I didn't see this addressed in the comments on Part 1: how is it possible that expensive marriages had more effect on women's divorce rate than it had on men's?

I didn't see this addressed in the comments on Part 1: how is it possible that expensive marriages had more effect on women's divorce rate than it had on men's? For multiple reasons it is not plausible that a strong correlation within lesbian weddings skewed the results. Was the data self-reported, and this reflects imperfect memory of the expense? Were they only looking at the expense from that partner's side of the family?