“Using 26,000 diary entries to show ovulatory changes in sexual desire and behavior”

Kevin Lewis points us to this research paper by Ruben Arslan, Katharina Schilling, Tanja Gerlach, and Lars Penke, which begins:

Previous research reported ovulatory changes in women’s appearance, mate preferences, extra- and in-pair sexual desire, and behavior, but has been criticized for small sample sizes, inappropriate designs, and undisclosed flexibility in analyses.

Examples of such criticism are here and here.

Arslan et al. continue:

In the present study, we sought to address these criticisms by preregistering our hypotheses and analysis plan and by collecting a large diary sample. We gathered more than 26,000 usable online self-reports in a diary format from 1,043 women, of which 421 were naturally cycling. We inferred the fertile period from menstrual onset reports. We used hormonal contraceptive users as a quasi-control group, as they experience menstruation, but not ovulation.

And:

We found robust evidence supporting previously reported ovulatory increases in extra-pair desire and behavior, in-pair desire, and self-perceived desirability, as well as no unexpected associations. Yet, we did not find predicted effects on partner mate retention behavior, clothing choices, or narcissism. Contrary to some of the earlier literature, partners’ sexual attractiveness did not moderate the cycle shifts. Taken together, the replicability of the existing literature on ovulatory changes was mixed.

I have not looked at this paper in detail, but just speaking generally I like what they’re doing. Instead of gathering one more set of noisy data and going for the quick tabloid win (or, conversely, the so-what failed replication), they designed a study to gather high-quality data with enough granularity to allow estimation of within-person comparisons. That’s what we’ve been talkin bout!

3 thoughts on ““Using 26,000 diary entries to show ovulatory changes in sexual desire and behavior”

  1. This area of research is the mirror image of what is usually discussed on this website, so there is an opportunity to learn something.

    Usually, this website looks at papers with dubious results and the game is to point out the flaws and how the authors were able to tease out a publishable paper from noise. But here we have a paper that is looking for results where we have strong priors that there should be something real here. How could there be no relation between ovulation and sexual behavior in humans? In many animals, the relations are so strong that they are obvious to the naked eye.

    So why might this be an opportunity to learn something? Maybe we can learn something by observing how research in this area differs from research in other areas we have found to be bogus. Then we could use those observed differences as signals to spot bogusness in other areas.

    One thing bothers me though. Why isn’t there a large literature on this topic already? If it is so obvious that are real relationships here, why haven’t studies already been done? Serious, reliable, convincing studies. I should really read the paper to see what the state of the literature is on this topic. I’m going to pretend I haven’t done that because the paper is behind a paywall, but I’m really just too lazy to do the legwork.

  2. Happy to see our paper mentioned positively here – your criticism of some of the literature shaped my thinking about how to work on this topic and how to review the literature.
    A preprinted version without paywall is here: https://psyarxiv.com/jp2ym/
    Dan Engber wrote a nice piece with some context about this literature’s own replication crisis here:
    https://slate.com/technology/2018/10/ovulation-research-women-replication-crisis.html

    By the way, I used Stan (via brms) for the first time for some of the robustness checks in this paper, and I think there are some aspects of the data in this and later work that I wouldn’t have been able to model without Stan and brms. So, in many ways, new tools (massive automated online data collection and new statistical tools) made this work much easier.

Leave a Reply to Ruben Arslan Cancel reply

Your email address will not be published. Required fields are marked *