How to Skew a Poll

On January 26, Brian Brown announced the results of a survey conducted by NOM in Washington state. He spoke with the grace and clarity befitting a man devoting his life to a moral cause.

Ha! No, just kidding. He lied. Here it is:

If the Washington Legislature wants to change the definition of marriage, which 57% of voters oppose, NOM calls on them to give this decision to voters.

Sorry, but NOM’s poll says that 57% don’t think a law permtting same-sex marriage is necessary, given the state’s civil union law. When it comes to actual opposition, though, only 49% said they would vote against it. (42% would vote for equality, while 12% don’t know.)

Funny thing:  When Brian made his announcement, NOM didn’t release the survey for us to read. Jeremy Hooper over at NOMExposed dinged them for this at the time. As it happens, NOM’s updated press release now links to the document, but Google Cache indicates this wasn’t always the case (though Google Cache might be showing the new version by now).

In other words, Brian Brown distorted the results of his own survey on a day when we couldn’t check the honor of his word. It’s hard to blame him. One of NOM’s consistent talking points is that  a clear majority of voters oppose marriage equality, so any number under 51% embarrasses them.

Still, even the 49-42 sentiment against us is odd. Prior polls found the numbers going 47-46 or even 54-35 in our favor. So how did NOM get its results?

That turns out to be an interesting lesson in “priming.”

You can rig a poll by priming your respondent with a series of leading questions. For instance, this is how the poll introduces the idea of marriage equality:

As you probably know, since 2010 Washington has had a civil union law which gives gay couples all the legal rights of married couples. Now some people want to pass a new law, which changes the definition of marriage, so that it is no longer between a man and woman, but between any two people. Do you feel it is necessary or not necessary to pass now a new law which changes the definition of marriage in this way?

57% said the law is not necessary. This is where Brian gets his “57% of voters oppose” falsehood, even though the question isn’t asking that.

Back to priming, though. Check out a few things. NOM uses “changes the definition of marriage” twice — language they promote as as good PR for their side. Also the long, confusing, syntactically tortured question makes it sound like marriage would be “no longer between a man and a woman.”

Beyond that, though, the question lies when it says the law would allow marriage “between any two people.” They didn’t even bother to say “between any two adults“! Lord, I’d oppose that: I don’t want a 50-year-old woman marrying her 4-year-old neighbor. You can bet this distortion — this outright lie — is meant to trigger the incest and pedophilia fears that anti-gay forces work so hard to implant.

The survey’s next question:

Who do you think should decide what the definition of marriage is in Washington state: should it be defined by the courts, or should it be defined by the state legislature, or should it be defined by the voters of the state?

Most folks who passed civics class understand that the legislature exists to, well, legislate. But with this question, NOM plants the notion that passing a law is not what legislators are supposed to do.

Next up:

If you were able to speak today with your local state legislator, would you tell him or her that passing a new which charges [sic] the definition of marriage is something you want the state legislature to work on at this time, or would you tell him or her that the state legislature should work on solving other problems?

As if they can’t do both. The question sets up a pair of false alternatives, and does so with loaded language.

Finally, we get to the key question:

And if you were able to speak today with your local state legislator, would you tell him or her to vote for this new law which changes the definition of marriage, so that it is no longer between a man and a woman but between any two people, or to vote against this change?

Loaded and dishonest — they might as well have asked:

Can you seriously tell me you think the state legislature should drop everything they’re doing to grab power from the people and pass an unnecessary law that would allow a creepy old man to marry his sweet and trusting grandson?

Priming works. You can’t fool all the people all the time, but partisan pollsters make a science of it. In this case, it’s heavy-handed to the point of hilarity.

Yet with all that, NOM was only able to get 49% of the respondents to oppose same-sex marriage. And it forced Brian Brown to tell what’s clearly a lie while hiding the data that would expose him as a fraud.

In other words, business as usual for NOM.

  • Digg
  • Facebook
  • email
  • Reddit

7 comments to How to Skew a Poll

  • 1
    Mrs. Chili says:

    I’d be surprised if NOM ever called me for another poll again.  They tried twice, and both times I didn’t get past the third question before the recording thanked me for my time and hung up on me.  How much you want to bet that my answers didn’t make it into their data?

  • 2
    tavdy79 says:

    Yes Prime Minister (the British political sitcom) covers precisely this topic. Enjoy!

  • 3
    clayton says:

    How can any self-respecting individual continue to support and promote a cause if doing so means that he or she must continually lie, distort, misrepresent or skew the facts?  Is blind hatred that much stronger than reason?  Is there an alternative I’m overlooking?

  • 4
    MIkey says:

    Because they are not self respecting.. they are out to get what they want, and are willing to say or do whatever they can to achieve it.. NOM has been well known for this.. and they will continue to do this if it continues to serve their purpose..

  • 5
    Neil says:

    @clayton: Donations, which is to say Brian Brown’s salary.  OR, that they feel so strongly about their “mission to protect marriage” that they’ll feel any means justifies the end.

  • 6

    […] Brown dishonestly claims that 57% of Washington state voters oppose changing the definition of marriage (in fact, less than […]

  • 7

    […] Brown dishonestly claims that 57% of Washington state voters oppose changing the definition of marriage (in fact, less than […]

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>