Saturday, August 21, 2010

Futility of Polls

In a now infamous poll from Pew Research Center, 18% of Americans claimed that President Obama is a Muslim. As shocking as those numbers are, it's a good reminder of exactly what opinion polls are, opinions. More than that, they are non-expert opinions. Even more than that, they are non-expert opinions on issues that are mostly opinion. What is Obama's religion? Are in for a double-dip recession? How likely is another major terrorist attack? All of these questions are are pure speculation.

In a democracy, I can understand the desire to know the opinions of the people, but we aren't a democracy, we are a republic. This is an important distinction in understanding sensitive issues like religion or the recent controversy over gay marriage in California. For the purpose of law, it doesn't matter what a majority of Californians voted for, the government is restricted by the Constitution. The courts claimed that limiting marriage in this way violated the right to equal protection and due process. Now, whether those clauses were interpreted correctly is a different matter.

Another final reason why I'm weary of pollsters is how important the wording of question are. For issues like gay rights there are red flag words that can get desired results. For example most Americans oppose same-sex "marriage", but half support "civil unions". Depending on which word is used, results vary significantly. This not only shows bias in pollster agenda, but just how blurry many political lines are. Opinion polls may be valuable for politicians trying to understand their constituents, but voters shouldn't concern themselves with them.


  1. Is Obama's religion speculation?

  2. Well speculation in the sense that you can't prove belief. I assume the poll responders know that he claims to be a Christian, but just don't believe him.

  3. Amike1:36 PM

    More accurately, half are opposed to "marriage" and a majority support "civil unions."

    Polls do provide valuable information, in spite of their inherent weaknesses...the ANES study, for instance, provides political scientists with a heap of important data, simply by asking the same questions (worded the same way) every four years and allowing us to track changes. And of course the fact that the wording of questions *does* make such a difference is itself telling...recognizing that should make us more attuned to political rhetoric and how it works.

    Thought: the Constitution doesn't actually restrict the government, because the government is the agent with the power of enforcement (or non-enforcement)...and it also has the power to change the text whenever it likes (provided it's sufficiently united). What really restricts the government, more than anything, is the willingness of government officials to restrain themselves (see the back half of Federalist 10) and to keep their fellow officials in line (see the front half). It's morality, first and foremost--a shared sense of public duty.

  4. I'm curious Aaron, what "valuable information" do you think they provide?

    I agree the Constitution isn't perfect at restricting government, but I would say it's more because of things like the necessary and proper clause, not politicians. If we had to rely on the will of politicians we'd be in trouble.

  5. Amike1:08 AM

    Then we're in trouble, 'cause we really, really do.

    The classic example is Brown v. Board of Education--which accomplished virtually nothing for a decade, in terms of actual school desegregation, because officials in the South decided to keep on keeping on, Constitution be damned, and federal officials were unwilling to take the necessary action to hold them in line.

    The Constitution's just a piece of paper--it takes people to make it matter.

  6. Great example. Maybe we are in trouble.

    I wonder how much impact the words of the Constitution have on American political preferences.


You are the reason why I do not write privately. I would love to hear your thoughts, whether you agree or not.