In the past couple of weeks, the Albuquerque Journal has published some poll results regarding hot-button topics from this legislative session.
The surveys, conducted by New Mexican company Research and Polling Inc., asked registered voters about right-to-work, minimum wage increases, driver’s licenses for undocumented immigrants and third-grade retention legislation.
Some have used the results to support their causes, while others have called the questions flawed. New Mexico Political Report spoke with the company’s president Brian Sanderoff about how his company’s polling questions are written and how he strives for well-balanced questions.
Sanderoff started working in polling after he graduated from the University of New Mexico. His first company, Sanderoff and Associates later became Research and Polling.
While much of the attention he gets from the public comes from election and political polls he creates for New Mexico news outlets, he said about 99 percent of his work is actually with non-political polls.
What is the general process for one of your polls?
Typically a client will tell us what they want to learn. Then we recommend to them a certain data collection method. Is it survey research? Is it focus groups? If it’s survey research, what’ the best way? Is it phone or internet or direct mail?
We write the questions and then we send a draft to the client. Often times we’ll go through a first, second, third, fourth, fifth draft until we get something that both sides can live with.
What goes into coming up with poll questions for media outlets like the Albuquerque Journal?
The Journal will tell us what topics they want to cover and we’ll cover them. The client always decides what they want to cover. It’s their poll. Then we draft the questions and we often times engage in a discussion with our clients about the questions. Our goal at Research and Polling Inc. is to be as objective as possible.
Some people have said questions can be drafted to result in a desired outcome. How accurate is that?
Yes, the questions can be designed in such a way to affect the outcome. Sometimes you see some questions that are slanted.
Sometimes when I craft a question on a complex issue, I’ll look around and see what’s been done out there by other polling companies. With right-to-work, I had a hard time. Questions were always slanted to the left or to the right, so I tried my best to write a question that I feel I can defend. There are times when we can’t come to terms with clients and we will not do the survey or they will reject us.
How much input do clients like the Albuquerque Journal have on how questions are worded?
All of our clients have input on the questions. We will go back and forth until we come up with something that we both feel comfortable with. That’s the nature of polling. We write the questions and send them the draft. Then, they will give us the feedback and we’ll discuss it until we have something that both sides think are appropriate questions.
It sounds like there’s a balance between a fair question and the client wanting a certain product.
Right, but I won’t balance my integrity. I have to live with the questions at the end of the day. It’s Research and Polling’s question, not the client’s question. So if somebody wants to criticize the question, I want to feel good that the question is something that I take full responsibility for and can defend.
Some recent polls that state republicans have used to support their causes are right-to-work, third grade retention and driver’s licenses for undocumented immigrants. What kind of work went into drafting questions for these topics?
On the driver’s licence question, we’ve asked it probably three other times for the Journal, so on this question, I made the call to ask it the identical way as the prior time and the time before that so that [the Albuquerque Journal] can track the results and the results were very similar to the past. The opposed went down a couple of points because this time we did a poll of registered voters and registered voters are a more inclusive group. When you survey registered voters compared to likely voters, you have a younger population, so opposition went down a couple of points.
(The question Research and Polling Inc. asked was “Do you support or oppose New Mexico’s state law that allows driver’s licenses to be issued to immigrants who are in the country illegally?”)
Notice it doesn’t say “Do you support it?” or “Do you oppose it?”, it says “Do you support or oppose..?”
You’re making no value judgements trying to push people to it one way or the other. People are more likely to say ‘yes’ to anything. You want to have the support and oppose in there.
Notice that I didn’t ask, “Do you oppose or support the legislation to repeal [the law].” The reason is, I’ve learned a long time ago that people get confused by double negatives.
Notice it says “..immigrants who are in the country illegally.” I don’t say illegal immigrants. I think there’s some connotation to that.
(In another poll, Research and Polling asked, “Do you support or oppose a state law that would require public schools to hold back third-graders from advancing to the fourth grade if they do not have adequate reading skills?”)
Again, notice it doesn’t just ask, “Are you for this?”
