What it’s like to be a pollster?

What It’s Like to Be a Pollster? A Working Podcast Transcript.

What It’s Like to Be a Pollster? A Working Podcast Transcript.

Slate Plus
Your all-access pass
March 8 2017 11:11 AM

The “How Does a Pollster Work?” Transcript

Read what Jim Gerstein had to say about how polling is more important now than ever.

Jim Gerstein.
Jim Gerstein.

Photo illustration by Slate. Photo by GBA Strategies.

This is a transcript of the Feb. 13 edition of Working. These transcripts are lightly edited and may contain errors. For the definitive record, consult the podcast.

Jacob Brogan: You’re listening to Working, the podcast about what people do all day. I’m Jacob Brogan.

This season on Working, we’re talking to people employed in fields threatened in one way or another by the Trump administration’s agenda. These are the stories of people doing difficult, important jobs. Jobs that may get a lot more difficult and a lot more important in the years ahead. Trump himself recently tweeted that all negative polls are fake news. While that statement is obviously, indefensibly ridiculous, we wanted to get a better sense of why it was so ridiculous. To do that, we talked to a pollster. We spoke to Jim Gerstein, this episode. He’s of GBA Strategies, a company that conducts public opinion research and qualitative polling for Democratic candidates, advocacy groups, and others. He led us through his daily efforts from developing good questions to conducting polls to analyzing the results, to help us better understand what actually goes into a poll—positive, negative, or otherwise. Of course, we also talked about how the recent election is changing the way people look at polling data.

What is your name and what do you do?

Jim Gerstein: My name is Jim Gerstein. I am a public opinion researcher for a company called GBA Strategies.

Brogan: Is your name one of the initials?

Gerstein: I am the G in GBA Strategies. We picked our names out of a hat to see what order we could go in and I drew the lucky straw.

Brogan: What is your actual work?

Gerstein: Most of our clients are candidates for political office on the Democratic side, advocacy groups, nonprofit organizations. We also do work for some civic institutions and businesses as well. The thing that ties all of the work together is getting public opinion to help inform the strategies, and try and figure out how they can lead people to where they want them to go.

Brogan: Is it fair to characterize this, by and large, then, if we’re talking about public opinion, as polling, as we would understand it, or is that an inaccurate term?

Gerstein: Yeah, polling is part of it. It’s a variety of survey research. We also do qualitative research, so we will do focus groups or one-on-one interviews with people within those audiences that we want to learn from.

Brogan: What are the biggest differences between the qualitative focus-group-based research that you do and phone surveys with large bodies of people that you’re analyzing?

Gerstein: The big difference is how you use it and we like to say that surveys, those give you the what, and focus groups give you the why. It’s a little bit of a simplification because you’re getting what and why in both, but these surveys will give you a snapshot in time of where the public is quantitatively in a representative sample. You can say, “OK, 60 percent of the American public believes X.” The focus groups, again, they really tell you why do they believe X, and let’s say X is 60 percent of the public supports having Syrian refugees come into the country, we know that now. I’m making up the number, but it’s an illustrative example. The focus people say, “Well, why?” Because their families were refugees, were immigrants, or because they see the images on TV and it draws them into accepting because for whatever reasons we learn in the focus groups. That’s how they’re used differently and you can see how they complement each other because to know what the number is, yeah, it’s important, it’s valuable, but to understand what’s behind it will help you communicate it better. It will help you understand it better, it will help you understand how this debate can shift people’s views on it. That’s why we like to do both whenever possible.

Brogan: What are your clients doing with the information once they receive it?

Gerstein: It’s multifold. First piece is for them to understand the concept in which they’re working, so if it is a candidate for office, it’s to understand the mood of the electorate before we get into the details, but that’s the next part of it. Once we understand the context in which we’re working, to then get into the other pieces that may drive decisions. How they view an individual or view an issue or view an organization, and what messages you can use to lead the people to arrive at the conclusions that you are seeking to advance.

Brogan: Are you involved at all with that kind of downstream application of the research?

Gerstein: Yeah, so we as a company try to be involved both on the front end and on the back end in order to make the research as useful for the clients as possible.

Brogan: How do you figure out what you’re going to pursue for any given client? Is that part of conversations that you’re having with them about strategy and planning?

Gerstein: Yeah, it’s funny, the same exact question the clients ask us when we first start the process. We’ll start out any project where we sit down with the clients to get their input, to understand what it is that they’re seeking to learn from this, what kind of disputes they have within their team over which direction to go. The research is a great arbiter of that but it’s only a great arbiter if everybody feels that it’s legitimate, and they feel that it’s legitimate if their voice is heard in the survey itself. We draft a questionnaire based on their input and there’s a process where we get their feedback on what they. … Oh, this was missing from the survey, or I love this, make sure we don’t cut it and if all goes well, it goes into the field quickly.

