Multinomial logit latent-class regression models: An analysis of the predictors of gender-role attitudes among Japanese women

This article describes the method and application of multinomial legit late nt-class regression models in sociological research. This group of models i s the categorical analogue of latent-variable regression for continuous lat ent variables such as those employed in LISREL or LISCOMP; therefore, thoug h never yet employed in sociological research, the potential usefulness of these models in such research is very high. These models are regression ext ensions of loglinear latent-class models with group variables. An applicati on focuses on predictors of three latent classes of gender-role attitudes a mong Japanese women. These classes are labeled "traditional gender-role sup porters," "prowork gender-equality supporters," and "antiwork gender-equali ty supporters." The analysis illustrates the usefulness of the method by id entifying the characteristics of each class compared with the others. Other possible applications of the models are also discussed.