Thanks for that analysis, the ways these studies can be and are corrupted is vast. It takes a lot of time and effort to look through them and find the flaws/fraud. I don't enjoy anything to do with numbers and I'm not good at it so I'm delighted to find your work.
Thank you. I had separate probability and statistics classes (the math major track classes) going on 40 years ago now. The probability class has been useful as an actuary, and I remembered a significant amount of it. I don't remember much of the stats class, and haven't been motivated to brush up on it.
I will play around with that calculator and get familiar with it.
I took a look at the calculator, and it makes this pretty easy.
Since I took the math track prob and stats classes, we were doing the mathematical derivations of those formulas. There would be very few problems related to actually calculating what the actual confidence intervals and p values were.
Worse, since this was the mid 80s, it was assumed we using pretty basic calculators, not calculators with stats functions, and no access to computers.
I've just spent a month skilling up on Odds Ratio stats, because I saw a problem/inconsistencies in a vax study. Generally we don't get enough data to check other tests and have to take their word for it, but the OR can usually be checked.
I usually have a problem retaining statistical memory because stats are seldom derived from first principles, but I have a very intuitive feel for them. I've spent a lot of time experimenting in stats - doing repeated trials, making distributions and testing outcomes - so I can usually feel when something's wrong, but have to go out of my way to work out why.
Thanks for that analysis, the ways these studies can be and are corrupted is vast. It takes a lot of time and effort to look through them and find the flaws/fraud. I don't enjoy anything to do with numbers and I'm not good at it so I'm delighted to find your work.
Total Vaxd: 530,000
Tested positive to POTS? Y: 155, N: (530,000 - 155) = 529,845
Odds of testing positive = 155/529,845
----
Total Unvaxd: 811,000
Tested positive to POTS? Y: 43, N: (811,000 - 43) = 810,957
Odds of testing positive = 43/810,597
----
Odds Ratio (Vaxd compared to Unvaxd) = (155/529,845) / (43/810,597) = 5.52
Confidence Interval (95%) in excel language...
= exp (ln(5.52) -/+ 1.96 * (1/155 + 1/529,845 + 1/43 + 1/810597)^.5)
= (3.93, 7.93)
That has a p value of <0.0001
----
Little online calculator here: https://www.medcalc.org/calc/odds_ratio.php
----
The issue would be controlling for everything else. The vaxd probably had a million other vaxes.
Thank you. I had separate probability and statistics classes (the math major track classes) going on 40 years ago now. The probability class has been useful as an actuary, and I remembered a significant amount of it. I don't remember much of the stats class, and haven't been motivated to brush up on it.
I will play around with that calculator and get familiar with it.
I took a look at the calculator, and it makes this pretty easy.
Since I took the math track prob and stats classes, we were doing the mathematical derivations of those formulas. There would be very few problems related to actually calculating what the actual confidence intervals and p values were.
Worse, since this was the mid 80s, it was assumed we using pretty basic calculators, not calculators with stats functions, and no access to computers.
I've just spent a month skilling up on Odds Ratio stats, because I saw a problem/inconsistencies in a vax study. Generally we don't get enough data to check other tests and have to take their word for it, but the OR can usually be checked.
I usually have a problem retaining statistical memory because stats are seldom derived from first principles, but I have a very intuitive feel for them. I've spent a lot of time experimenting in stats - doing repeated trials, making distributions and testing outcomes - so I can usually feel when something's wrong, but have to go out of my way to work out why.