stefd

4 ms·

1. 1.▲
   by stefd 7y ago

   This is assuming this 0.62% actually shows a TRUE effect. If it is noise/invalid/a mistake, it means nothing; the actual effect might actually be a DECREASE of CVD as far as we know. And it is important to also look at all cause m
2. 2.▲
   by stefd 7y ago

   > They use statistical methods to eliminate the confounding variables. Some of those anyway. You can only account for confounders that you know about a priori. And given they used Food Frequency Questionnaires, it's mostly wasted ef
3. 3.▲
   by stefd 7y ago

   Wondering if this might explain similar spying around the Canadian parliament... https://www.cbc.ca/news/politics/rcmp-csis-cell-surveillance...