Pollster Bias

Pollster Predicament
“… the creators of the survey may deliberately try to influence the results by using questions that are purposefully worded in a biased manner.” – college-course excerpt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The above-quoted excerpt comes from a hand-out pertaining to a statistics course taught at Winona State University in Minnesota, and it refers to deliberate bias in a survey conducted by pollsters.

The entire statement says, “If the objective of a survey study is to support a certain cause, the creators of the survey may deliberately try to influence the results by using questions that are purposefully worded in a biased manner.”

The hand-out cites the following survey question as an example of deliberate bias in a survey question.

The administration of Democrat Barack Obama has still not satisfied congressional and media questions about just what it knew and when it knew it about the terrorist attack on U.S. diplomats in Benghazi, Libya, last September 11. That attack killed four Americans, including the U.S. ambassador to Libya. The Obama administration has changed its explanation of that attack several times since and has so far refused to identify those officials who made key decisions not to send help to stop the attacks, and who decided not to initially call the killings a terrorist attack. Knowing that and anything else you may be aware of about this issue, do you agree or disagree that President Obama should be impeached over his handling of this situation?

The hand-out goes on to say the following:

Given the wording of this question, how much faith do you have in the results of this survey study?

In general, avoid writing loaded or leading questions. You should always make sure your questions are worded neutrally.

Deliberate bias is not always introduced as blatantly as it was in the previous example. Some survey creators with an agenda simply take advantage of the proven fact that human beings, in general, like to be agreeable and are inclined to answer “yes” when a question leads them in that direction.

So, it is possible that a pollster’s survey is deliberately created to produce a particular result. It would not be a mistake for one to keep this fact in mind when one reads about a survey conducted by just one pollster, especially if the pollster has a reputation for having a strong political bias.

By the way, if you want to know which pollster is cited in the above-quoted college-course hand-out, then click here.

Charles Rangel Loses in Court Again
OPEN THREAD: Al Gores Says He Could Become Catholic After Pope Joins Global Warming Band Wagon
  • Brucehenry

    Wenzel Strategies comes to mind, lol.

    • Commander_Chico

      I downloaded the item. You are correct.

      World Nut Daily.

  • Par4Course

    Actually, there are a couple of examples of bias in the materials, one from MSNBC.com. The text teaches that bias can be introduced to skew a poll in whatever direction the pollster wants, right or left. In considering polls, it is particularly important to read beyond the headline, which often doesn’t accurately describe the poll results, let alone the details of who was asked what.

    • Brucehenry

      Yes of course you are absolutely right, as is David.

      As I’ve said before, MSNBC is unwatchable to me. The naked bias embarrasses me as a liberal. I would trust a poll commissioned by MSNBC about as far as one commissioned by WND, which puts me a little ahead on the skepticism scale of Warner, who thinks Wenzel Strategies’ polls are peachy.

      PS I do like Rachel Maddow’s style, however, in 5 minute increments. I don’t believe I’ve ever been able to sit through a whole hour of ANY program on MSNBC.

      • Retired military

        Bruce
        I have to say I am shocked. Pleasantly shocked but shocked.
        There is some hope for you yet. 🙂

  • Hank_M

    The MSM uses biased polls to get the results they want and then they use the results to influence peoples opinions.

    Truth is I’m sick of polls. But I suppose it’s easier than doing actual journalism.

  • WHO’S THE BUSTER

    Well-played David Robertson.

    I was going to guess Rasmussen.

    Hey, is Romney still President after winning in a landslde?

  • jim_m

    It doesn’t matter about the pollster. What matters is if the pollster will give you access to the demographics and the question sets. DJ Drummond used to do some really good dissections of polls by analyzing how they were conducted. That more than anything can demonstrate bias.

    • Brucehenry

      But if I recall correctly, Drummond’s dissections of the polls in ’08 and ’12 proved inaccurate. Correct me if I’m wrong, but I seem to recall him siding with the “unskewed” guy. Maybe I’m misremembering.

      • jim_m

        Polling is an art, not a science. In elections you never see thhe same connsditions twice so accurate polling is more luck than anything else. Social conditions change and people behave in surprising ways.

        Take a look at this article about the UK elections yesterday The experts all missed the Tory landslide. Why? The explanation is as good as any.

        Polling is subject to personal bias since much of it relies upon the ability of the polster to correctly guage the temper of society. If they get that wrong then they miss on their predictions.

  • WHO’S THE BUSTER

    Although Nate Silver is incredibly adept at distilling the results as he is only interested in accuracy.

    That is why it doesn’t matter if it is baseball stats (where he first made his name) or politics or examining market trends for private companies; he knows if he is wrong or demonstrates bias that his business will suffer.

  • yetanotherjohn

    If you had told me 7 years ago that answers I gave to a media poll would be passed on (along with information about me) to political operatives who would in turn mention my name to bureaucrats so that I could come into some grief (e.g. a tax audit), I would have laughed and called you a conspiracy nut. With the ever leftward trend of the media, politicization of the bureaucracies and the belief on the left that those who disagree aren’t just wrong but morally reprehensible to the point of deserving any ill that befalls them, I wouldn’t laugh anymore. I wouldn’t believe you without some further proof, but I wouldn’t laugh.
    I mentioned the polls under performing when sampling for conservatives in the US 2014 election, UK and Israel in the last 6 months to three people today. The universal response was long the lines “of course, why should a conservative answer truthfully in a poll?”

    • WHO’S THE BUSTER

      A tax audit because you replied to queries for a media poll?

      • jim_m

        Before obama it would have been considered highly unlikely that a person would have been targeted for an audit based on political statements. Now it is considered almost a given that if a person is being audited it is for political payback. The long list of democrat tax cheats who go unpunished and conservatives who are unfairly harassed speaks to the believability that the IRS is now a political weapon aimed directly at people who oppose the dems.

        • WHO’S THE BUSTER

          So it is your contention that if you respond to a media poll with the “wrong” answers that you are at a higher risk for an audit?

          Oh boy, the sky is falling again.

          • yetanotherjohn

            I didn’t say the if you respond to a media poll with the “right” answer that you are a a higher risk for an audit. What I said was that it is no longer unthinkable that such could happen. It is quite clear that the IRS has audited, employed extra scrutiny and delayed applications for organizations because of their political positions. Since the left wants so much to increase government, perhaps they should consider how abusing government functions for political aims undercuts their cries to increase the size of government.

          • jim_m

            I didnn’t say that it was my contention. I said that such a contention wasn’t completely unbelievable.