1,352
edits
Changes
Jump to navigation
Jump to search
Line 65:
Line 65: + +
Talk:United States Presidential Election, 2008 (view source)
Revision as of 16:04, July 16, 2008
, 16:04, July 16, 2008→Polling
I've simplified the polling numbers to the two largest polling firms. The previous information from Real Politics was misleading as it did not weight poll results based upon sample size. In other words if one poll of 3,000 people found the spread to be 3 points and another poll of 1,000 people found the spread to be 9 points, rather than merging them to be 4,000 people and a spread of 4.5 points, they simply treat them as equal and make it 6 points -- poor math and unworthy of our site. The "poll or polls" didn't have any numbers recorded at all, and therefore is not included. [[User:Learn together|Learn together]] 12:01, 16 July 2008 (EDT)
I've simplified the polling numbers to the two largest polling firms. The previous information from Real Politics was misleading as it did not weight poll results based upon sample size. In other words if one poll of 3,000 people found the spread to be 3 points and another poll of 1,000 people found the spread to be 9 points, rather than merging them to be 4,000 people and a spread of 4.5 points, they simply treat them as equal and make it 6 points -- poor math and unworthy of our site. The "poll or polls" didn't have any numbers recorded at all, and therefore is not included. [[User:Learn together|Learn together]] 12:01, 16 July 2008 (EDT)
:To put your mind to rest then, I'd be happy to add a table using RCP's results that show the most recent polls, as well as results, MoE, and polling size. Of course, your numbers for Rasmussen didn't explain that you included "Leaners", whereas most polls when publishing their main results will include those separately. So perhaps you'd like to change the number to not reflect the leaners? --[[User:Jareddr|Jareddr]] 12:04, 16 July 2008 (EDT)