Trick Polls: Obnoxious, Unfair, and Wrong


Public Policy Polling, a Democratic outfit that specializes in asking Republicans questions specifically designed to allow the mainstream media to mock their ignorance, and smug progressives to puff their little pigeon chests up with pride, really hit the public relations jackpot with a recent survey indicating that 30% of Republican primary voters would support bombing Agrabah, which is where Aladdin hangs out in the Disney movie. In other words, it’s not a real place. (I know: all my Republican acquaintances are exclaiming, “Wait, you mean Aladdin isn’t real?) This gave a real chuckle to the left-wing websites and blogs, the  mainstream media and all those brilliant news anchors who don’t know what to say unless a teleprompter lays it out for them, and who believe people who look to them for enlightenment are smart.

Not taking this lying down, a conservative polling groups called WPA Research devised another deceptive poll that revealed that 44% of Democrats would support taking refugees from Agrabah.  So there.

Now conservatives can puff up their pigeon chests, I guess.

Message to pollsters: I know we’re talking about stupid and ignorant people here, but even they will eventually figure out that a certain percentage of poll questions aren’t honest, but are tricks designed to prove they are dumb, violent, stupid, greedy, bigoted, mean, or likely Trump supporters. The joke is on WPA and PPP: people trusted them, as they have traditionally trusted pollsters to be seeking useful opinion data, not proof of knee-jerk partisan idiocy. If a pollster asked about an individual, issue or nation,  those polled never dreamed that the question was setting them up to be scorned. Now both of these organizations have proclaimed that they can’t be trusted, that they aren’t neutral truth-seekers but adversaries with an agenda.

Good to know.

I could analyze why this phenomenon is becoming more common—James O’Keefe, YouTube, the vicious polarization that increasingly causes differing views to be treated as The Great Undeniable Truth vs. The Babbling of Evil Fools. I don’t care what spawned it. The practice is wrong in every way.

I’m giving the polling community an undisclosed amount of time to deal with this phenomenon by revising their Codes of Conduct to include specific prohibitions against this kind of toxic polling, which I would define as “Polling that secretly aims to make those polled appear ignorant by using deceitful questions implying that something is true when it is not.” I’m sure the industry can come up with a more concise term. Both the Code of Professional Conduct of the American Association for Public Opinion Research (AAPOR) and  the Code of  Standards of the Council of American Survey Research Organizations (CASRO) have general provisions (“transparency;” “respect for participants”) that could be interpreted as banning this practice, but they are too vague to have any impact. If the profession won’t police itself, nobody should trust any pollsters, or any polling organization. When we are polled, we currently don’t know whether we are volunteers in a valid fact-finding project, or targets of a political or ideological agenda.

Pollsters can go poll themselves. At this point, trusting to them is proof enough of ignorance.


Pointer: The Blaze

11 thoughts on “Trick Polls: Obnoxious, Unfair, and Wrong

  1. I wonder if any of the options in the answers to the question in either poll was: “You’ve got to be kidding; Agrabah doesn’t exist,” instead of “yes,” “no,” or “uncertain.”

    • Oh no, that might tip them off.

      There are plenty of ways polls show that a substantial chunk of the public just repeats talking points, follows the crowd, doesn’t know what its talking about and has no rational basis for strongly held opinions. They don’t have to be tricked to find that out.

  2. Sometimes I believe that H. L. Mencken was right. Then again, it is definitely unethical to fool people in polls and then generalizing from the results to make members of a political party look ridiculous.

  3. One thing that should be done is that there should be a repository where polling plans get filed before the poll is carried out. That way, at least, if you run phony poll after phony poll, it becomes obvious that that is your business and that you’re only publishing the one in ten that gets the result you wanted all along.

  4. Statistics, especially politically based polls, are hokum.

    I’ve said it before an I’ll say it again…

    You can use statistics to “prove” just about anything, it just depends on the subjects chosen to participate in the survey, how the survey is conducted, and how the data is compiled; all you have to do is use your imagination.

    Consider the following:
    A statistical analysis was done to find out what cat turds taste like, and the results showed that cat turds tastes like chicken.

    Just because a statistical analysis has been conducted and a conclusion reached does NOT mean that what “The Survey Says” actually represents reality.

  5. I think the issue is that the pols ans purpose of the polls are set by the ones who paid. The professional sounding neutrality hides the biased self confirmation goals of the sponsor. To try to force a specific answer, many questions have NO good answer.

  6. “To try to force a specific answer, many questions have NO good answer.”

    Many polls, similar to many psychological or educational evaluation tests have no “good” answers. The main reason for the failure of these lies in devising a test which calls for opinion (informed or not) in the first place: essentially, that is, measuring people’s subjective answers in order to formulate objective statistics. Formulas are then applied to the statistics to “correct” for the statistical results so that they appear to be such -and-such percent “accurate” reflections of percentages of the people who responded.

    “You can’t do that! It’s not a valid question,” said a grad student in psychometrics (virtually, testing the tests) to the professor who placed the question, Who do you love more, your father or your mother? near the beginning of a two-hour final. “It’s not valid; there are too many varia…”.

    “Of course I can,” interrupted the prof. “It’s your choice whether you want to check one box or the other. Please return to your seat.”

    …. [long story short] When the test-takers reached the half-way point in the exam, one by one they discovered that the last hour’s worth of time would be given over to an essay on why that question had been removed from children’s psychological evaluation tests back in 1955. When the exams were graded, there was a notation on the posted list that extra credit had been given to those who had stated — verbally or indelibly on paper — their objection to the question in the first place.

    I’m sure any pollsters coming out of that class would have been able to construct a valid, objective, trustworthy political questionnaire. Unfortunately, I think they all switched courses of study to electrical engineering or cruise ship food management. I’ve never yet seen a questionnaire that was free of impossible, unanswerable questions; the ones online are invalidated for any blanks with no margins to scribble in. As Nero Wolfe would say, “PFUI!”.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.