes using multiple-group analyses. Examining the equivalency of the directional relationship between the 2 behaviors across sex involves comparing 2 sets of parameters: path coefficients (cross-lagged and autoregressive paths) and the measurement parameters (factor loadings and thresholds for categorical indicators) of ALC and ASB factors. Path coefficients, especially of the cross-lagged paths across time points, estimate the strength of the longitudinal relationship between the 2 behaviors. Even when the path coefficients are equivalent across sexes, the relationship between the 2 behaviors may not be comparable if the measurements of alcohol use and antisocial behavior are not equivalent across sexes. Thus, when we examined the equivalency of the path coefficients, we also examined the equivalency of the measurement parameters of ALC and ASB factors. We fit multiple-group cross-lagged models with male and female participants as subgroups. Models with and without equality constraints on path coefficients and measurement parameters were compared and their differences of chi-squared statistics were tested. Significant differences in chi-squared statistics between models with and
