Marital assortment and phenotypic convergence: longitudinal evidence.

This study provides a direct test of whether the observed similarity of spouses is due to initial assortment rather than to convergence of phenotypes. With data from three well-known longitudinal studies, phenotypic convergence is examined using both variable- and person-centered analyses. The longitudinal evidence does not support the hypothesis that couples increasingly resemble each other with time. Spouse correlations most likely reflect initial assortment at marriage and not the convergence of phenotypes.