The word in the legislation is “proficient” reading skills. We pretested it and some people didn’t know what the word proficient meant, so I went with the word adequate. I probably spent an hour coming up with the word adequate.
Some people say you can work for desired results. Actually, it’s pretty aggressive. Imagine people agreeing to a state law that would require the schools to do something. So I thought the question was fair.
The people that support the governor’s position are using the word retention.That’s a pretty well-crafted word. I said what it is, holding back.
So, I can defend that question. For people that say the question is designed to help the governor’s position, I totally disagree. I did not use the terminology that the supporters are using.
(On Sunday the Albuquerque Journal published a poll with the question “Do you think that workers should or should not be required to pay union dues or fees, as a condition of employment, in organizations that have collective bargaining agreements with unions?”)
I was trying to get at the issue in the legislation. There’s a distinction between the issue and the best pro-business or pro-union arguments.
I went up to Santa Fe for [former New Mexico Lt. Governor] Mike Runnels’ memorial service and I got an earful from both pro-business and pro-union people. People asked, “What if you put in the question that people who opt out of a union get to enjoy the benefits that are negotiated by the union members? Is that fair?”
The pro-business people asked, “Why didn’t you put in there that you’re forced to join the union, even if you don’t want to?”
There’s a difference between the issue of the legislation and the pro-union and pro-business arguments. Both the pro-business and pro-union have some excellent, effective arguments. I stuck to the issue.
The issue of right-to-work is whether or not all workers have to join unions in union shops. I put in the question, “..in organizations that have collective bargaining agreements with unions.” In other words, do you have to join the union in a shop that is already union? Also, technically you don’t have to join the union. A lot of the questions that are skewed to the anti-union position, ask “Do you think people should be forced to join a union?” I didn’t use any word like force. You still don’t have to join the union, you just have to pay the fees under current law.
One person asked me, “Well what do you mean? It doesn’t have to be a condition of employment.”
Under state law, if there’s a collective bargaining agreement between a particular unit of state government employees and the union, you’ve got to pay and if you refuse, you get fired. It is a condition of employment.
So you might ask, “Why didn’t you put the best two union arguments in the question?” and “Why didn’t you put the two best pro-business arguments in the question?”
Well, it’s complicated and based on 30 years of experience, the [questions] would have washed.
The unions have this fairness argument, which is a very good argument. Businesses have some good arguments too. “Why should you force me to join a union if I don’t want to?”
Then they have their economic arguments on both sides. Had I put the best two pro-business arguments and the best two pro-union arguments, the results would have been the same within a couple of points. I’ve done it with the driver’s license [polls]. The very first time I asked the driver’s license question, I put in there the best arguments from both sides and the results were the same.
When I went to Santa Fe, my union friends saw me and asked, “Why didn’t you ask about the opt in?”
I said, “That’s not the issue. That is one of your very effective arguments as to why you support your position.”
Yes I could have asked a follow up question on that and people would have said, “Yeah its unfair. People shouldn’t have to be able to opt out of the union and not get the benefits.”
I could have asked a pro-business question and gotten, “Yeah, people shouldn’t be forced to join the union.”
You can write the question any way you want to get a desired outcome. I try to write it to be as balanced as possible.
Bruce King used to jokingly say, “I try to treat everyone equally unfair.”
In other words, you just make sure people groan on both sides of the fence. Then you know you’ve probably done the right thing.
How long did you have to work on the right-to-work question? It seems like a complicated issue.
I worked on that question for three hours. That’s literally just on that one question. The first thing I did was research other questions that were asked throughout the nation and I couldn’t find any good questions. They were always skewed to the left or the right. Then I worked on my own and reviewed the legislation and decided not to do the double negative.
Frankly, it’s hard to surprise me anymore with results. I pretty much know what’s going to happen, I’ve been doing this for too long. I wasn’t sure what the results were going to be on this question. I was just as concerned about getting heat from right as from the left.
The questions and responses were edited for clarity and style.
Correction: This piece previously said “subjective” when it should have said “objective.”