I'm not certain how you're defining the word "anecdote" in the context of your answer. My statement discourages believing worldviews based on informal observations and "what I remember to be true" rather than formal data collection.
Anecdote is "I'm looking around me and I don't see anyone dying of COVID. I don't know anyone who died of COVID. The people I know who had COVID said it was just a flu."
That's an anecdote. It's just a story you heard. You didn't collect formal data on it. You didn't verify anything. You establish and compare demographics or classify variables. You just are suggesting that you hear a lot of stories, and you assume you'd know a bunch of people who died if COVID was so dangerous. You can't use that data to establish whether or not COVID is actually dangerous, because you don't know anything except your subjective interpretation of what you informally observe.
If you were taking data and formally entering it into a database and running algorithms on that, then you weren't using anecdotes to draw conclusions about the world. You were formally turning anecdotal reporting into formal data and then doing math on it.
It's not a problem just because data comes from an anecdotal source. That's how surveys are collected. It's a problem when you try to suggest a reality about the world based on your informal observations and connections you build based on what YOU notice and what YOU happen to remember. That's not useful scientifically.
The problem is in your fundamental assumptions about VAERS, which again, go against what VAERS itself says about its own data.
Yes, the data is underreported, but that doesn't suggest that the underreported data is filled with legitimate vaccine injuries. That is an assumption you make because of your already-established conclusion that they're trying to hide something.
This is what it's saying:
VAERS wants a LOT of data. It literally wants to know about every single medical problem that happens within a few months of a vaccination for EVERY SINGLE vaccinated person in the country. Every case of the sniffles. Every headache. Every tooth that falls out of a child's head (just in case). Every time someone has to update their glasses. Every time someone gets food poisoning.
EVERYTHING.
This is why everyone is allowed to submit a report, and why doctors are required to submit a report. Because if a 98 year old alcoholic with liver cancer dies three weeks after a vaccine, VAERS wants to know about it. JUST IN CASE.
The reason that the data is considered underreported is because VAERS does not get all of this data. In fact, when you consider that it wants to know every time you have an allergy attack to pollen randomly after getting the vaccine, then yes, it absolutely makes sense it's only getting 1% of the data it wants.
Because it wants to know fucking everything. JUST IN CASE. Because that's the point of VAERS. It's just a tip line so that researchers can establish a baseline, and the more of these random aches and pains they can record, the more stable and robust that baseline will be.
There is nothing in VAERS to suggest that the underreported data consists of even a single vaccine injury. VAERS cannot confirm it's ever registered a single vaccine injury. It asks for data that it knows can be 100% comprised of unrelated coincidences, and it still wants that data. Because that's what VAERS is for.
Your fundamental assumption about the data you're using and what it represents (and therefore, how you can work with it) is incorrect, and this confirmed by VAERS itself.
If you think you can build an empirical picture of confirmed vaccine injuries from data that is incapable and unwilling to take any measure to confirm that ANY vaccine injuries exist within its dataset, then you are trying to do magic, not math.
I absolutely don't underestimate what you can do with the tiniest fragment of data. I say this in the absolute most objective way that people who tend toward believing in conspiracy theories tend to be people who can establish patterns from miniscule amounts of data. They can see narratives from scatterplots.
I am definitely one of those people, in fact, despite not following you down this particular rabbit hole.
The problem is, of course, that those patterns are often subjective, and sometimes patterns can be drawn from the random and routine.
Such people also tend to have masterfully beautiful and complicated interpretations of the shapes of clouds, but that doesn't necessarily mean that the cloud is arranged in that particular shape for any particular reason.
I don't reject the complexity of the tapestry that you believe supports your beliefs. I just don't currently believe in the robustness of many of the threads with which the tapestry is woven. I am willing to be convinced, but you tend to want to stay in the more abstract top-level conversation rather than establish whether most of the facts you're using are actually as factual as you believe.
I'm not certain how you're defining the word "anecdote" in the context of your answer. My statement discourages believing worldviews based on informal observations and "what I remember to be true" rather than formal data collection.
Anecdote is "I'm looking around me and I don't see anyone dying of COVID. I don't know anyone who died of COVID. The people I know who had COVID said it was just a flu."
That's an anecdote. It's just a story you heard. You didn't collect formal data on it. You didn't verify anything. You establish and compare demographics or classify variables. You just are suggesting that you hear a lot of stories, and you assume you'd know a bunch of people who died if COVID was so dangerous. You can't use that data to establish whether or not COVID is actually dangerous, because you don't know anything except your subjective interpretation of what you informally observe.
If you were taking data and formally entering it into a database and running algorithms on that, then you weren't using anecdotes to draw conclusions about the world. You were formally turning anecdotal reporting into formal data and then doing math on it.
It's not a problem just because data comes from an anecdotal source. That's how surveys are collected. It's a problem when you try to suggest a reality about the world based on your informal observations and connections you build based on what YOU notice and what YOU happen to remember. That's not useful scientifically.
The problem is in your fundamental assumptions about VAERS, which again, go against what VAERS itself says about its own data.
Yes, the data is underreported, but that doesn't suggest that the underreported data is filled with legitimate vaccine injuries. That is an assumption you make because of your already-established conclusion that they're trying to hide something.
This is what it's saying:
VAERS wants a LOT of data. It literally wants to know about every single medical problem that happens within a few months of a vaccination for EVERY SINGLE vaccinated person in the country. Every case of the sniffles. Every headache. Every tooth that falls out of a child's head (just in case). Every time someone has to update their glasses. Every time someone gets food poisoning.
EVERYTHING.
This is why everyone is allowed to submit a report, and why doctors are required to submit a report. Because if a 98 year old alcoholic with liver cancer dies three weeks after a vaccine, VAERS wants to know about it. JUST IN CASE.
The reason that the data is considered underreported is because VAERS does not get all of this data. In fact, when you consider that it wants to know every time you have an allergy attack to pollen randomly after getting the vaccine, then yes, it absolutely makes sense it's only getting 1% of the data it wants.
Because it wants to know fucking everything. JUST IN CASE. Because that's the point of VAERS. It's just a tip line so that researchers can establish a baseline, and the more of these random aches and pains they can record, the more stable and robust that baseline will be.
There is nothing in VAERS to suggest that the underreported data consists of even a single vaccine injury. VAERS cannot confirm it's ever registered a single vaccine injury. It asks for data that it knows can be 100% comprised of unrelated coincidences, and it still wants that data. Because that's what VAERS is for.
Your fundamental assumption about the data you're using and what it represents (and therefore, how you can work with it) is incorrect, and this confirmed by VAERS itself.
If you think you can build an empirical picture of confirmed vaccine injuries from data that is incapable and unwilling to take any measure to confirm that ANY vaccine injuries exist within its dataset, then you are trying to do magic, not math.
I absolutely don't underestimate what you can do with the tiniest fragment of data. I say this in the absolute most objective way that people who tend toward believing in conspiracy theories tend to be people who can establish patterns from miniscule amounts of data. They can see narratives from scatterplots.
I am definitely one of those people, in fact, despite not following you down this particular rabbit hole.
The problem is, of course, that those patterns are often subjective, and sometimes patterns can be drawn from the random and routine.
Such people also tend to have masterfully beautiful and complicated interpretations of the shapes of clouds, but that doesn't necessarily mean that the cloud is arranged in that particular shape for any particular reason.
I don't reject the complexity of the tapestry that you believe supports your beliefs. I just don't currently believe in the robustness of many of the threads with which the tapestry is woven. I am willing to be convinced, but you tend to want to stay in the more abstract top-level conversation rather than establish whether most of the facts you're using are actually as factual as you believe.