Click on graphics to embiggen for easier reading.
The Left Stream Media and Democrat politicians like Harry Reid got a lot of mileage out of a recent CNN poll that showed 47% of Adult Americans have an Unfavorable opinion of the Tea Party movement.
That same CNN poll showed only 32% Favorable to the Tea Party movement, but a poll done by GALLUP just two months ago showed 71% considered Tea Parties ideas either “Somewhat important” or “Very important.”
Did 4 out of 10 Americans really switch their opinion of the Tea Party in just two months?! Not likely.
Now, y’all know how much I love digging into polls! And having dug into this one … methinks CNN got what it paid for. I.e., a biased poll result they could use to sway the swayable.
There’s nothing Dems like better than a sound bite they can blat to the stupid Sheeple who have been voting Democrat for years without having a clue what Democrats actually get up to.
And you know human nature. A hint that some group has started to show signs of growing a Big Red L on its forehead quells enthusiasm around the water cooler Pretty Darn Quick.
These sleazy games have been working for the Left for decades. Why would they change strategies now? … Hmm, lemme think. How about the risk that the Conservative Loud Mouthed Right will use the internet to wise up the Sheeple that they’ve been conned?
And whoa Nelly … Hell hath no fury, especially in America! A young lawyer told me one of the most dangerous court room strategies is for a lawyer to try and scam a jury. Americans HATE being conned! If they get even a whiff of an effort by one side to scam them, the jury’ll switch en masse to favor the other side.
So … on that note, let me explain just exactly how we can all see for ourselves that the CNN poll was a Big Fat Lie.
First, a bit about polling. Most polls are done over the phone. Good polls aim for more than 1,000 randomly selected respondents, because the math wonks have proven that if you get that many, your results can be extrapolated to the entire population polled with a ± 3 Margin of Error (MOE).
So these are key items to look for:
- What population was polled? Both CNN and Gallup polled Americans over 18 years of age who have either land lines or cell phones. These two polls can be compared.
- Did the survey include more than 1,000 respondents who were randomly selected from the target population? Yes and yes.
- Did the survey report an MOE of ± 3? Yes and yes.
It’s not uncommon for news reports about polls to ignore the total respondents and MOE. However, the CNN poll reports I saw highlighted BOTH of these items. In other words, the LSM REALLY wanted the Sheeple to BELIEVE their meme — i.e., the Tea Party is growing a Big Red L.
Except … what nobody but the polling organization acknowledged was that the ± 3 MOE was a fraud. The poll was improperly weighted to give the Left a handy sound bite that would make the Tea Party look bad.
Pollsters have to call people randomly, so they are not going to get an exact match with the population for various affiliations. Thus, they ask the people who don’t hang up on them to not just answer their questions, but also to give personal info like age, gender, race, income level and political affiliation. This information allows the pollster’s math wonks to “weight” the respondents’ opinions so the results will more accurately represent the opinions of the total population.
Weighting sounds ominous, but it’s not. Let’s say a pollster gathers answers on a political issue (like the Tea Party) from more than 1,000 phone calls. And let’s say the people who agreed to give those answers happened to be 50% Democrats and 25% Republicans. The pollster could not report their answers as meaningful for the American population until he first did some math magic to give a lot less weight to the Democrats’ answers and a little more weight to the Republicans’ answers. Done properly, this would get the responses to ± 3 match the current affiliation numbers, which are 29% Democrats and 29% Republicans.
Party affiliation moves around all the time. Gallup tracks it regularly and posts the results on-line. No polling organization has any excuse for not properly weighting the results they get from their respondents … except maybe, “We were broke and CNN offered us a lot of money to throw our ethics in the trash, but hey, we reported the results in our data and posted it on-line, so it’s not like we really lied … much.”
The Tea Party is very much a Right-wing group. Look at how the CNN poll deliberately biased the political weighting and how the polling organization buried the bias in a ream of numbers:
Here’s another way to look at the bias. The top bar graph shows the current Gallup affiliation results with the independent leaners added to the Right and Left for simplicity’s sake. The bottom two bar graphs show the CNN poll results for “Do you have a favorable or unfavorable view of the [Democratic, Republican] Party?”