Brogan: How long does that process, from those initial conversations to the actual deployment, usually take?

Gerstein: Each case is different. If something’s breaking in the news really quickly and you need to see how people are reacting to it in order to inform your understanding of what’s going on, we can quickly put together a questionnaire and get it into the field in a day or two. Other projects, there’s not such an urgency to it. If it’s a campaign, you know when the election’s going to be, you know when you want to do your baseline research, and you know when you want to do your tracking surveys so you can plan ahead and take your time. That could be anywhere from a week to two months, depending on how quickly the client wants to move.

Brogan: Can you give us an example of one of those more hurried operations where you have to respond quickly to what’s breaking, something that’s happening in the news?

Gerstein: Sure. The Comey letter, right? The call me letter was a surprise, an October surprise that happened really on the eve of the election and that’s one where you’re not sure how people may react to it. We suspect that when the lead law enforcement official in the country is impugning the integrity or honesty of the candidate that that’s going to have an impact on people. You want to understand more, really, is it important to them at this point in the election? How can you possibly react to something like that given the position or the credibility of the lead law enforcement officer in the country?

Brogan: In that circumstance, what questions would you be asking and who would you be asking them to?

Gerstein: I think we would be targeting the electorate as a whole or you could be looking at your battleground states or you could zero in on the voters that you think are going to make a difference in the election. What you’d want to understand is does this change their views on how they were going to vote, specifically the undecided voters. Which we can see, there was a lot of evidence that this did move votes. You understand why, though, what is it about what he did, what he’s saying that’s raising the doubts, that’s causing this slippage that you see in the actual head-to-head horse race. When you can understand what it is that’s driving that, that helps you know how best to address those concerns that voters would have. That late in the race, it’s very hard to do. I am definitely in the school of thought that the Comey letter made a big difference in the election.

Brogan: Are you looking for phrases that a candidate or campaign might use in response to a sudden event? What kind of questions are you asking at that point?

Gerstein: First thing you want to do is assess the impact. You’ve got all this tracking data overtime in which you can see on key measure whether it’s the actual head-to-head vote, opinions toward candidates that they have on certain attributes. Whether it is leadership or honesty or being an advocate or fighting for the voter, you can see what things are moving and which aspects of this you need to address because you know certain things are very important in determining people’s vote. The other thing you want to see is enthusiasm, does this have an impact on people’s enthusiasm for each candidate and if it’s an enthusiasm question, you need to address that. If one segment of the population is moving, well, that means that that’s who you need to target your communications to in a way that helps address the concerns that that segment of the population has.

Brogan: Does the research that a firm like yours does differ significantly from the work that a big-name firm like Rasmussen or something like this is doing that shows up in the news all the time?

Gerstein: I think the distinction is not necessarily in the sizes of the organization but the purpose of the companies. For example, Gallup. Gallup asks a lot of contextual questions about what’s the job approval of the president, they ask that on a daily rolling basis and they’ll ask certain things about how do you feel about this issue or that issue. How do you feel about the Muslim ban or the Access Hollywood tape, they’ll ask a series of questions. What’s different between that and what the polling companies that work for candidates as opposed to news organizations is we’re trying to help figure out a strategy that will move voters, move Americans to the positions that your client wants. That’s what we do, that’s what the Republican pollsters do. We’re on the Democratic side, it’s funny, we’ll look at poll results and have more in common and more appreciation for what the Republican pollsters are doing because we know they’re coming at it from a similar approach. Then we would say a Gallup or a news organization.

Brogan: You are in a profession that is heavily focused on demographics that has to be extremely attentive to breakdowns by age, by gender, by ethnicity, and so on. What are the demographic breakdowns of your field like?

Gerstein: I haven’t done a survey of them but I guess I can look around the room at various meetings. There is a range of generations, good, nice, age range. The other demographics in terms of gender, traditionally—like many industries especially in politics—has been more male than female, but fortunately we’re starting to see that change a lot in the last 10 years or so. In terms of racial demographics, again, it’s been mostly white for a long time but again we’re seeing more diversity and more African Americans, more Hispanics, Asian Pacific Islanders as well. I’d say it’s still not representative of the country as a whole and a lot of work needs to be done. Something that we’re focused on as a company, for sure.

