Do you believe in the polls? What about the tooth fairy?
One of the hot topics today is the polls. Most have Obama pulling away in the toss up states and ahead nationally. The conservative talking heads are expressing disbelief claiming that the polls are run by democrats and are trying to determine the outcome of the election. However, I find it hard to believe that the Wall Street Journal pollsters are a bunch of libs. Nevertheless, from a purely scientific viewpoint, one can call the polls into question. For whatever reason, most of the polls oversample democrats. The polls that don't actually have Romney ahead. Nonetheless, for the polls to be accurate, those who actually vote must reflect the composition of the polls. Thus, if the polls oversample democrats say 40 percent democrat, 30 percent republican and 30 percent independent, then the poll is only accurate if the actual voters keep the same percentages. While this may be likely, the likelihood that oversampling democrats is accurate is dubious at best. Consider the following: since the election in 2008, democrat registration has fallen dramatically. In the swing states consider the following: democrats are down over 400,000 in Ohio, 5 percent in Florida, 9.5 percent in Iowa and 19.7 percent in New Hampshire ( http://www.foxnews.com/politics/2012/09/27/drop-in-ohio-voter-registration-especially-in-dem-strongholds-mirrors/#ixzz27mDHMdtX). The pollsters admit that their models are largely unchanged from 2008 and reflect voting patterns from that election. Consider this, first the fall in democrat registrations should call for a revision of the samples. Second, common sense tells you that Obama will not do as well among any voting bloc save the progressive wing of the democrat party. He has to do worse among Catholics for the administration's stand regarding mandated birth control for employers. He has to do worse among Jews for his anti-Isreal stance and the anti-semitism in the Occupy Wall Street crowd - a democratic favorite. He may still get 95 percent of the black vote but the number of black voters will be significantly less. All this is common sense. For the polls to ignore all of this lessens their value and leads me to conclude that this election won't even be close. Romney in a landslide.
Harold A. Black is professor emeritus in the Department of Finance, University of Tennessee, Knoxville having retired after 24 years of service. He has served on the faculties of American University, Howard University, the University of North Carolina - Chapel Hill and the University of Florida. His government service includes the Office of the Comptroller of the Currency and as a Board Member of the National Credit Union Administration. He also has served on the boards of directors Home Savings of America and its parent company, H. F. Ahmanson & Co., Irwindale, California prior to its merger with Washington Mutual Savings Bank, on the board of New Century Financial Corporation, Irvine, California, then the nation’s largest real estate investment trust and as director and later chairman of the Nashville Branch of the Federal Reserve Bank of Atlanta. He writes an occasional article for the Knoxville News-Sentinel at http://www.knoxnews.com/staff/dr-harold-black/. His web page is haroldablackphd.com