Expect more warped pre-election polls as media supports Obama

You will see more and more of this as we get closer to the fall election. You might as well forget about quality polling for the time being, not that it was all that good anyway. In this case, NBC and the Wall Street Journal figure you don’t care about the sample break-down between Democrats and Republicans.

The media love polls. They are “easy” for the audience to listen to and the so-called “facts” are so easy to “digest.” They are straight forward, with generally four pieces of information; two names and two numbers. But the facts are much more complicated, as polls can be “massaged” to change results to the point where they are absolutely meaningless.

An example. If I head off to an elementary school and ask 100 first-graders the following question, what do you think the results would be? For your birthday, would you like to have a Princess Party or a Bounce House Party? My report: 85% of first-graders would prefer to have a Princess Party for their birthday this year.

This is an extreme example, but I could easily get that result by increasing the number of girls in the sample. If the normal boy to girl ratio in the United States is 51 to 49, all I’d need to do was to use a sample including 90 girls and 10 boys to get my result. Again, this is an extreme example of poll manipulation to make a point, but the media and pollsters do this on a smaller scale all the time, but nobody complains … except for the alternative media. From AP at Hot Air, with my emphasis in bold.

The 2008 national exit poll sample, taken when Hopenchange fever was at its zenith, was 39D/32R/29I, or D+7. This one, after three years of Obamanomics dreck, is somehow D+11 if you include leaners and D+12(!) if you don’t. Anyone feel like taking these results seriously?

So in-other-words, this poll assumes 4 percent more people are more in line with Democrats right now as compared to Nov. 2008. Right, sure, I believe that!

AP is pointing out this headline from MSNBC last night. NBC/WSJ poll: Negative campaign takes toll on candidates; Obama up six points. So in other words, they totally screwed with the sample to get a result that fits their agenda (Obama up), then created a reason – out of thin air –  for the Obama bump that the masses would understand … “negative” campaigns.

Of course, “negative campaigning” has been the standard operating procedure since the first election in the United States, and probably forever, but the media keeps telling us this is something new. Every decade, they tell is it’s new. They treat you like idiots.

Don’t believe anything to see in the media when it comes to polls. Instead, start looking at the actual poll and the sample. Registered voters? Likely registered voters? Just adults? Lean conservative? Lean liberal? Democrat? Republican? Libertarian? Put their feet to the fire and demand quality polling that represents reality if they are going to do it at all.

15 replies
  1. Dimsdale
    Dimsdale says:

    It is soundbite reporting at its worst: make up a poll, sample as you wish, interpret the results as you see fit with a leading headline.? The average schmuck sees a “winner” and a “loser”, and that influences their opinion.? BAM!? Move on to the 20 minute tribute to Sherman Helmsley.? Mission accomplished.?
    Unfortunately, the real loser is the economy and the country, but I guess that isn’t news anymore.

  2. Benjamin Less
    Benjamin Less says:

    SteveM, thanks so much for clarifying the Poll-Dance that we’re been witnessing since 2010.? When the Republicans took over House that whole poll-data during the process was obscured.? Then we witnessed Gov. Walker with his recall vote and blatant misrepresentation of “polling data.”? Geez, I remember Chris Matthews announcing on TV that Tom Barrett was the winner over Walker at 21h00 during the recount.
    Anyway, I will pay little heed to anything I hear and only half of what I see.? You brought up great points on searching the sources.

  3. sammy22
    sammy22 says:

    Perhaps the issue is not with the poll sample selection and question selection, but with the interpretation and spin given to the results. There are polls taken all the time for all kinds of reasons. There is even a daily poll on this blog.

    • Dimsdale
      Dimsdale says:

      I disagree: the poll sample is integral to the validity of the poll.? I mean, if they polled all Dems, the results would be skewed left, and vice versa if all Reps.? These polls are taken specifically to report the alleged mood of the country with regard to the upcoming election.? A skewed poll is worse than useless, it is unethical to the point of criminal.

    • Steve M
      Steve M says:

      Completely disagree. These polls are represented as a non-biased poll sample and it’s total hogwash because nobody in the past has questioned methodology. A majority of the visitors to this site would lean to the right. If I asked the “who do you support” question in one of our polls, of course my results would be skewed, but I wouldn’t claim Romney is up by 12 points in a headline (if our results showed that). ?Heck, I don’t even collect data on party association or political viewpoint, ours is a “not scientific results” kind of one-question hit, where they are passing their’s off as a scientific poll. Totally different.

  4. JBS
    JBS says:

    If the LSM is so blatantly biased and so slavishly lazy, why even continue the charade of having done the work to take a poll? Just bend whatever topical “news” there is and make it fit the moment.
    I have never been called for my answers to any poll. I have never had a canvasser stop at my house and ask about a poll. No one has asked me at anytime what I think about politics, the economy, etc for use in a poll.
    I see these polls as hack Revisionist Goebellization, telling a lie over and over, insisting it is the truth and eventually a lot of people will accept the lie as truth.
    It would be good for the conservatives to begin publishing their own polls and surveys:
    57% say they will vote for Romney
    61% think Democrats are wrong for the country
    86% of 50 and older will vote Republican
    59% feel Obama has had his chance
    You read it here first!

  5. sammy22
    sammy22 says:

    I would think that an NBC/Wall Street Journal poll would be neither right nor left as far as sampling and questioning goes.

    • Steve M
      Steve M says:

      So you’re just ignoring the fact an exit poll from Nov. 2008?(CNN) measured the lean D+7, and now (a different group) using a sample that’s D+11 or D+12? OK…

    • Dimsdale
      Dimsdale says:

      Now you know why it is important to examine the poll internals and questios asked. ?You frequently find that the headline doesn’t actually match the poll results. ? But how many people get past the headline? ?Answer: not many, and that is what they are counting on.

  6. ricbee
    ricbee says:

    As Rush pointed out today polls are only manipulated news stories in today’s world. There is so little public integrity left it’s heart breaking. Only “internal” polls taken by campaigns are believable but not released.

  7. JBS
    JBS says:

    Media integrity has been plummeting for a long time. The media — plural, most of them — have sold to the left and are simply trumpeting the views of the Progressives. In the follow the money scenario, they are just keeping their jobs, self interest first, truth, veracity, and journalistic pride is gone. Journalists are pandering to their bosses and currying favor at every turn. Their god is airtime, bylines, credit and ultimately, salaries. Carrying water for the Regime is what they do.
    That’s why I don’t watch their “entertainment.” They are story readers, news-shapers.

Comments are closed.