Are So Called Reputable Pollsters Cooking Their Numbers

In the last election we were manipulated by the main stream media, concerning Obama. I'm wondering if that manipulation isn't on going (stupid question)  in other areas we are  unaware of in the battle for hearts and minds AND votes.

I'm going to focus on one poll reporting site-Real Clear Politics.

Most all of the MSM are citing Real Clear Politics as an authority on election polling data, A) that should be our first tip off. NBC ABC CBS are not our friends, they are our enemies. FOX plays for both sides.

Take the California senate race, between Boxer and Fiorina. If Boxer is reported as leading in the polls week after week, at some point some of the non Boxer likely voters might decide to throw in the towel, and stay home election day.

Peal Clear Politics takes a group of polling firms, adds their gross "leading by" percents together and averages the total, then declares Boxer ahead by X percent for the week..

Using such a method  seems odd. First of all, there is no weighting of surveys. ABC takes a survey of 500 likely voters and Rasmusenn samples 1500 likely voters. A  +9  ABC and a -4   Rasmusenn result is averaged to give Boxer a +2.5 That's not right. The same 2000 likely voters if grouped together might reveal a completely different weighted  result. It would likely show a result in favor of Fiorina.

Another thing I've noticed, RCP is not consistently using the same polling compaines. One week it's three compaines X, Y, and Z. The next week X, A, and R.

Sometimes polling data from current weeks are combined with data from two or three weeks past.

Some races RCP averages 3 polling firms, other races 5 polling firms. No consistinceny whatsoever.

Pennsylvania is cropping up from what was once considered a safe race for Republican Toomey, to a race that has suddenly turned t very close.  Sestak's late surge,can only be all credited to Obama's  recent visits. Since when does radioactive Obama suddenly change the tide so dramtically? It sure didn't happen in Massachusetts, New Jersey or Virginia.

RCP had Toomey leading handily from the primary, suddenly not so. Let's see Morning Call 9/23 +7 Toomey, Morning Call 10/4 +7 Toomey, Morning Call Tracking 10/19 +3 Sestak. Huh?? What happened to Morning Call?

RCP used PPP (D) in this race for a while....PPP(D). 6/21 Tie, PPP(D) 8/16 Toomey +9 Suddenly PPP(D) gets dropped by Real Clear Politics like a hot rock. Two months later they are picked back up. PPP(D) 10/18 Sestak +1

Some of us are aware polls can be skewed by the questions, and how they are asked or even the order they are asked etc. but not everyone is, and that's a problem.

Polls do change that's not the question, but picking and chosing which data a firm uses in it's reporting is akin to stacking the deck. Instead of reporting the results, they become an arm of the party they favor. RCP asks rather "What would you like the results to say?"

Manipulating the masses comes in many forms. Some more subtle than others.

Steve