3 ms·
If anyone wants to see an abuse if statistics, this is a fantastic case. Take a look at the conclusions drawn, and how the p values prove insignificant conclus
by throwayEngineer 7y ago
If anyone wants to see an abuse if statistics, this is a fantastic case.
Take a look at the conclusions drawn, and how the p values prove insignificant conclusions. Ie, less than 20kcal/day changes.
You can see that at best there is a small correlation. But it seems like they found numbers to prove their theory.
- tracker1 7y agoI was a bit surprised by the conclusions as well... the abstract chart looks to be close to a 500kcal difference, but the chart is 34kcal difference. a 500kcal difference would account for the weight difference ~7000kcal over 2 weeks, but not 34. About the only really interesting numbers, to me were the insulin levels at around the 120min mark.