I work with real data too. That's why I know about selection biases and similar issues. Most statistics are junk.
The VAERS system only covers 1% of actual cases, this is well known, not contrived. Most doctors don't even know what it is. Any statistics you try to draw from VAERS is going to slanted towards those who know about and use the system.
It also doesn't provide a full picture of anything you look at. If one batch is 10 times the size of the other, and the amount of adverse effects is 10 times the size of the other, then both batches are just as deadly and have the same percentage of adverse effects, although one of them is going to look 10 times worse, when it isn't.
Since the batch data is woefully incomplete in a variety of aspects, I have the luxury to tell you any conclusions drawn with it is pure fantasy.
I recommend reading "How to lie with statistics" and "Statistics done wrong".
I work with real data too. That's why I know about selection biases and similar issues. Most statistics are junk.
The VAERS system only covers 1% of actual cases, this is well known, not contrived. Most doctors don't even know what it is. Any statistics you try to draw from VAERS is going to slanted towards those who know about and use the system.
It also doesn't provide a full picture of anything you look at. If one batch is 10 times the size of the other, and the amount of adverse effects is 10 times the size of the other, then both batches are just as deadly and have the same percentage of adverse effects, although one of them is going to look 10 times worse, when it isn't.
Since the batch data is woefully incomplete in a variety of aspects, I have the luxury to tell you any conclusions drawn with it is pure fantasy.
I recommend reading "How to lie with statistics" and "Statistics done wrong".