Actually, sex without consent is the definition of rape. Consenting to sex when you didn’t want it isn’t rape. Both men and women give into having sex to satisfy their partners when they don’t really want it, I’d hardly call that rape because consent was involved and I don’t think most reasonable people would. Considering things like that “rape” gives you the inflated rape statistics the OP points out and does a real disservice to actual rape victims when you pretend that is equal to the real assault they faced.

1. As others have mentioned, it is vaguely plausible that in an era in which the contextual background of the question was less clear (because issues like date rape were less openly discussed), some respondents might have interpreted it in terms of wishing they hadn’t done something that they in fact consented to at the time. I’m not sure that is very plausible, but it is worth considering.

2. Another interpretation that I think is more plausible could be paraphrased as the question “Have you ever consented to sex under the influence of alcohol or drugs even if you didn’t really want to do it?” There is a discernible difference between consent and wanting. The question could be interpreted as asking if respondents had ever consented to sex to make their partners happy when they didn’t feel like doing it. This may be an unrewarding experience (resembling prostitution), but if it’s consensual, it wouldn’t meet a legal definition of rape.

No study should be taken in isolation even if its methods are flawless. To assess the prevalence of rape, it’s not even sufficient to point out that other studies have found similar results. One has to take the totality of evidence that is available so as to avoid the charge of cherry picking just those results that agree with whatever agendas one might have.

Publication bias is also an issue that might make it more likely that studies with alarming results are published more often than studies that produce less alarming results. The publication bias might also go the other way with studies that report more conservative statistics being taken more seriously. As it stands, rape statistics vary widely from study to study, so the field is very difficult to negotiate. If nothing else, this variability suggests serious problems of credibility within the field.

Theories of what to make of college costs abound, of course. Here is one of mine:

We can think of a resume as a currency used to “purchase” jobs in the labor market. This is a nominal asset, like money, since it only matters that you have more than everyone else.

There are a number of expensive things that people do–volunteering in Africa, taking unpaid internships, attending expensive schools–to boost their resume, presumably in the hopes of trading it in for a good job.

Thinking of resumes as currencies can allow us to come up with a resume-to-dollar exchange rate, which reflects the amount of money one could spend to ceteris peribus boost one’s resume.

(There are a lot of measurement problems here that I’ll gloss over. But this is the intuition.)

You can read more about the idea at my own blog:
http://kmthinking.blogspot.com/

(BTW the link to the Econometrica paper doesn’t work).