Battle of the Polltsters

NATIONAL POLLS Site

The battle for Ohio in this presidential race comes down to pollsters. In a very tight race since the presidential race of 2000, pollsters are doing the heavy fighting; some for political gains, others mainly for business promotions. Only a few days left and the pollsters are hard at work trying to predict and perhaps affect the outcome of the presidential election by concentrating on battlegrounds. In this case, various pollsters put out some interesting polls which demonstrated just how inconsistent the polls have been. Ohio is a battleground that political pundits see as a do-or-die state for the GOP, and one that accurately predicted GOP victories over the past presidential races going back decades. Obama knows just how important Ohio is to the GOP and spent quite a deal of time and cash in Ohio to ensure the state doesn't go for Romney come November 6, 2012. 
The polls are interesting because it shows tremendous efforts to help President Barrack Hussein Obama retains his job. For instance CBS, PPP, CNN, and Gravis Marketing polls gave President Obama 50-51 advantage over Gov Mitt Romney who is not too far behind. The GM is a clear statistical tie as always; a single point advantage is not a sweeping victory. The other polls gave President Obama taking a slight four points lead over his opponent if the differential falls outside of the margin of error. For that reason, it is important to note who/what these orgs are. CNN and CBS are clearly proDemocrat news organizations with an obligation to sway public opinion toward President Obama's way. We've seen how CNN portrays the GOP and Gov Romney over the years with the likes of Piers Morgan and Soledad Obrien, while turning a blind eye on President Obama's abysmal performance. CBS and PPP are clearly Liberal pollsters attempting to bring President Obama up to the surface after being submerged for a straight two years. It is also important to note that Obama's approval rating when he took office was 70%. In just four years, his approval fell substantially by 30 points. He was in the 40s for sometimes.
What is the significance of the 50-51 points? Pundits noted that for decades, no incumbent president ever won reelection with polls below 50%. With all the scandals: the Benghazi coverup; the Fast & Furious gun-walking scandal; the unemployment rate not recovering; the stubborn recovering economy etc., these liberal - leftwing pollsters decided to just give the president a point or two above the competitive mark. This will give the nation the impression that President Obama is ahead and is doing incredibly well. To give him more points - eg. 56% as that of President Bush this time of the year 2004 - would appear extremely unrealistic and biased. 
The only poll that seemed to give Gov Romney the edge over the incumbent is Rasmussen, which is an independent pollster not tying to any networks or political party. When all pollsters sampled registered voters and often times over sampled registered Democrat voters, Rasmussen concentrates on the "likely voters" and "independents". His sample over the year shows a neck to-neck tie in battlegrounds and nationwide. In one of the Liberal polls, Democrat voters were over sampled by 30%, while Independents were under sampled by a single digit. People are smart enough to figure out what a controlled poll is and what a credible one looks like. 
In November 2010, the Democrat leaning polls predicted that the Democrat party would gain a huge majority in the House of Representative. These polls may have motivated the [then] speaker of the House Rep.Nancy Pelosi (D) to declare she would retain her position and that the Democrat party would retain both the Senate and the House of Representative. Off course we all know how it ended. The GOP threw her out of Speakership and replaced her with Rep. John Boehner. 
What is a controlled poll and why putting them out anyway? A controlled poll is one that is obtained through carefully designed questionnaires. Example: a 2008 pollster asked this this question: (paraphrase) - will you vote for Obama or someone not Obama. This is a controlled question because it controls choices. If 50 out of 100 (universe) participants answered "NO" to the question, the result would construed as 50% - not for Obama but not for the other candidate either. Another controlled question: How do you rate Obama's handing of War Against Terror? The options are: 1. Excellent 2. Very good 3. Good 4. Somewhat good. 5. Not sure. The question is driving the universe to decide between 5 choices, all of them give Obama a clear advantage. In other words, the choices don't give the universe the opportunity to rank President Obama's handling of the War Against Terror - Extremely Bad or Poor! It doesn't give the universe the benefit of expressing real feelings. Furthermore, for those who don't agree with President Obama, the only choice is "unsure", which means that 4/5 for and 1/5 against Obama. If the universe is 100, the chances of Obama getting a good result is 80% and the probability of giving Obama a bad choice is either zero because there's no choice, or 20% if the anti-Obama vote settles for "unsure". 
As professor Sabato of University of Virginia said, its all about TURN-OUT. The polls are designed to affect public perception about the parties and their candidates, but if people who never voted and never actively engaged in public political discourses turned out to vote, the winner will crash all predictions.