Sex-differentiated regression models were run for each of the dependent variables
(see Tables 3-5). The full model, which incorporates controls for all of the independent
variables—household attributes, individual characteristics, individual pursuits, and
measures of values, is a good fit for each of the dependent variables. The model accounts
for 8.8% of the variation in the women’s responses and 7.4% of the variation in the men’s
responses on the first dependent variable, which measures agreement with a traditional
gendered division of labor. When applied to responses concerning the perceived effects
of maternal employment on child well-being, the model explains 11.8% of the variation
in the women’s attitudes and 11.4% of the variation in the men’s attitudes. When
inspecting attitudes concerning the quality of a working mother’s relationship with herchild, the full model accounts for 7.7% of the variation in the women’s responses and
9.8% of the variation in the men’s responses