So the new Public Policy Polling has Ohio tied at 45% to 45% with 10% undecided. This poll also has McCain winning 18% of the African-American vote and winning 17% of the Democratic vote. Last month's Public Policy Polling, however, had McCain winning only 6% of the African-American vote and had McCain winning the white vote by a margin of 4%. This latest poll shows McCain winning the white vote by 11%. So, here's the question: how did McCain effectively triple his African-American vote and more than double his lead among white Ohioans?
Well, the answer may have less to do with McCain ads and more to do with the people being polled. This is from the PPP media release on the poll:
PPP surveyed 950 likely voters from August 12th to 14th. The survey’s margin of error is +/-3.2%. Other factors, such as refusal to be interviewed and weighting, may introduce additional error that is more difficult to quantify.
Obviously the people being polled this time are not the same as the people being polled last time. So the question is which poll is off? Was it the July poll, which had McCain getting only 6% of the African-American vote and Obama getting 42% of the white vote? Or was it the August poll, which has McCain getting 18% of the African-American vote and Obama only getting 39% of the white vote? Well, if we knew the answer to that question, we would be in the polling business.
Of course, the best answer may be not to worry about polls at all. In David Broder's column on Sunday in the Washington Post, an Obama staffer said that right now he doesn't worry about polls, he worries about increasing the number of registered voters. He said that increasing the number of registered voters is how they are going to win this election in places like Ohio, Florida, and other battleground states.