Poll: It’s how you ask the question

In case you have not heard, the State of Minnesota is heading for a seemingly inevitable state shutdown if state leaders cannot agree on a budget before July 1.  Anecdotally, much of the discussion seems to have shifted from budget priorities to assigning blame for the pending cease of operations.  And, I am seeing more public opinion polls that ask some variant of “Who is to blame?” or “Should we raise taxes or cut spending?” 

First, polls can be subject to several sampling and statistical errors that can lead to bias, whether or not it is intended.  Worse, however, is that some polls are framed to get intended answers.  I came across this link via a Minnesota State Chamber of Commerce tweet (@MCC_THesse).  A KSTP poll that, shows that 87 percent of Minnesotans do not favor increased spending.  The Chamber translates this message as no new taxes.  That is fair insofar as the question was phrased.  “Going forward, should Minnesota’s government increase spending? Decrease spending? Or continue to spend about the same amount as it has been?”

Unfortunately, in a 5-10 minute telephone poll, the questioner, who probably is not a policy wonk, cannot explain what an increase really means.  Does “increased spending” include inflation adjustments and cost-inflators like health care and energy?  Does increase spending include federal stimulus funds that were used in the last legislative session but are now unavailable?  I surmise that there are very few folks (13 percent according to this survey) that actually favor increased spending relative to the previous session.  But, there is not a consensus on what an increase truly means.  I wrote a piece at the beginning of the sesson that explained my take on When an Increase may not be an increase.  Others may have been more eloquent on the topic.  But, it should explain how these questions are not quite as obvious as “Should the government increase spending?”  I would not expect persons who do not work in government to be aware of these nuances.  Therefore, I don’t expect such a poll to provide much validity.

Finally, other polls show seemingly contradictory results.  This Star-Tribune poll from May showed that 63 percent favord a mix of tax hikes and spending cuts to balance the budget.  How can both polls be right?

The answer, of course, is how you ask the question.  I firmly believe that most citizens do not favor new program spending.  However, the state has less money today than two years ago in terms of real revenue, inflation-adjusted revenue, and federal revenue.  As such, the state will require “more spending” (or more taxes) just to provide the same services as the previous biennium. 

So, after digging deeper, the polls are not contradictory at all.  But, they also do not tell us very much we do not already know and largely agree upon.   Therefore, I will continue to view with much skepticism any a public opinion poll that purports to overwhelmingly support one position or the other in terms of taxes and spending.