Alex Blaze

A short lesson on polling would do the fundies good

Filed By Alex Blaze | May 12, 2009 3:00 PM | comments

Filed in: Fundie Watch, Marriage Equality, Media, Politics
Tags: marriage, marriage equality, New Hampshire, New Jersey, peoople home, polling, Questions, same-sex marriage, survey

Don Sherfick emailed me a link to Cornerstone Policy Research's site where they claim they polled everyone in the state of New Hampshire and found that 64% of that state's population opposes same-sex marriage.

Yes, someone wrote that on their site, and no one's gone to that person's house and asked if she or he needs to lie down a bit.

Jeremy Hooper is asking for proof, which makes sense since this poll has already made it into the local media.

The organization has refused to release any proof showing that they actually did call everyone up, and the math shows what a monumental task this is. Let's assume that they polled a diligent 12 hours a day (since they can't be calling people at 3am), for the two days they said they polled, and called all 432,398 households in New Hampshire. I was polled last year during the presidential run, and it took a good 9-10 minutes, but considering how few questions the CPR was asking, let's say it took a full 2 minutes to dial, wait for a response, introduce the topic, ask the questions, wait for responses, collect demographic information (ok, it's doubtful they were that scientific), thank the person, take a breath and begin again.

That would mean that they had to have 601 staffers polling for both the full 12 hour sessions, without breaks.

And that's assuming that everyone was home during that time period and that everyone picked up the phone and finished the interview. The people who do that are a fraction of the people called. Daytime polling, which they must have engaged in, is considered by pollsters to be a way to cut corners and results in less accurate numbers, but I don't see any other way for this poll to have been completed without hiring several thousand callers. (Update: mcc in the comments pointed out that they could have robocalled all these people, which would have meant that their completion rate was really, really low. People don't like answering computer's questions. That would make their claim that they asked everyone in NH even less likely, considering how many people would hang up on a computer.)

(This article provides some more information about how polling a large sample in a short time period might be an impossible feat, and it discusses polling a little over 6000 people over two days and how implausible that might be, not the around 430,000 households the CPR claimed to call in two days. And the article estimates that an interviewer can do only about 5-15 interviews an hour, which is a lot less than I thought someone could accomplish in my estimate, meaning that 1803 callers would have to work without breaks for two 12-hour sessions.)

I also wonder about why it's important to interview every "household" instead of every person, since there's no reason to believe that everyone who lives under the same roof shares the same beliefs.

Last, there's the question of the questions themselves. They weren't worded objectively, professionally, or clearly at all, saying that the same-sex marriage bill before the New Hampshire legislature right now is "dangerous."

In the interest of no-spin, here's how the first question went that gay marriage opponents said led to 64 percent agreement.

"This survey concerns a new law the state Legislature just passed that will affect marriage in New Hampshire. Do you agree that marriage between only one man and one woman should be legal in New Hampshire?'' the question asked.

It's not a law, and the question chooses the positive course rather than to simply ask, "Do you oppose a legal marriage between two men or two women?''

This is also when even word placement makes a difference. Look to the question above and change "marriage between only'' to "only marriage between'' and the emphasis is stronger and gives a better clue as to the level of opposition to gay marriage.

Here's the question about calling Lynch:

"During his tenure, Governor Lynch repeatedly promised us that he would oppose gay marriage in New Hampshire. Now the state Legislature is sending him HB 436: a dangerous bill that will legalize same-sex marriage in New Hampshire, would you be willing to call Governor Lynch today and tell him to oppose same-sex marriage?''

Lynch never promised or pledged anything. He did say in the 2006 campaign he opposed gay marriage and, throughout this late winter and early spring, said marriage should be reserved as an institution for a male and female union.

Using a lightning rod word like "dangerous'' in a poll question always cranks up a desired response. The N.H. Freedom to Marry Coalition's own pollster - the UNH Survey Center in this case - found 55 percent said they support same-sex marriage.

Let's see, who am I going to believe: a respected university polling outfit that polled a representative sample of the state with objectively-worded questions or an advocacy group's in-house polling where they ridiculously claim that they called everyone in New Hampshire and found out what they think about the "dangerous" bill before the state government?

There's a reason the big kidz outsource their polling: they want people to have confidence in the results. These claims are simply laughable.

But they'll make their way into the media, I'm sure, or at least the Religious Right's own publications. It'll turn into "a poll found that 64% of New Hampshire believe marriage is between one man and one woman," and it'll become the truth.

Although I do have to say that I'm puzzled by their choice to say that 64% of New Hampshire households oppose same-sex marriage instead of going for broke and saying that 95% of them do. If they're going to lie, why not do it with style?

Leave a comment

We want to know your opinion on this issue! While arguing about an opinion or idea is encouraged, personal attacks will not be tolerated. Please be respectful of others.

The editorial team will delete a comment that is off-topic, abusive, exceptionally incoherent, includes a slur or is soliciting and/or advertising. Repeated violations of the policy will result in revocation of your user account. Please keep in mind that this is our online home; ill-mannered house guests will be shown the door.

My experience phonebanking is that very, very few people pick up the phone on a number they don't recognize. The idea of surveying "every household" in a state is just ridiculous.

This said, it is possible their polling was done by a computer (which would have lead to even fewer poll completions-- since how many people would hang up immediately upon hearing a recording at the other end of the line, or hang up once the recording started ranting about "one man and one woman"?).

Don Sherfick Don Sherfick | May 12, 2009 3:40 PM

Most likely if they ever do provide a more detailed explanation it will be to say that they just said they contacted every household and that they never said that every household responded with an opinion.

But concerning their overcoming some big logistical problems in getting enough personnel to do the polling, that's not such a biggie for them for the precedent: See "Loaves and Fishes; Biblical Multiplication Techniques".

Angela Brightfeather | May 13, 2009 12:51 PM


Come on!!! This is news? The RR has been doing this for years, and before that Senator McCarthy was doing it for years and before him the Nazi's were doing it for years. Throwing out false numbers from polls that no one ever remembers taking and using those numbers to make their point and make other people get of the fence and join the multitude, is nothing new.

What would be news is finding a good way to fight it.

Last night I saw Joe Solomonese on CNN up against some jerk from Friends of Marriage talking about our favorite bimbo in the news lately. The guy was throwing out poll numbers like the announcement of the winning numbers for the lottery in all 50 states. Joe did a commendable job, but he had no numbers to throw back at this idiot and when the other guy said that the majority of Americans are in favor of SSM, Joe never said that those numbers, while reflecting the ideas of 52%, totally ignore the fact that 48% think another way. Joe was right on every point that he made, but where are our numbers? How about the change in the numbers a month after Prop 8?

The bottom line is that we don't use our polling numbers as well as the RR uses theirs. We have to start using the difference in the numbers between now and 4 years ago to really paint the picture and shut them up, while energizing our base and making their base feel like their losing this fight, along with hate crimes, ENDA and DADT.

You have to give them the feeling that it's not going their way any longer and that they won't be able to use their numbers in the near future if the trend continues. Whatever they are throwing out there today, will be written on their tombstones as proof of the progress against them.
These people are as conservative as a Charlotte banker sitting on his bailout money when it comes to polling numbers and I think they will be hung by those numbers a year from now.