Brogan: Do you think that those limits on diversity and representation affect the kind of work that your profession more generally can do?

Gerstein: I think a company with diversity in it benefits from the diverse backgrounds that people have because people bring different perspectives to the table and I think the more diverse a company is, the better it should be. Especially in this field where you’re trying to understand different people’s perspectives across the country, it’s important to have those perspectives within your office.

Brogan: How much of your time is spent in business development? Are you having to seek out new clients a lot of the time?

Gerstein: Often our business development strategy has been do good work for the existing clientele because if you’re doing good work for the existing clientele, they’re going to come back to do more work and they’re going to tell other people to hire you and people at that existing company are going to move to another company and that’s how—

Brogan: Or another campaign, I imagine too. Politics, especially, you see a lot of movement between.

Gerstein: Absolutely. If you’re doing good work and people enjoyed the experience of working with you and felt that you provided a good product to us, that’s the best way that you can develop your business.

* * *

Brogan: What’s an ordinary day like for you?

Gerstein: The days can range from traveling across the country to go to focus groups—we’re moderating these discussions and listening to voters.

Brogan: You still do that kind of work?

Gerstein: Yeah. My partners and I, we all agree that hearing directly from voters is invaluable and it puts a face on the numbers when you see the results in a survey, you can think of the different people in the focus group and why they’re answering these questions and the survey the way they are. One day could be taking the commuter jets to small airports in swing states around the country, it could be a day where you’ve got a bunch of phone calls with clients all day, it could be a day where you have none and you’re writing and you’re analyzing and you’re preparing data. It varies each day, which keeps it interesting.

Brogan: Do you have a standard schedule at all? Is there a time when you usually get started, a time when you usually end the day?

Gerstein: The schedule’s often dictated by family needs, right? In the school year, it’s easy to get to the office after the kids are off to school and in the summer it’s after they’re off to camp, which is earlier, so I get to the office earlier in the summer. I try to get home in time to have some dinner with family, get the kids to bed, and then depending on the season and how much work we’ve got going, it may involve a lot of work after they’re in bed.

Brogan: I assume a lot of the focus groups, a lot of polling, is conducted in the evening when people are available. Does that mean that you are often on the job in the evening?

Gerstein: When we’re doing focus groups, yes. I would say that those are the days that are the longest because you have your full day of work, then the focus groups start at 5:30, 6 o’clock, and they’ll go until 10 o’clock. That is a long day. The polls themselves, yes they’re conducted at night, but I’m not sitting on the phone calling a thousand people across the country. We have a calling house that we subcontract to and we will monitor the calls, we have folks in our office who will listen to some of the calls to make sure that funny names are being pronounced correctly and that the survey is being administered properly as it was written. In the morning is when we look at the data as it comes in to see if there were any problems in the calls or in anything.

Brogan: You’re then perusing it and at that point already starting to think toward analysis?

Gerstein: We really don’t do any of the analysis on the partial data because results change. There’s a reason why we have a large sample to reduce the error. The other thing that we do when the survey’s complete is we will weight the date because while we try to produce as random sample as possible and as representative a sample as possible, it doesn’t always come back that way. You have some response bias and it’s harder to reach young people, for example on the phone. They are doing more interesting things at night than older people.

Brogan: If I can speak for my generation, we try not to answer the phone if possible.

Gerstein: I’m fully aware of that, actually. It takes longer to reach your generation. We do reach them but sometimes we won’t reach enough of them. We will need to weight the data up a little bit so that that group is sufficiently represented in the sample.

Brogan: What point are you able to really start thinking about total analysis?

Gerstein: If we finish a survey, 10 o’clock Eastern, we can then run the data overnight if it’s urgent or we will run it first thing in the morning and be able to produce the top line results and the cross tabs—

Brogan: What are cross tabs, for our listeners?

Gerstein: Sure, cross tabs. First the top line results, that’s, OK, what percentage of people approve of President Obama and what percentage disapprove. That’s something that we will usually produce by noon the next day. The cross tabs are, OK, we know that 58 percent approve of Obama and 40 percent disapprove, and 2 percent are undecided, but we’re really curious to see what is that among voters under the age of 30. A cross tab will have what’s the result on job approval by age, under 30, 30 to 49, 50 to 64, over 64, and we will have a book of cross tabs for each question in the survey and that book takes a little longer to produce but we can have that in the afternoon. Then after all that’s done, we’ll also produce something, if it’s a tracking survey we’ll produce something called the time series where we can see, OK, back on January 3 his job approval was 50 but then March 1 it was 53 and then June 1 was 56 and you can see the trend. We can look at that by demographic, what’s driving this movement? That’s another piece. Once all these pieces are put together, we’ll start analyzing it and coming up with our theory of the case.

