JulianMorrison:

Seems pretty obvious that hypergamy is what poor women do in societies that only let them gain control of resources through marriage. It's a rational adjustment to a sexist, unequal society, not some sort of instinct.

This is a hypothesis worth investigating, but how much data seems to support it? The research I've read supports the existence of hypergamy in both modern societies, and in pre-agricultural societies without high levels of gender inequality.

The Dalmia study cited on Wikipedia supposedly doesn't find women "marrying up," but since I can't read the full text I'm not sure how they were operationalizing "marries up." For instance, perhaps the study found that women don't marry up in wealth. But that doesn't mean they don't marry up in education, which is what this study found:

Contrary to popular beliefs, the increased concentration of women at the top of the education distribution has not resulted in a worsening of the marriage market prospects of more educated women. The “success gap” declined substantially in the 1980’s and 1990’s. The marriage market accommodated the shift through a decline in hypergamy at the upper end of the education distribution.

On the other hand, the declining economic prospects of men at the bottom of the education distribution have rendered many below the threshold of marriagiability. The likelihood of a 40-44 year old man with 11 years of education being married fell by over 20 percentage points over the 20-year period, a greater decline than that for women of the same education level. There was no decline in hypergamy at this end of the spectrum; in fact, some measures indicate an increase in hypergamy for this group, as women have increasingly been reaching upward in the education distribution for husbands.

In short, education hypergamy exists, but it’s getting weaker at the top (presumably because there is a shortage of higher-education men to date up to), and may be getting stronger at the bottom.

For women of high socioeconomic status, hypergamy does appear to decrease. For instance, women in some college samples tend to not care about men's wealth very much. Though this could also be partly because those women are more oriented towards short-term mating.

Even in a modern, short-term mating context, it's not clear that hypergamy disappears. In a speed dating study, Asendorpf & Penke found

The key finding for popularity was that both men and women’s popularity was largely based on easily perceivable physical attributes such as facial and vocal attractiveness, height and weight. This was already the full story for women’s popularity in speed-dating, that is, men used only physical cues for their choices. In contrast, women included more criteria, namely men’s sociosexuality and shyness as well as cues for current or future resource providing potential, such as education, income, and openness to experience (but not cues of steady resource striving like conscientiousness).

Note how eduction and income mattered to women, but not to men. Those are elements of hypergamy. Avoiding shy men is also hypergamy because shyness is low-status in Western men.

For another example of modern hypergamy, observe the attraction of women to rockstars and actors. Yet do women become groupies of rockstars merely in hope of gaining their resources through marriage, as a rational adjustment to a sexist society? I doubt it.

Is modern hypergamy merely a hold-over due to outdated norms? No. In pre-agricultural societies where women don't economically depend on men, hypergamy still exists. Anthropologists used to be bamboozled by the discovery that the lioness' share of calories in some cultures is supplied by the women. So why were the men hunting, if it was so inefficient? Anthroplogists eventually came up with the hypothesis that male hunting isn't (just) about providing meat.

Hawkes and Bird argue that a large function of men’s hunting isn’t putting food on the table for their families, but rather showing off to gain social status and mating success. The researchers observe that competent male Ache hunters have greater mating success:

The families of better hunters end up with no more meat than other families. Hill and Hurtado’s demographic data show little difference in survival risk for the children of better hunters. But men rated as better hunters had much higher fertility. In a smaller data set, better Ache hunters were more often named by women as lovers and as secondary fathers of more children. (Secondary fathers are men other than a mother’s husband who were sexually involved with her at the time of her pregnancy). Ache women did not nominate hunting skill as a criterion for choosing a mate, but men emphasized its importance for success with women.

Since the hunter's skill doesn't translate into more provisioning for his family, the it's difficult to explain women's preference for hunters as a response to economic deprivation. Women don't have to date good hunters to feed their children, but they do anyway.

In other ethnographic cases, hunting success is also associated with advantages in male competition. Hadza men foraging in northern Tanzania are big game specialists (Fig. 1). As among the Ache, hunters do not control the distribution of meat. In this case, the wives and children of better hunters do have more positive weight gains, and those wives have surviving children faster. But these differences are directly associated with the foraging effort of the women themselves. , As with the Ache, the wide sharing of meat means that Hadza women and children receive little of their meat from kills by their husband and father. Consistent with this, a father’s death or parental divorce has no effect on child survival. However, better Hadza hunters tend to be married to harder-working wives. Older men who are better hunters have younger wives, suggesting they are more likely to leave an older wife to raise a second family—another way they have increased success in competing for paternity. Meriam turtle hunters also have higher age-specific reproductive success than do nonhunters and, as with the Hadza, this seems due to assortative mating: hunters claim more fertile wives than do nonhunters.

Successful hunters gain high status, have more partners, and experience greater reproductive success. That's hypergamy.

The "economic inequality" hypothesis does not explain this pattern of women "dating up" in terms of <insert culturally valued pursuit/trait here>. It does seem plausible that women start caring less about men's economic status in prosperous societies, but that doesn't mean that women have stopped being hypergamous.

Without the need to mate with good providers to put roofs over their heads, women are free to go after the men they are attracted to, which seems to mean dating up on other dimensions they care about such as personality traits, education, status, intelligence, and accomplishments (some of these traits have been discussed in this comment, and others will have to wait for another time). This appears to be a generalized phenomenon; for instance, women care more about humor in their partners than men do, another culturally-valued trait.

As far as I can tell, this pattern of evidence looks a lot more like some sort of instinct on the part of women than merely a response to economic inequality (those as I've mentioned above, economic inequality is a factor in how hypergamy is expressed). The other problem with the sociocultural inequality hypothesis is that it can't explain how gender inequality came about in the first place: clearly there are some pre-cultural forces in play. It's difficult to make any sense out of this data without invoking evolutionary theories like sexual selection.