Educational heterogamy: a measurement question

The effect of educational differences between partners on marital quality and stability is commonly analysed using difference, or compound measures. This article considers the theoretical foundations for these approaches and examines an alternative approach, diagonal reference models (DRMs). The three methods are then compared empirically, using data on 629 married couples from the survey Child-Rearing and Family in the Netherlands. Difference measures show no effect on marital satisfaction, whereas compound measures and DRMs do show an effect. As DRMs also allow for the quantification of this effect, while being theoretically and methodologically appropriate, they are affirmed the best method.