Brogan: When you are setting out to do a focus group, let’s say, how do you find the people who you’re actually going to engage with, solicit information from?

Gerstein: Focus groups have a fairly straightforward process to it. First of all, we’ll sit down with the client, we’ll say, OK the budget will allow us to do two focus groups or four focus groups. We know that our, let’s say, it’s for an issue advocacy group, and they’re very interested in seeing how young people are going to drive the debate. We’ll divide up the group and say what’s different about the different young people? Let’s do groups in different regions of the country. We’ll say, “OK, well in Florida, we’ll do women and then we’ll split them up in two groups and in Nevada, we’ll split it up into Hispanics and whites.” We’ll mix the gender in Nevada and we’ll mix the race in Florida. Now we’ve gotten our criteria, our specifications for what we want. Then—

Brogan: Those are conversations you have with the client?

Gerstein: Yes. Then we will contact the focus group facility in Florida and one in Nevada and—

Brogan: Those are facilities that you contract out to?

Gerstein: Yes, yes. There are focus group facilities all over the United States. Especially in markets where corporations want to do a lot of market research because they’re trying to sell candy bars. There are focus group facilities all over the country servicing this industry and they have databases. We say, “OK, for this facility in Orlando, Florida, we’re going to have a group of men ages 18 to 30 who are not partisan Democrats, not partisan Republicans and are undecided in their vote for president.” The focus group facility has a database and they will call through that database and ask them a set of screening questions that we’ve provided them and when people pass that screener, their name gets filled out on a list and their demographic criteria and they’ll send us that list. We’ll look and we’ll see, “OK, it looks like this group, you’re filling this one really well but this one it looks like we’re a little heavy on the independent leaning Democrat. For the rest of it, we want you to do Independent or lean Republican.” Then we show up.

Brogan: At what point in the process have you crafted the questions that you’ll be asking?

Gerstein: For a focus group, we call it a discussion guideline and during that 10 days to two weeks of recruiting is when we write the discussion guideline with the client. The discussions last about two hours to make sure of open-ended questions where we’re really trying to get people, their input, in their own words. The questions become less general, more specific, and we can be testing messages or we could be showing them television commercials or we could be giving them video of candidates talking and getting their reactions to that. It starts general, moves to specific.

Brogan: What’s the rationale behind that?

Gerstein: You don’t want to influence their perceptions of individuals or things that you want to know. You don’t want to bias the responses based on information that you gave them previously, but you still want to understand the context in which we’re working and you want to get. … Before you start putting your ideas in front of them, you want to hear in their language and in their point of view what they’re thinking.

Brogan: Do you find when you’re developing these questions that you have to balance what your client wants to hear with what they need to hear?

Gerstein: That’s a gentle way of putting it, Jacob. Yes. The hardest part really is when you’re working with clients, they only often want to hear their point of view but a good survey or a focus group discussion needs to be a balanced discussion in order to get a high-quality response. What I mean by that is if we’re only presenting one side’s arguments throughout the survey, you’re going to start to see people on the other side drop out of that survey and skew your sample.

Brogan: They’re going to lose focus, they’re not going—

Gerstein: They’re going to lose interest. They’re going to think, “Why am I getting this questionnaire and all it’s doing is attacking Hillary Clinton? That doesn’t make any sense, this must be a Trump thing,” and they’ll hang up so your sample will be proportionately Republican, or vice versa.

Brogan: You are, though, personally coming at this from a partisan perspective. You work exclusively with Democratic candidates as I understand it?

Gerstein: Yes.

Brogan: At your firm now, do you ever have to worry about your own biases or those of your employees and your partners when you’re crafting questions, thinking through your research?

Gerstein: Right. We aren’t doing our Democratic clients or progressive organization clients or whoever, we’re not doing them any benefit by bringing a biased perspective to the table. We have a goal of winning but the process has to be very balanced and unbiased in order to be good at achieving the goal and developing the path because it doesn’t do anybody any good to have blinders on and not—

Brogan: Is there a trick to taking those blinders off and checking your bias at the door?

Gerstein: One of the best tools is going to those focus groups and hearing people with different perspectives and how they’re talking about things and understanding where they’re coming from. That helps a lot, to be able to craft different messages for both sides of the debate.

Brogan: Is it ever hard for you to hear the other side?

