Oct 4, 2022·edited Oct 4, 2022Liked by Mathew Crawford
Hi Mathew,
Way beyond my knowledge of stats, but both a good indication of how clever liars can be (and as a predatory species, I am guessing there is a high correlation between 'clever' and 'capacity / willingness to lie), and the necessity of finding an even rarer breed, those capable of seeing through the dust thrown in our eyes, and with the moral autonomy to call it out.
Though no substitute for continuing education, I was just wondering if you have a favorite 'How to lie with statistics' book for the mathematically challenged?
" And then I proved that it is impossible to trigger a PRR safety signal for some AE in the aggregate without it existing in at least one subgroup. I'll leave that as an exercise for the mathophiles in the audience except to say that there are multiple solutions and I think the easiest one uses Jensen's Inequality."
Not to dis your math skills but that sounds like a common sense problem. If we have ten batches of cookies we need chocolate chips in some individual batches for the whole group to have some chips. It won't ding my mathophile ego to learn there's a more complex element at play. :~)
I think Steve mentioned in his post last night that the variables should be capitalized, because they are in the table.
And I thought, "Maybe it is an accident, or maybe, just maybr, the lower case varmaybe, have another definition, thst they are a subset of the capitalized variables." I'd put nothing past the charlatans. Of course, I am getting old and my last mathematics course was almost 35 years ago in college, so I could be wrong. Of course, I am willing to admit I am wrong. And I'm not a murdering psychopath, so, by definition, I don't work for the CDC
That's why I am creating a synthetic placebo group. I figure out the prr's for all descriptors and assign vaccine specific weights to the descriptors in the reports of the reference groups.
Vaccine Safety Signal Analysis: Does the Ecological Fallacy Apply?
Apologies for the several edits. My first writing was lazy.
Maybe this is not a math problem?
Maybe it is a psychopath problem?
numbers don't lie, but it's very important to understand what exactly they are saying...
Hi Mathew,
Way beyond my knowledge of stats, but both a good indication of how clever liars can be (and as a predatory species, I am guessing there is a high correlation between 'clever' and 'capacity / willingness to lie), and the necessity of finding an even rarer breed, those capable of seeing through the dust thrown in our eyes, and with the moral autonomy to call it out.
Though no substitute for continuing education, I was just wondering if you have a favorite 'How to lie with statistics' book for the mathematically challenged?
Cheers from Japan,
— another, more financially challenged, Steve 🤣
" And then I proved that it is impossible to trigger a PRR safety signal for some AE in the aggregate without it existing in at least one subgroup. I'll leave that as an exercise for the mathophiles in the audience except to say that there are multiple solutions and I think the easiest one uses Jensen's Inequality."
Not to dis your math skills but that sounds like a common sense problem. If we have ten batches of cookies we need chocolate chips in some individual batches for the whole group to have some chips. It won't ding my mathophile ego to learn there's a more complex element at play. :~)
Cool 😎 I loved being taken through your work and thought process.
I think Steve mentioned in his post last night that the variables should be capitalized, because they are in the table.
And I thought, "Maybe it is an accident, or maybe, just maybr, the lower case varmaybe, have another definition, thst they are a subset of the capitalized variables." I'd put nothing past the charlatans. Of course, I am getting old and my last mathematics course was almost 35 years ago in college, so I could be wrong. Of course, I am willing to admit I am wrong. And I'm not a murdering psychopath, so, by definition, I don't work for the CDC
That's why I am creating a synthetic placebo group. I figure out the prr's for all descriptors and assign vaccine specific weights to the descriptors in the reports of the reference groups.
See here: https://help.pervaers.com
"Simpson's Paradox"....
Slowly I turn...step by step...inch by inch
In 2021, US aggregate mortality was only up 0.21% from 2020.
However, 85+ y.o. mortality was DOWN 14% in 2021 from 2020 and
working age mortality was up 18% from 2020
The 2021 US working age mortality was up a massive 30% from 2019.
Can you use these figures somehow, Crawford?