Why Bother Doing the Poll

The joint polling effort of ABC News and the Washington Post released a national poll of voters regarding the election choice of Obama against Romney. One has to wonder why they bothered to do the poll. Let me explain.

Jay Cost did a detailed analysis of this poll in an article in the Weekly Standard. The poll shows Obama up by 7% over Romney, but the composition of the group polled was skewed strongly towards the Democrats. The sample used was 34% Democrats and only 23% Republicans with the rest Independents. In 2008 when the Democrats pulled many people to the polls who normally did not vote and when the GOP voters were depressed, the voters still self identified as Democrats by just 7% more than Republicans. In 2010, voters self identified by about 1% more Republicans than Democrats. Do not get confused; the numbers I just listed are not statements about which party the people voted for, but rather they are responses to the question as to which party the person considered himself to be a member.

There is no reason to believe that the composition of the electorate in 2012 will duplicate the extreme skew towards the Democrats that existed in 2008. The GOP is much more energized and the novelty of the first African American president no longer will draw folks to the polls to the same extent. Even were the electorate to stay as it was in 2008, however, the ABC poll is still substantially more Democrat than that. In other words, the ABC/Washington Post poll rigged the outcome just by the sample it chose. Normally, voters of one party support that party's nominee overwhelmingly. The winner is chosen by the Independents. As a result, if you count Democrats to a much greater extent than Republicans, the poll will inevitably favor a win by the Democrat.

Are the Obama folks so desperate for some good news that they are now having their friends in the media rig polls so that Obama will look like a winner? How sad.