Fun With Polling

Polling is a tricky thing.

The average person may think it is as simple as finding a question and asking it.  It’s not and that’s why most reputable pollsters have a combination of education and/or experience that give them the background to do the job right.  Most people would be surprised at how easy it is a change the wording of a question in subtle ways to get an answer you want.  It’s why every now and then you see polling results that purport to represent the feelings of the electorate that may make you shake your head in wonder.

Interest groups realize this and that’s why they will often work in conjunction with a pollster to craft questions that they feel will have a reasonble likelihood of giving them the answer they want. You simply throw in options in the question that you know people like to gin up support for your issue.  I tend to call this “PR polling.”

For example – Let’s say you were behind an effort to raise gas taxes by 25 percent.  Now common sense tells you that most people would oppose that.  So you don’t ask the question “Would you support raising gas taxes by 25 percent to build more roads”  You don’t need a poll to know the answer to that one.  Instead you conjure up something like this:

“Would support raising gas taxes 25 percent if it would reduce your commute by 75 percent and allow you to spend more time with your family?”

I suspect the results on that question would be quite different.

Which brings me to the recent poll conducted by Margaret Kenski and High Ground and just released by the AZ GOP today.  The press release purports to say that Arizona voters support increasing taxes.

Here is the question asked by Kenski, who by the way is a very reputable pollster.

“I’d like to ask you how acceptable or unacceptable you find the following measures which would be considered in addition to spending cuts in order to preserve critical education and public health funding while the economy begins to recover. These taxes would be in the form of a constitutionally limited tax that would last for no more than three years and would automatically be eliminated without another public vote.

Let’s assume for the purposes of these questions that the Legislature and the Governor have already cut nearly $1 billion dollars in State spending by 2010, used all of the federal stimulus funds to balance the budget and still have a billion dollar per year budget deficit. All of the funds raised from these taxes would be dedicated towards maintaining 2010 spending levels for K-12 education, universities, community colleges, and health care for the poor. Each of the following measures would raise approximately a billion dollars per year.”

I’ve highlighted some of the “sweeteners” included in the question to ensure the answer they want.  First we have “critical”  This is in there for one reason – to scare people.  It’s a word that when used before education and public health implies an emergency situation.  Not, mind you an emergency situation in government terms, but an emergency situation in real terms – IE if we don’t do this then our schools will close and people will die.

Second, we have “while the economy begins to recover.” this at first blush may seem extraneous to the question, but it’s not.  It’s there to say “hey things will get better soon so this wont be so bad.” because people don’t like higher taxes in tough economic times.

Finally, we have the all time favorite term to use when you want to gin up support for something – “education” and the “poor”.  After all who wouldn’t want to support our kids and those less fortunate than us.  So these terms are used in the question.  Now I am not saying they shouldn’t be used for descriptive purposes, but ask yourself, why didn’t the question just say “education and healthcare” instead of “K-12 education, universities, community colleges, and healthcare for the poor.”  The answer is simple, by listing all of these things you are attempting to touch in some way the respondent.  They may not have a child in school, but maybe they care a lot about a community college or a university.  Same thing with the healthcare question.

Now if you really wanted to find out what the citizens of Arizona feel about how to deal with the budget situation, you would probably ask the following and you would probably get a much different answer.

“Some people say that we need to raise taxes to balance our state budget so that we can limit budget cuts for government services like education and healthcare, others say that we should reduce government spending to balance the state budget.

“which position is closer to your view?”

I suspect we would get much different results.


  1. A person in my office privy to the survey says that the your sample question was asked in a similar form. Results were the same.

  2. Think Independent says

    What about specificity? Wouldn’t your argument be the same if they had gone with “education and healthcare?” Wouldn’t those have been too broad?

  3. Jerry,

    Yah and my friend said that their poll said something else.
    What your comment and the paragraph above have in common is they are nothing more than useless supposition

  4. Unless Coughlin wants to get on and post another response, we will have to work with what we have. No survey is perfect. Not my foray but I have never heard a negative thing about Margaret Kinski.

  5. Question: How many households actually make $250,000 a year?

  6. Frequently, small business owners create sole proprietorships. Sole proprietorshipss mean that business revenue is considered personal income of the owner and is taxed at the owner’s personal tax rate. Basically, if a small business owner is only earning $40,000 a year from their small business, but the business is earning $300,000, they would be taxed at a $340,000 level, even though they are not actually using the $300,000 for their household. That means that there are plenty of households that technically make $250,000 or more a year, but are not earning $250,000 a year.

Leave a Reply