Originally posted by northernladyuk
View Post
Yes, you are technically correct (but the point is moot by the end of this post) with regards to causation and StatSig, but I doubt, for example, many people would like a wording change to 'smoking causes cancer' on cigarette packets...due to this same technicality....
"Many other reasons were suggested for the link between lung cancer and smoking, including sleep deprivation or alcoholism. In layman’s terms, it’s now known that smoking causes lung cancer. But in scientific (or statistical) terms, you can’t really say “cause” as that would mean every single person who smoked even just one cigarette would get lung cancer. As statisticians, we say that there is a very strong correlation between smoking and lung cancer."
However, plug in the study figures to the link below and you will arrive at a confidence interval of 4.6-5.4% - NOT really statistically significant is it.....so not even a correlation has been found
https://select-statistics.co.uk/calc...on-proportion/
But you don't really need to use the calculator to work that out, when the sample (11205) is so low (0.03%) compared to the target group (34000000)
So to sum up;
1. Low sample size compared to population size
2. Low confidence interval
3. No correlation can be derived due to 1 & 2
4. Extremely unlikely (but not impossible) to be causation without any correlation
5. Study is yet to be formally presented (and therefore not yet peer-reviewed)
But thanks for linking it anyway, made me chuckle with all those 'remainer' symptoms
Leave a comment: