[rexy]

from their website
With pure probability samples, with 100 percent response rates, it is possible to calculate the probability that the sampling error (but not other sources of error) is not greater than some number. With a pure probability sample of 2,747 adults one could say with a 95 percent probability that the overall results have a sampling error of +/-2 percentage points. However that does not take other sources of error into account. This online survey is not based on a probability sample and therefore no theoretical sampling error can be calculated.

50% means pretty much nothing without the error interval posted. depending on the survey statistics the difference between a measurement of 47% and 50% is negleable. just remember the % is an average based on the survey of a small selection out of a large population, as such the result from the small subset is invariable incorrect.(but become more accurate as you increase your test set)

they indicate a variance of 2 and a 95% accuracy. they dont prove their 95% accuracy and given the sampling size that's questionable at best even.also they perform an online survey. which means their subset is likely not to be a true crosssection from american society, and as such is the margin of error larger even.

note i dont question the validity of 50%, which i am sure they have measured. however based on their test there is no proof to claim the difference in measurements as a significant bounce, because the chance of it actually being better then before is just as likely as it being worse.