-
-
Show this thread
-
The study was very simple. Divide men into 2 groups, give half of them nuts, half of them no nuts, see who does better on measures of sexual functionpic.twitter.com/kC3amS4388
Show this thread -
People who ate more nuts reported a statistically significant improvement (p=0.040/0.037) on two metrics of sexual function
Show this thread -
First problem: these differences were statistically significant, but TINY Here are the graphs from the study. You need a magnifying glass to tell which tests would've been significant, because there's barely anything therepic.twitter.com/IIXXWK0wtS
Show this thread -
This gets into the idea of a statistically significant difference, which means that numbers diverge, and a clinically significant difference, which means people should care
Show this thread -
Suffice it to say that if nuts improve your sex life by an amount barely big enough to be measured, most people aren't going to care
Show this thread -
There was another mammoth issue with the study - they used what's called a "per protocol" analysispic.twitter.com/Im9ZA40mOE
Show this thread -
Per protocol means that the researchers ignored dropouts from the intervention and analysed only those who completed the trial This is well-known to overestimate treatment effect and bias study results in randomized controlled trialspic.twitter.com/aFhDlO1u1i
Show this thread -
The problem is, if you ignore dropouts - particularly when they make up >30% of all your study participants as in this case - you lose a lot of data
Show this thread -
Furthermore, we KNOW that dropouts are less likely to achieve the outcome of interest (have better bangs) than people who finish the study
Show this thread -
So it's very likely that the only reason there were any results at all from this study, given how marginal the treatment effect was, is that they used this per protocol analysis methodology
Show this thread -
At the absolute minimum, trials like this usually present per-protocol analyses alongside less misleading results from an intention-to-treat analysispic.twitter.com/olSGRYJQ8b
Show this thread -
Why did the researchers choose such a problematic method to analyse their results? Impossible to know. But it might be to do with the funding bodypic.twitter.com/S28MCBBeq4
Show this thread -
Yes, the study on semen and sex quality was funded by - and I cannot believe I get to say this - Big Nut You cannot make this up, it's too good
Show this thread -
Also worth noting that the presentation in the media of this as a positive trial conflicts with the results, because for the majority of the things the researchers tested nuts made no difference
Show this thread -
I remain unconvinced that eating 60g of nuts a day will do anything whatsoever for your sex life, based on this researchpic.twitter.com/WvBWzZJVAi
Show this thread
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.