@eadu4 wrote:
@pitdakota And again I respect your reply. Agreed operator errors occur. I'm not saying every area reporting had massive errors, but it would only take each area to have a number of errors to add up to a significant number. I understand you have faith that the system worked as it should. I don't. I feel there were more errors than we will ever know about. And I don't have faith that it would ever be reported. Since inconsistencies and errors are rarely reported or they are simply a footnote correction. You have more faith in the system than I do. I respect that but I can't do that. That's a leap of faith I just can't take anymore. :smileyh
___________________________________________________
@eadu4, and there we have it. "That's a leap of faith I just can't take anymore". I respect that is your opinion. Problem is you try to back that opinion with all kinds of discussion that is not factual.
First, it was that mortuaries, doctors' offices, clinics, hospitals, etc reported data and it was suspect for corruption coming from all those different places. Well no, mortuaries, doctors' offices, and clinics are not doing testing. They never have. That is not how the testing system in this country works.
Then it is assumptions and every state sets their own requirements for a positive test. Well no, states don't set those requirements. So we finally get to the issue that you just don't have faith in the system, no matter what. Ok.
It isn't about me having faith. It is about the fact that I have done this type of work for years. I have experience with this so I am a little familiar with how it works. There are others here in this discussion that have experience in this area as well. Doesn't mean we know everything, but it does mean we understand how the basics work.
Here is an example I will use about data. Although understand I am a stickler for accurate data.
AAA routinely reports average cost of gasoline for different areas. For a large area, they may routinely survey 30 gas stations and average the cost of the 20 gas stations. One day the person doing the calculation omits a gas station by accident but still divides the total by 20 and reports the average cost. Later, the mistake is found. They re-enter the data to add all 30 gas stations before dividing by 30. But, it only changes the cost of gasoline by 0.03 cents/gallon. Not a major error because all gas stations had the same range of cost and the denominator was a large number so it doesn't distort the information by much.
If on the other hand, there were only 3 gas stations in the mix, so that the cost per gallon for only 2 gas stations is added before dividing by 3, the mistake results in a greater deviation from the true number which results in the cost reported being much higher than it actually is.
Now, think about numbers of tests performed per day in any state that are in the thousands. Some states have tested on a daily basis in the tens of thousands of cases. So it would have to be a really BiG deviation and error to grossly impact the data. And it would also probably stick out like a sore thumb from data obtained any other day, unless there are circumstances such as severe weather closing testing sites, etc.
Now being on the receiving end of that data whether for disease surveillance in public health or doing research, I want that data to be as accurate as humanly possible. But what people are alluding to here on this thread is just not founded in reality.
I get you don't have faith in the system. I just take issue with using talking points that aren't factual to support that opinion.
I hope that makes sense, because my goal is not to be onerous. Although it is hard in a discussion like this to actually get the sincerity across. There are things that no matter what people are just not going to believe. Fact of life. The problem for me comes when misinformation starts becoming a part of a reason to support an opinion, just because.
Blogs & Forums
Forums
Blogs
Since there is no way to accurately prove either side conclusively without knowing the parameters for each and every statistic that was counted. And I can't blindly accept numbers where no one can show me exactly how each state, county, city, town, hospital, clinic, test site, doctor's office, morgue counted each occurance or outcome. There is just too much risk for inaccuracy.
