THE ENTIRE THREAD IS BASED ON A POLL. WHAT ARE YOU DOING IN THE DISCUSSION?
Pointing out once again that polls are completely meaningless.
The problem with polls like this is that they do not follow the requirements of statistical mathematics.
* The raw data MUST be available. The raw data MUST be free of biasing influences. Failure to publish unbiased raw data invalidates the summary.
* A summary MUST be accompanied by the margin of error, or the summary is completely meaningless.
* Data used in the summary MUST be selected from the raw data by randN. Any other method renders the summary meaningless.
* Data MUST be normalized by paired randR. Failure to do so incorporates out of band data, and renders the summary meaningless.
* The variance MUST be declared and justified. Failure to do so prevents calculation of the margin of error value, which is required.
* Statistical math is incapable of prediction, normally inherent in mathematics, due to it's use of random numbers. Even two summaries, performed over the SAME raw data will produce different results.
News polls do NONE of these things. They simply announce some value (and possibly something that looks like a margin of error value) as if it were some kind of decree.
Statistical math has it's uses, but prediction isn't one of them. Simply specifying a summary without showing the data is just making stuff up. Pollsters make a lot of money out of making stuff up. Someone is paying them to do it. The stuff they make up will tend to bend toward whoever is paying the bill. They are not a crystal ball or a set of holy entrails. They cannot predict the future. They cannot show a 'trend' so popular in the Fake News and government today.
I do not even consider ANY raw data as possibly useful without some pretty specific requirements. Too many people spew numbers as 'facts' these days. Among those requirements are:
* The source of the raw data must be known. I must know who collected it, when it was collected, how it was collected, and the purpose of collecting it.
* The raw data MUST be published. The collection method MUST be published.
* If instrumentation was used, I must know what instrumentation it was, its tolerances, when it was last calibrated and how, and how the instrument is maintained and used.
My standards of accepting numbers from people is considerably higher than the average Joe here. They are not unreasonable. I recommend more people adopt these standards at the very least. It would do much to shut out lies and people making shit up.