Field of study and partner choice
There is strong assortative mating by field of study. To examine to which extent this is due to self selection or to a causal effect of access to specific "marriage markets", we use data from participants in admission lotteries of four oversubscribed studies in the Netherlands. For each of the four studies, we find that the winning compliers of an admission lottery are significantly more likely than the losing compliers to have a partner from the lottery study, whereas losing compliers are only marginally more likely to have a partner from the lottery study than would occur under random matching. These results indicate that assortative mating by field of study is largely due to marriage market access and that self selection plays a minor role.