All I can add is that I suspect that the small sample size would add a lot more to the variance of the results than might otherwise be expected; anecdotal evidence suggests that women who consider themselves to be ‘desirable’ are less reliable than average, which is given credence where the researchers note “Participants only logged on to the web site on about half of the days when they worked (27% of the 60 requested days).” Further, Lap dancers are probably not skilled accountants, and so the self-reported earnings might also be questionable, especially if any of them were trying to avoid paying taxes.

Unfortunately, very little mention is made of the 14-page questionnaire that the dancers were given to complete. Possibly it asked whether the dancers were in a relationship or otherwise sexually active, although I suspect that since money would be the prime motivation for becoming a Lap dancer, the women would be incentivised to make themselves as attractive as possible, whilst a boyfriend effect would be minimal, and possibly even counteracted if the sample was random enough. I’d also hazard a guess that the median lap dancer would tend to rate as more attractive than a median member of the public.

However, the problem with your suggestion for a well-controlled study is that the women selected may feel compelled to try to ‘help’ the researchers get the ‘correct’ result or otherwise act differently when they start taking the pill, which would bias the results in the opposite direction; and a double-blind trial wouldn’t really be possible either.

So, I would class the experiment as pilot research that would be worth further investigation. The statistics do seem to show something, this might be an estrus cycle, but as you point out, the researchers can’t be sure they’ve eliminated every other possibility.

Finally, the main claim of the paper is not demonstrated by the data, viz., that there is an estrus cycle in women. What it can claim to show is that there is a difference in pill-taking and non-pill-taking women’s lap-dancing abilities (although given the scant statistics, it can’t even necessarily claim that. We’re also not presented with individual data, so it might merely be the case that the majority of these effects are due to one or two individuals who are high takers.) In order to demonstrate estrus, they would have to show a significant difference between peak and non-peak days in non-pill taking women that wasn’t the result of menstruation-related problems. This would be pretty hard to do, methinks, and at least from inspection the data doesn’t even bear out the idea that there IS a difference.

All we can get out of this, MAYBE, is that the pill depresses libido. But that claim has been made before, and with much better evidence.