An important paper by Chiappori et al. (2012) has proposed an elegant and parsimonious model of spousal matching over multi-dimensional characteristics. Importantly, the model suggests specific testable assumptions that allow researchers to uncover marginal rates of substitution (MRS) between spousal traits, and the authors use the Panel Study of Income Dynamics to show that the model is not rejected by the data. In this paper, we extend the analysis to two additional representative samples of US couples and find in each case, the data reject the MRS implication of the model.