in the model. To test the significance of each of the interaction parameters, we dropped each one in turn and compared the fit of the nested models (those in which the parameter was dropped) with the fit of the model in which all parameters were freely estimated. Furthermore, differences across sex were examined by comparing the fit of a model in which male and female estimates were free to differ with a model in which they are constrained to be the same. Comparison of the fit of nested models with the full model was tested using a likelihood ratio test. That is, the difference in the −2LL between the full and nested models was treated as a χ2 statistic, with degrees of freedom equal to the difference in the number of estimated parameters. A significant χ2 (p < 0.05) indicates that the nested model provides an unsatisfactory fit to the data, and thus, it should be rejected in favor of the fuller model. A non-significant χ2 indicates that the constraints assumed for the nested model do not significantly worsen the fit to the data; the nested model would be preferred as more parsimonious.
