3 ms·
N=49 Interesting results, but I'd take them with a grain of sugar until they get replicated several times over. Especially since they match my bias that sugar=
by esperent 6y ago
N=49
Interesting results, but I'd take them with a grain of sugar until they get replicated several times over. Especially since they match my bias that sugar=bad.
- chrisweekly 6y agowordplay on the idiom "take it with a grain of salt" ^ for the sake of non-native speakers I appreciate what you did there.
- disgruntledphd2 6y agoIt's actually even worse than that. I don't have access to the paper, but they had 8 conditions (3 sugars + placebo * fasting|non fasting). That's basically six people per condition. I assume that they did some kind of repeated measures design (although the abstract doesn't say), but the numbers reported are most likely to be statistical noise, especially without preregistration or standard analytical methods.