|
One of the explanations pundits have given for the Ohio exit polls being so far off is that African Americans were over-sampled. Please read the following analysis so you know why that is pure bunk.
In the original exit poll, among White Men (40% of survey), Bush gets 53% and Kerry gets 47%. Among White Women (45% of survey), Bush gets 53% and Kerry gets 47%. So among these subgroups, the percentages are the same.
Now, on the corrected data, now posted on cnn.com, suddenly we find that among White Men (40% of survey), Bush gets 56% to Kerry's 43%. And among White Women (46% of survey), Bush gets 55% to Kerry's 45%.
Chuck Todd, respected journalist, stated during the election coverage that the original exit poll had over-sampled African-Americans. Whether or not the original exit polls over-sampled African-Americans has no relevance whatsoever to this unexplained shift toward Bush among White Women and White Men. These numbers clearly look fudged.
If cnn.com had simply corrected their exit poll to show a smaller number of African-Americans, that would be one thing, but shifting the percentages of Whites who voted for Bush and Kerry is something completely different.
Additionally, I distinctly remember that in the original exit poll Bush garnered 12% of the African American vote. Now, suddenly, it's 16% in this new exit poll.
|