You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Now out: In a preregistered study of 26k online diary entries, we found evidence for ovulatory changes in sexual desire and self-perceived attractiveness only for women who weren't on the pill. Little else replicated. http://psycnet.apa.org/psycinfo/2018-41799-001/ … OA version: https://psyarxiv.com/jp2ym/ pic.twitter.com/9Hr3hNCNIz
This was work with the wonderful Katharina Schilling, @TanjaMGerlach and @LarsPenke. The first preregistration we did in Goettingen (in 2014)! We learned a lot since then. If you find yourself regretting preregistered decisions, maybe our experience can be of help.
I think this was the first preregistered test of ovulatory changes and the field has reacted mostly well to the evidence that we, @julia_juenger and @Ben_C_J presented that some previously reported effects do not replicate in larger samples and less researcher flexibility.
However, this hasn't always trickled down into public perception. Just last week @haselton talked about studies on ovulatory changes in voting and in mate preferences on @SamHarrisOrg as if the nonreplications never happened.https://twitter.com/SamHarrisOrg/status/1031571915939278848 …
The study of ovulatory changes is very interesting both to inform evolutionary approaches to human psychology, but also because the pill is such a common medication – and it flattens ovulatory changes, as the graph shows.
I hope we can learn to predict individual differences in the pill's side effects by measuring ovulatory changes within subjects. To this end, I've partnered with @clue, a cycle tracking app with millions of users.
Because our work (& @julia_juenger's & @Ben_C_J's & other recent work) contradicts predictions by dominant theories in the lit, the work continues.
Our project spawned another, larger diary study by us (analysing right now) and a multisite effort by @Ben_C_J which you can join!
I should also note that the mean differences between hormonal contraceptive users and ovulating women in my graph above should not be interpreted causally. RCTs on sexual desire tend to find a decrease on HC on AVERAGE (individual differences may exist):https://academic.oup.com/jcem/article/101/11/4046/2764940 …
Getting a few responses saying most women know this about themselves already. True! The changes that replicate are the intuitive ones. But we also showed that some supposed causes of differences (e.g., partner attractiveness) don't hold. Harder to have intuitions about this.
In our 2nd study (not this one), we gave individual cycle feedback. Some participants told me it made them question intuitions about their cycle. It's easier to notice effects of (pre-)menstruation without actively tracking your fertility & not everyone experiences large changes.pic.twitter.com/SnNI2LCs1i
In summary, my mom predicted many of these results quite well, but some results in the literature needed replicating and to understand individual differences and disentangle effects of HC from concurrent changes (e.g., in relationship status), we need this quantification.
Almost forgot: If you want to tell me about problems with this study (conceptual, methodological, statistical, me being a nincompoop), but would prefer not to tell me directly, I would still love to hear about it, so I can learn:https://tellmeimwrong.formr.org/
Here's what I meant by "Little else replicated": No evidence for changes in style of dress, partner mate retention (~jealousy) and no evidence that women who find their partners sexually unattractive show stronger ovulatory increases in extra-pair desire.pic.twitter.com/WM6YJGUTfp
@danengber alerted me that there is room for misinterpreting one of our graphs. In the by-item plot, we standardised separately for HC users and cycling women to zoom in on diffs in cyclical _changes_. Not stdising by group leads to the 2nd graph. Details:
https://rubenarslan.github.io/ovulatory_shifts/3_stan_brms_long2.html …pic.twitter.com/bx2i7xwKZC
And Steve Gangestad wrote to me saying that we had undersold the evidence for moderation in our own data. We think he's wrong, but in case others have the same concerns, here's my write-up https://rubenarslan.github.io/ovulatory_shifts/3_fertility_robustness2.html … It's long, but I swear it's only 10% of the email that he sent :-)
Briefly, calculating p values differently than we did, one model in one robustness check would show significant support for moderation. I think this is not robust evidence and we reported it truthfully. Still, our data cannot rule out all moderation, happy to repeat this.
We have collected more data and are planning a collaborative project to collect even more to get narrow confidence intervals even for those pesky dual three-way interactions. Write me to know more and join!
Steve Gangestad has pointed out to me that I test two different questions here. He's right – I rushed to run these analyses and didn't think much about it. I've edited the website to clarify. My conclusions what the pattern of results mean are unaltered though (as are his).
He has also requested the data on further moderators from me (I've shared parts of it with him a while back). So, I guess we'll see whether he can squeeze more (replicable) insight out of the data than we did.
Layman here. Am I to takeaway from this study that females taking oral contraceptives experience both less sexual drive and a sense of less desirability (by their own claims)? Or have I totally missed the mark?
We did not randomise women to the method of birth control which is the gold standard for making such causal claims https://twitter.com/rubenarslan/status/1034408764961710081 … Results from randomised studies are consistent with lower desire on _average_, but individuals experiences may well differ.
I don't know a randomised study looking at attractiveness, but if we're right about the mechanism, I'd expect similar. There is one finding a small negative _average_ effect on well-being. http://www.fertstert.org/article/S0015-0282(17)30247-9/abstract … I think we need more research to give individuals good advice.
It may well happen that the benefits of a comparably hassle-free, effective contraceptive outweigh negative side effects for many women, even given these results on psychological side effects. Stories likely differ.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