The purple box visually drops the proper Gallup weight down to the bottom two bar graphs to make them easier to interpret. If the CNN poll had been weighted properly, the purple box would line up closely with the red and blue boxes. Instead of lining up, the purple box demonstrates a pronounced pro-Left, anti-Right bias.
There’s one more thing I’d like to teach you today about polls and that is to pay attention to who the poll purports to have sampled.
A statement like “3 out of 4 doctors recommend Bayer” is completely meaningless. Bayer would obviously like you to fill in “all doctors in the country” but minus a statement of who was sampled, it could easily mean 3 out of the 4 doctors in some golf four-some said, “Sure, I’d recommend Bayer to my patients. It’s as good as any other aspirin out there. Now can we play?”
On-line polls are almost as meaningless, though you can glean some knowledge from them if you keep in mind that they ONLY poll people who come to that website AND answer the poll questions. However, if you’re interested in, say, who Hot Air readers who respond to Hot Air polls are most interested in for the 2012 GOP ticket, then this poll would have some value for you:
Most of the big pollsters — e.g., GALLUP, Pew — claim to poll Adult Americans. Often, this means they called a random selection of people who have land lines or cell phones in the continental U.S., who tell the pollster they are over 18 years of age and who consent to answer all the questions. Given the difficulties of polling, it’s as good a compromise as any and, since all the pollsters work this way, their results (properly math wonked) are interesting and useful.
Occasionally, a pollster will go for Registered Voters, but that’s rare.
My favorite is Rasmussen, who seems to be the only group that frequently polls Likely Voters. I like them because, when it comes to political issues, it’s really only the voters who matter. When looking at poll results, keep in mind not only that (outside of election season), the Left routinely biases polls to make us look bad, but also that even in a good poll, Adult Americans routinely poll more Liberal than do Likely Voters. You can see this clearly in this graph.
As for reporting, keep in mind that the Left Stream Media reports what it wants the Sheeple to see. They mostly ignore Rasmussen or cherry pick data out of polls while ignoring more significant numbers that don’t fit their meme. For example, the reporting on that CNN poll was almost exclusively devoted to the negative Tea Party number. When I googled “unfavorable view of tea party CNN”, I got 152,000 hits in .13 seconds.
The March 11-13, 2011 CNN poll didn’t just ask about the Tea Party. It also asked about the Democratic and Republican Parties.
- Democrats: 46% Favorable v. 48% Unfavorable
- Republicans: 44% Favorable v. 48% Unfavorable
So they made a Big Hairy Deal about 47% Unfavorable Tea Party, while burying the 48% Unfavorable Democrat!
And there was another set of significant data they ignored. Back in November 6-9, 2008 — i.e., the same week Obama and the Democrats swept the elections — CNN got these results to the same questions:
- Democrats: 62% Favorable v. 31% Unfavorable
- Republicans: 38% Favorable v. 54% Unfavorable
Let’s see those lined up another way:
- Nov 2008: Democrats: 62% Favorable v. 31% Unfavorable
- Mar 2011: Democrats: 46% Favorable v. 48% Unfavorable
That’s a SIXTEEN PERCENT shift in the Democrat Favorables! And it all happened during the period when Dems held the majority in the House, the Senate and the Oval. Ouch.
It’s no wonder the O’drooly Left Stream Media didn’t find it “news worthy.” ::snort::
—————
SOURCES:
http://i2.cdn.turner.com/cnn/2011/images/03/29/rel4l.pdf
http://www.gallup.com/poll/145838/Americans-Believe-GOP-Consider-Tea-Party-Ideas.aspx
The Left Stream Media and Democrat politicians like Harry Reid got a lot of mileage out of a recent CNN poll that showed 47% of Adult Americans have an Unfavorable opinion of the Tea Party movement.
That same CNN poll showed only 32% Favorable to the Tea Party movement, but a poll done by GALLUP just two months ago showed 71% considered Tea Parties ideas either “Somewhat important” or “Very important.”
Did 4 out of 10 Americans really switch their opinion of the Tea Party in just two months?!
Not likely.
When there’s a discrepancy THIS big, smart people suspect a DemocRAT.
