A sample size of 11 males? With over 14 choices? Actually codifying 15 different possible behaviours from which an "interactions" score is determined by resear

A sample size of 11 males? With over 14 choices? Actually codifying 15 different possible behaviours from which an "interactions" score is determined by researchers? not blind, let alone double blind? It reminds me of the chocolate paper [1] Come on! If you actually open the paper, https://michaelfergusonphd.files.wordpress.com/2016/08/sex-differences-in-rhesus-monkey-toy-preferences-parallel-those-of-children.pdf https://michaelfergusonphd.files.wordpress.com/2016/08/sex-d... You will see that the "total interactions" with boy toys is the same for male and female. You might say: No, the women's one is lower. Well, that's what error bars are for. Error bars tell you: Be honest, you actually don't know. So for all intents and purposes they are equal. It's also surprising that they are comparing humans and monkeys, but the don't use the same metric. The paper for humans uses "Mean seconds of play-time", but the study for monkeys had to come up with a new metric "total interactions" to achieve a similar looking graph. Why? For the plush toy, the female result was 1.5 minutes, ... but the standard deviation was 3.8 minutes ... so ... what does that tell us? It tells me, design a better experiment! Here's a better methodology: Take 2 "gendered" toys with accelerometers and take 200 monkeys one-by-one, with trackers. Then automatically measure (in a fully blind study) how long each monkey spends which each type of toy presented. I'm not saying the effect does not exist. But this paper does not show it does. [1] https://io9.gizmodo.com/i-fooled-millions-into-thinking-chocolate-helps-weight-1707251800 https://io9.gizmodo.com/i-fooled-millions-into-thinking-choc...