It's self reported. N=41 and even in the self reporting they give that men that were in a relationship had few interactions with people of the opposite sex.

Twenty-seven participants reported being single at the start of the study

Given the rarity of DSI occurrences in partnered men, and the expectation that mate attraction efforts are more important for single men, we tested whether DSI moderated the relationship between testosterone and courtship efforts specifically in single participants.

So only the data of 27 persons were relevant for most of the conclusions of the study.

It was a study of one month:

we collected daily measures of salivary testosterone for one month (31 days), as well as self-reports of sexual desire and other states or events relevant to mating effort on days corresponding to the hormone measures.

It looks like not terribly significant a study. Don't read too much into it.

[–] [email protected] 2 points 2 weeks ago (2 children)

Almost every single study I see has a sample size that is too damn small. It's such a chronic problem.

[–] TheSambassador 4 points 2 weeks ago (1 children)

People see a sample size of 30 and assume it's instantly a bad study. A small sample size can still have considerable power depending on the experimental design and assuming that the sample is actually random across the target population. At the very least, it's information that can be used to guide the experimental design of future studies.

[–] sinceasdf 2 points 2 weeks ago

For sure, I think the research is more relevant to other research professionals rather than us plebs on social media though. Posting it here is kinda puts it out of context I think