Now, y’all know how much I love digging into polls! And having dug into this one … methinks CNN got what it paid for. I.e., a biased poll result they could use to sway the swayable.
There’s nothing Dems like better than a sound bite they can blat to the stupid Sheeple who have been voting Democrat for years without having a clue what Democrats actually get up to.
And you know human nature. Nothing quells enthusiasm among Sheeple faster than a hint that some group has started to show signs of growing a Big Red L on its forehead. The reaction around the water cooler is immediate. “Better not act like I like them or their friends or people might think I’m a Loser too!”
Right? And heck … it’s been working for the Left for decades. Why would they change strategies now? Well, let me think. How about finding out they’ve been conned? And too bad the Conservative Loud Mouthed Right now has the internet and polling organizations that helpfully post their reports so we can point out to said Sheeple that the Left lies. A lot.
And whoa Nelly … Hell hath no fury! A young lawyer told me one of the most dangerous court room strategies is to try and scam a jury. Americans HATE being conned! If they get even a whiff of an effort by one side to scam them, they’ll switch en masse to the other side in a heart beat.
So … on that note, let me explain just exactly how we can all see for ourselves that that CNN poll was a Big Fat Lie.
First, a bit about polling. Most polls are done over the phone. Good polls aim for more than 1,000 randomly selected respondents, because the math wonks have proven that if you get that many, your results can be extrapolated to the entire population polled with a ± 3 Margin of Error (MOE).
- So these are key items to look for:
- What population was polled? Both CNN and Gallup polled Americans over 18 years of age who have either land lines or cell phones. These two polls can be compared.
- Did the survey include more than 1,000 respondents who were randomly selected from the target population? Yes and yes.
- Did the survey report an MOE of ± 3? Yes and yes.
It’s not uncommon for news reports about polls to ignore the total respondents and MOE. However, the CNN poll reports I saw highlighted BOTH of these items. In other words, the LSM REALLY wanted the Sheeple to BELIEVE their meme — i.e., the Tea Party is growing a Big Red L.
Except … that ± 3 MOE was only for the TOTAL poll. What nobody but the polling organization reported was that the weighting was deliberately biased to make the Tea Party look bad.
Pollsters have to call people randomly, so they are not going to get an exact match with the population for various affiliations. Thus, they ask the people who don’t hang up on them to answer not just their questions, but also to give personal info like age, gender, race, income level, political affiliation and whatever else is deemed interesting and important. This information allows the pollster’s math wonks to “weight” the respondents so the results better represent the opinions of the total population.
Weighting sounds ominous, but it’s not. Let’s say a pollster gathers answers on a political issue (like the Tea Party) from more than 1,000 phone calls. And let’s say the people who agreed to give those answers happened to be 50% Democrats and 25% Republicans. The pollster could not report their answers as meaningful for the American population until he first weighted those responses so that they better matched current percentages, which at the moment are 29% Democrats and 29% Republicans.
GRAPHIC: 2011_03 29 Latest Gallup Survey of Adult Americans
CSB @ http://news.webshots.com/photo/2189061510056011884SMAFFA
Party affiliation moves around all the time and Gallup tracks it regularly and posts the results on-line. No polling organization has any excuse for not properly weighting the results they get from their respondents … except maybe “we were broke and CNN offered us a lot of money to throw out ethics in the trash, but hey, we reported the results in our data and posted it on-line, so it’s not like we really lied … much.”
The Tea Party is very much a Right-wing group. But look how biased the weighting is. Conservative respondents have an MOE of ± 4.5 while Liberal respondents have an MOE of ± 7.0. (And get a load of all the hits I got when I googled “Unfavorable view of tea party CNN”!)
GRAPHIC: 2011_03 CNN poll – MOE’s show left bias
CSB @ http://news.webshots.com/photo/2866925570056011884XdvOFZ
Here’s another way to look at the bias. The top bar graph shows the current Gallup affiliation results with the independent leaners added to the Right and Left for simplicity’s sake.
The bottom two bar graphs show the CNN poll results for “Do you have a favorable or unfavorable view of the [Democratic, Republican] Party?”
