There They Go Again – NYT / CBS / QPac Poll Undersamples GOP By Double Digits
![nyt[1]](http://virginiavirtucon.files.wordpress.com/2012/09/nyt1.jpeg?w=530&h=343&crop=1)
Another day, another Virginia presidential poll undersampling Republicans by double-digits. This time, it is the NY Times / CBS / Quinnipiac poll that, like yesterday’s WashPo poll, predicts a meager 24% GOP turnout in Virginia in November.
I’ll keep posting this until these morons get it through their thick skulls. Here is the partisan turnout in the last three Virginia statewide elections:
2006 – 39% Republican, 36% Democrats, and 26% Independents
2008 – 33% Republicans, 39% Democrats, and 27% Independents
2009 – 37% Republicans, 33% Democrats, and 30% Independents
That gives us an average of:
36.3% Republicans, 36.0% Democrats, and 27.6% Independents
Not even in the Democrat wave year of 2008 did Republicans drop below 33% and the rolling average for the last 3 statewide elections in Virginia pegs GOP turnout at an average of 36.3%. Republican turnout in IL and CA in 2008 was even higher than 24%.
This is the last straw and I’m going to call all these news organizations and polling companies out and suggest a drastic course of action that may very well rock them to the core.
Other than gross incompetence, the only other reason these media organizations have for putting out these polls showing Obama ahead in VA based upon partisan samples that so vastly diverge from the past 3 statewide elections or current party ID estimates is they are attempting to suppress Republican turnout so Obama can win Virginia again.
We have their numbers in print via their internals and we have the cold hard numbers from previous statewide elections. The evidence of manipulation is crystal clear and there can only be one motivation.
What happens when Republicans attempt to suppress the vote? They get sued or brought up on charges of violating the Voting Rights Act.
Don’t be surprised if a legal complaint against the WashPo, NY Times and CBS for attempted voter suppression is brought by some Virginia voters — particularly Republican minority voters. Perhaps these media organizations have forgotten, but the First Amendment isn’t absolute. I have a feeling they are about to rediscover that in a most ugly way.
10 Responses to “There They Go Again – NYT / CBS / QPac Poll Undersamples GOP By Double Digits”
Where’s David Axelrod when you need him?
The ends justify the means.
Not Larry Sabato argues that the Virginia electorate won’t look like the statewide elections, but more like 2008:http://notlarrysabato.typepad.com/doh/2012/09/the-changing-electorate-in-virginia.html
I note he doesn’t address 2010′s electorate.
2008 had 33% GOP, 9% more than what these polls are indicating.
Over estimating Democrats in the polls now will make it easier / more believable when they steal the election in November.
THAT scares me.
Well, the best thing to do (as I’m sure everyone here will) is not to let the polls get you down, show up on election day and VOTE!!!
I’m trying to educate myself on your theory that the CBS/NYT/Quin. poll is so bad. Looks like the poll questioned 1474 likely voters in VA (a statistically signif. sample size). Random people were called. Of those 1474 people who took the poll, when asked for their ‘party identification’, only 24% seld-ID’d as Repub, 35% as Dem, & 36% as Indep (& 6% as other/didn’t know).
Help me understand why this is some left-wing conspiracy to understate Repub support for Romney. Doesn’t the poll report what is reports? The poll asked 1474 randoms, and of those polled, 24% self-ID’d as Repub. What these suggest to me is that fewer people in VA are self-ID’ing as Repub & Dem (esp. fewer as Repub), with a third of likely VA voters are saying they are Independent.
Further, what incentive does a 3rd party pollster have to skew polls? Before you jump all over this, I readily concede that either party has an incentive to conduct a poll in a way that leads to more favorable results for their candidate, then to release their party poll as evidence of momentum for their guy. But for 3rd party polls like CBS/NYT/Quin., the pollster would be moved to be as objective as possible due to at two things: (1) the pollster’s reputation, and (2) the public disclosure of the poll’s details. Any possible short-term gain realized by a maverick 3rd party pollster (to skew a poll in favor of candidate #1), would be dwarfed by the pollster’s long-term desire to be credible in all future campaigns.
You assert that these polls are flawed, and it’ll be proven on election day. We don’t know now what’ll be on Election Day 2012, but we cold look back on recent elections. Have you looked back on polls by CBS/NYT/Quin., lets’ say from 2008 or 2004, and actually seen how their pre-election polls were clearly wrong compared to the actual election day turnout? Just sayin.
Legitimate pollsters know the approximate partisan ID of the electorate and then run screening questions to determine if the person they are polling is needed for the sample or whether they already have enough from that particular party / gender / ethnicity / etc. for purposes of the poll. You don’t randomly poll to determine the partisan makeup of the election. Polling is a mathematical science. I know enough pollsters and have had in depth discussions with them to know the basic fundamentals of what makes a good poll and what makes a junk poll.
If you want to see which polls have a history of being flawed and which have a history of being accurate, you need only scroll to the bottom of this site and look in the left-hand column to see it both for the nation in 2008 and Virginia in 2009.
Points well-taken. I did look at the linked chart of previous polls proven reliable or not. But it seems for every poll deemed reliable, there are numerous studies/articles/criticisms (by the other party, presumably) that say “no, actually, the so-&-so poll touted by party A is very flawed….”. We’ll have to agree to disagree for now, but I was surprised by the wealth of resources & opinions on this whole sub-issue of which polls are good & which ones stink.
Question– In the recent CBS/NYT/Quin. poll for VA in which the 1474 Virginians polled were comprised of 26% who self-ID’d as “Repub”, how can you KNOW whether the pollster simply went through 1474 randoms & it turned out 26% self-ID’d as Repub’s, -or- whether the pollster asked screening quesitons to inflate the % of self-ID’s Dem’s (&/or lower the % of self-ID’s Repub’s??? At this point, it seems you’re simply asserting that a truly objective random poll of 1474 Virginias WOULD HAVE a higher % of self-identifying Repub’s, apparently based on prior election exit polls of actual voters self-identiying as Repub’s, and therefore this poll must be flawed b sloppiness or by design?
Comments are closed.