Those metrics are culturally specific and the dataset of okcupid is limited to people who go on okcupid and volunteer answers to such questions. It's a classic case of participation bias:
https://en.wikipedia.org/wiki/Participation_bias (see "Alf Landon wins in a landslide" for the famous example).
The takeaway here is that straight single middle aged men who chase after woman, are on ok cupid looking for them, and who fill out this survey, are on average, looking for younger women.
Those results cannot be seen as representative of the general population based on the methodology used here.
It gets worse. Similar to the Alf Landon flaw, this could simply be exposing the demographic makeup of the site. Perhaps men think there's a promise on okcupid of younger women so it self-selects for those proclivities. Perhaps women think there's a different promise so they come accordingly.
You have to extricate that. Otherwise, it'd be like going to a gay dating site, running a survey and then come out with some earth-shattering ground-breaking claim of 100% of men saying they prefer intimacy with other men. I mean, duh, look who you're asking.
That's not to say these results don't hold generally, they might. But this is not evidence supporting that.