-
Show this thread
-
Replying to @eigenrobot
Looks like they tested 10 different foods for change in mortality to 3 different chronic diseases, across 4 different types of demographics (2 sexes, 6 age groups, 3 education levels, 4 races) . If I'm counting right, that's 4320 chances to get a positive result. Impressive!
3 replies 4 retweets 75 likes -
-
-
Replying to @ArturBuchhorn
Okay, so each one of those tests has a chance of looking like something interesting happened, even if there's nothing there. The study is set up so that if a 1 in 20 coincidence happens, they'll report it. By sheer weight of trials, we can reasonably expect 216 1 in 20 to happen!
1 reply 0 retweets 2 likes
That is to say, about 216 of the relations between health and diet found by the study are there by chance. So we can't say anything from this study because it's reported that if you roll a die a thousand times it comes up six on some of them.
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.