How do political polls work and how are they conducted?

How do political polls work and how are they conducted?

BALTIMORE -- In a world that feels like it's moving and changing faster than ever, one thing remains constant -- everyone will always have an opinion.

Each November, voters go to the polls to decide the leaders of our state, community and country.

However, before the election, polls are released to gauge how the public is feeling about certain candidates and issues.

In this week's Question Everything, WJZ's Rick Ritter finds out how political polls work and how they are conducted.

Surveying the public

Leading up to Election Day, we have been inundated with political polls, texts and phone calls.

Mileah Kromer and Ian Anson, associate professors of political science at the University of Maryland, Baltimore County, have been conducting polls for years.

"I think we need to know more accurately and truthfully what the actual public thinks,"  Anson said.

"Public opinion polls, like we produce, give the public a chance to weigh in on key policy issues and political issues," Kromer added.

How do the polls work?

"A poll, like we conduct at UMBC, is a scientific poll," Anson said. "I would say there's a lot of polling out there that falls short of that definition and that may be one of the reasons why people are so confused with that because they might see a poll on Facebook that says, "How many of you like cats?" and that is not quite a scientific poll."

Rick Ritter: "What is the difference between scientific and non-scientific?"

Ian Anson: "The power of randomness is the biggest part when we conduct the kinds of polls we do. If you live in Maryland, you have an equal probability as anyone else of being selected into that poll. A Facebook poll, or one that has some kind of political agenda, usually doesn't start with that type of premise."

Phone surveys

Ritter: "Twenty years ago, landlines probably dominated these polls, but that's not the case anymore?"

Mileah Kromer: "I think one of the things the public gets the most wrong, or the comment that is like nails on a chalkboard, is that you only poll landline users, and that's absolutely not true. We're calling cellphones and landlines and our polls and others do a heavy cell sample and that means 80% of our numbers are coming from random cellphone numbers."

Pollsters also use text-to-web, which is the same probability-based sampling.

Kromer: "It gives people the chance to answer a survey in a different mode, not just talking to the interviewer, but being able to click on their smartphone. Older individuals still love to take a survey on the phone no problem, but for those under 30, if you want to capture them, you have to do some sort of online component."

How are you selected for a poll?

Ritter: "What do you say to people who still argue that they are not being selected or never get a call?"

Anson: "If you think about the six million people who live in Maryland, you actually have a low probability of being selected in any poll. It's very unlikely that any given individual will be in our poll at any time. I think it's important to remember that even though we're selecting a small group of people, because of the power of randomness, that small group is still representative of this broad population."

Kromer: "If you think about a big pot of soup, you don't have to eat the entire pot of soup to tell me the soup is spicy. You just need one spoonful and that's what we're talking about in terms of random sampling, those 1,000 Marylanders are a spoonful of Marylanders."

Behind the scenes of polls

One of the most labor-intensive parts of polls is the hours and hours of work that goes into it behind the scenes. 

Question wording and question order can have a huge impact. Sometimes it can be as simple as changing a word, or the order of the questions, but once it's out there to the public, there's no going back if there's a mistake.

Anson: "Randomization is our friend a lot of time, where we will often shuffle order questions respondents will see so that we avoid any systematic biases."

Kromer: "You have to find a sweet spot in the question, you can't overload them with information. That balance between how much and how little to put in the question to avoid bias but also avoid leading them is something we grapple with and something we teach the students here to grapple with."

Observing poll making

WJZ sat in on a class at UMBC where the students were working on issuing an exit poll at more than a dozen precincts on Election Day.

"We just started our institute of politics here at UMBC, so this is kind of like our first big project," said sophomore Christian Jassani, a double major in political science and global studies. "It's kind of like a great way to get experience, to kind of go out and figure out how to go out and do polling in a much more like controlled safe environment. What I really think is cool and kind of legitimizes it, is it's not just being collected. We're actually applying this to future research."

It's a future that's already being crafted by the next generation of pollsters.

Kromer: "In the absence of polls, who gets to decide what the public thinks about important policy uses? There is nothing more democratic to me than asking a random sample, whatever your population of interest is, what they think about the politics of the day."

Importance of transparency

The margin of error is another important component of conducting polls. 

Since surveys only hit a sample of the population, the margin of error describes how close we can expect a result to fall true to the population value. 

When Kromer and Anson release their polls to the public, they are always transparent about the margin of error in their press releases.

Kromer: "It is important to show you transparency behind this and it shouldn't be a black box, so we can show you how exactly these polls are conducted so the viewers can decide whether they want to believe it."

Anson: "Every survey we conduct, because of randomness, has this element of wiggle room around the estimate. You need to be comfortable with a little ambiguity in science and the science will always involve some margin of error. We need to always remember, this is just a snapshot. We'd love to know for certain but ambiguity is the way we have to live our lives as scientists." 

Read more
f

We and our partners use cookies to understand how you use our site, improve your experience and serve you personalized content and advertising. Read about how we use cookies in our cookie policy and how you can control them by clicking Manage Settings. By continuing to use this site, you accept these cookies.