Puna-i was addressing what started this thread, and the discussion about statistics.
Q had stated that 90% on this forum want CCW.
To successfully demonstrate what 90% on this forum want, well 90% would have to vote for CCW. However, currently only 105/2403=4% not even close to 90% on this forum. Understand now?
What I understand now is that you have confirmed my suspicion that you have absolutely no understanding of statistics in general nor polling statistics in particular. Do you really believe for there to be a valid (accurate) poll of the American adult people or American registered voters (for examples) that ALL of them have to be polled? That would not merely be woefully mistaken, but also be an idiotic and moronic belief revealing (miraculously) a state of comprehension even lower than complete and total ignorance about statistics and polling. Please note that I did not call you and idiot or a moron or ignorant. The whole point of polling or surveys is that it is statistically possible to conclude with a reasonable probability what an entire group holds as a view while sampling only a minute percentage of that actual group.
So, according to your belief that to know what a percentage of members of this site likely believe, you have to poll ALL of them, you must then discount and/or invalidate any and all polls that sample microscopic fractions of the group they draw a conclusion about? Correct? Or is it just this poll on this site and not any other polls? If so, why? Please explain.
For instance, a recent Gallup poll (
http://www.gallup.com/poll/168770/half-illinois-connecticut-move-elsewhere.aspx#1) asking adult residents (600 per state) of all 50 states whether they would move to another state if they were able to, claimed that the
margin of sampling error is ±5 percentage points at the 95% confidence level. Using the state of California as an example, the sample polled there of 600 people out of an adult population of 29 million is .0000206% of the total group for which the conclusions of the poll had the margin of sampling error is ±5 percentage points at the 95% confidence level. Compared to the 4.5% sample of the current poll on this site/thread,
the poll here has a sample 218,447 TIMES larger than the Gallup poll.
And the current Fox poll re "How much does Obama lie?" (
http://www.foxnews.com/politics/2014/04/16/fox-news-poll-many-voters-say-obama-lies-to-country-on-important-matters/), which has a
margin of sampling error of plus or minus three percentage points, polled 1,012 registered voters out of the total 137 million registered voters in the U.S., which is .000000738% of the group for which conclusion are drawn with only a three per cent margin of error.
The percentage sample for the poll on this site/thread is 6,428,571 TIMES greater than the Fox poll sample.Is your polling statistical analysis methodology based upon revealed biblical mathematics, or what? Please explain.