In the three - way - analysis of variance with partnership, age, and education as factors and the AAS depend as continuous dependent variable (N = 1675), three small but
relevant significant main effects of partnership, F (1, 1663) = 24.41, p < 0.001, η2 = 0.01, age, F (2, 1663) = 8.56, p < 0.001, η2 = 0.01, and education, F (1, 1663) = 11.92, p < 0.001, η2 = 0.01 were observed.
To test the moderator hypothesis, it is important that the interaction term has a
significant effect on the criteria, for the two predictors (predictor and moderator) the
main effects are not
relevant (Baron and Kenny, 1986).