The purple box visually drops the proper Gallup weight down to the bottom two bar graphs to make them easier to interpret. If the CNN poll had been weighted properly, the purple box would line up closely with the red, green and blue boxes. But they don’t and they show the same pro-Left, anti-Right bias.
GRAPHIC: 2011_03 CNN poll improperly weighted to favor Left
CSB @ http://news.webshots.com/photo/2139443770056011884cAkYxg
There’s one more thing I’d like to teach you today about polls and that is to pay attention to who the poll purports to have sampled.
A statement like “3 out of 4 doctors recommend Bayer” is completely meaningless. Without a statement of who was sampled, it could mean “3 out of the 4 doctors in my golf four-some said, “Sure, I’d recommend Bayer to my patients. It’s as good as any other aspirin out there. Now can we play?”
On-line polls are almost as meaningless, though you can glean some knowledge from them if you keep in mind that they are ONLY polling people who came to that website AND answered the poll questions. However, if you’re interested in, say, who Hot Air readers who respond to Hot Air polls are most interested in for the 2012 GOP ticket, then this poll can be useful to you:
GRAPHIC: 2011_04 05 Monthly Hot Air GOP Prez polls
CSB @ http://news.webshots.com/photo/2279031650056011884EfXZcc
Most of the big pollsters, like GALLUP and Pew, poll All Americans over 18 who have phones and consent to talk to pollsters for a few minutes. Occasionally, you’ll see a poll that uses Registered Voters, but that’s rare.
The ones I’m most interested are done by Rasmussen, who seems to be the only group that frequently polls Likely Voters. When it comes to elections, I’m not that interested in what All Americans or Registered Voters think. I want to know how many people are on my side and what do the people I want to switch to my side care about.
The two main things to remember when you see or talk to your targets about poll headlines like the CNN 47% is not only that outside of election season, the Left routinely pays for biased polls and Adult Americans routinely poll more Liberal than Likely Voters. You can see this clearly in this graph.
GRAPHIC: 2011_03 Likely Voters v Adult Americans
CSB @ http://news.webshots.com/photo/2364231020056011884BuDGwW
The Left Stream Media would ignore the Rasmussen poll entirely and report the 47% Approve number from the Pew poll with a headline like, “Majority of Americans approve of President Obama.” Only way down in paragraph umpteen would they maybe admit to the 2% differential between Approve and Disapprove.
With the CNN poll, all the reporting was about that 47% unfavorable Tea Party number. But there were a lot of other numbers in that poll that they could have, but did not report on.
For example, CNN has been asking “Do you have a favorable or unfavorable opinion of the [Democratic, Republican] Party?” since September of 2006.
They ran this poll the same week (November 6-9, 2008) Obama and the Democrats swept the election, the results were:
Democrats: 62% Favorable v. 31% Unfavorable
Republicans: 38% Favorable v. 54% Unfavorable
(BTW, you can see a skew here. Obama won 52.9% v McCain won 45.7%. If the poll had accurately predicted the election, the Democrat Favorable should’ve been 9 points lower and the Republican favorable should’ve been 8 points higher.)
The Left Stream Media did NOT report that the March 11-13, 2011 CNN poll showed the following major shift:
Democrats: 46% Favorable v. 48% Unfavorable
Republicans: 44% Favorable v. 48% Unfavorable
Let’s see those lined up another way:
Nov 2008: Democrats: 62% Favorable v. 31% Unfavorable
Mar 2011: Democrats: 46% Favorable v. 48% Unfavorable
Nov 2008: Republicans: 38% Favorable v. 54% Unfavorable
Mar 2011: Republicans: 44% Favorable v. 48% Unfavorable
That’s a 6% shift from Left to Right, all during the period when Mr. Slow-the-Oceans-Rise was running the planet. Ouch. And let’s not forget that this 6% is on TOP of the 8% bias we saw in Nov 2008. Double ouch.
—————
SOURCES:
http://i2.cdn.turner.com/cnn/2011/images/03/29/rel4l.pdf
http://www.gallup.com/poll/145838/Americans-Believe-GOP-Consider-Tea-Party-Ideas.aspx
polination.blog 