Gerstein: No. Honestly, it is eye-opening to go around the country and hear different people’s perspectives and where they’re coming from and what motivates them and how their days are going and what’s important to them. I don’t mean to sound cliché or anything like that, but people have important things to say and I love hearing from them. Yes, I have been moderating focus groups where you hear things that you wouldn’t want your kids to hear, and I’m not talking about bad words, I’m talking about bad ideas. You have to have a straight face there and listen to folks and power through it. It is a professional challenge and interest to understand what’s behind that and how do you develop a strategy that’s going to change that.

Brogan: When you’re there for a focus group while it’s being conducted, how involved are you in the process? Are you in the room with the group or are you behind a glass wall? What’s your relationship?

Gerstein: The physical structure is that you’ve got a room with a conference table and about 10 people sitting around the conference table and you may have a television that they can watch a video or a blackboard where you can write on. Not a blackboard—

Brogan: Whiteboard?

Gerstein: —but a whiteboard, thank you.

Gerstein: That’s one room. In that room are the participants plus the moderator. On the other side of that wall, you’ll have a big glass that is a one-way mirror so the people can observe—the clients and the rest of our team of observers—can watch the focus group and listen to it and hear what they’re saying and see it all. The people in the room, while they know that there’s people watching them and we’re very clear and we tell them that upfront that they can’t see the people in the back room.

Brogan: Which side are you on?

Gerstein: I am generally on the side with the participants.

Brogan: OK.

Gerstein: Moderating the folks.

Brogan: You’re actually asking the questions and—

Gerstein: Yes.

Brogan: Is that a skill you’ve had to develop over time?

Gerstein: Yeah. In anything you’re always learning something, right? I don’t think anyone has mastered every particular skill that’s out there. First, I learned it from observing focus groups, you generally need to bring a certain skill set of being able to listen, and truly listen to people and want to know what they have to say and have a sincere interest in what they have to say in order to be good at it. A good focus group is when the moderator hardly talks at all but needs to steer it and get through the discussion guideline. It was a matter of learning it through observation and then doing it. The first time I did it wasn’t as good as the second time I did it. The third time was better.

Brogan: What form does the information that you pull out of those focus groups take when it goes to your clients? I feel like I have a pretty good sense of what forms of polling data for a phone survey might take but are you writing up a report of the focus group, are you giving them raw video, is it edited clips? What’s it look like?

Gerstein: That’s funny—different clients want different things. Some clients want a written report that includes quotes and feels very deep and qualitative and they want to read through it. They want to spend the time to really go through it. Other clients, they want to understand the big takeaway and they aren’t interested in all the detail. Instead of a detailed, lengthy memo, they want to have 10, 15 slides that have the big points and a couple of pieces of supporting evidence. Some clients want to see the videos themselves and some clients say, “I don’t have time for that,” and it varies, but we want them to know, “When you entered this you wanted to learn X, Y, and Z. This is what we learned about X, this is what we learned about Y, and this is what we learned about Z and here’s the big picture about how these things relate to each other and what are the other driving dynamics that are taking place here.”

Brogan: Is it ever difficult to present hard news to them? If the results are not what they were hoping to hear, do you have to soften the blow?

Gerstein: That’s the hardest conversation to have with a client in a political campaign situation is two weeks or one week out, and they’ve given it their all, they’ve poured their life into this campaign for a year and it’s clear that it’s highly unlikely that they’re not going to pull it out. It doesn’t do people any good to—let me give you an example: When a candidate is losing by 10 or 15 points with a week out and they say, “You know what, I’m thinking of putting in my own money to make up the gap here.” You know that it’s highly, highly unlikely that the money they put in—their savings, and that they’re going to have to take out loans for—is going to get them across the victory line.

You need to be honest with them and say, “Look, this is what the data shows. I’m not saying—anything can happen, we know that, but in our experience we’ve never seen somebody make up this big of a gap in this short of time and with these dynamics. We will be with you the whole way no matter what you decide and we will continue to fight and we only work for candidates that we believe in and we believe in what you’re doing and we want to be there by your side but before you dip into the savings, we want you to see where this all lies.” That’s a hard conversation to have but I think you need to be honest and speak to the data in a way that they’re making sound decisions about their own livelihoods and their own lives.

Brogan: Talk about how the results of the recent presidential election are changing the way that some of us are thinking about and looking at polls.

Gerstein: There’re a couple of different aspects to that. One is, as you said, it was unexpected. We need to learn from 2016 to see why was it unexpected and are there any changes we need to do in how we approach our methodology. The truth of the matter is that the national polls were actually on target. The average before the election was plus-three for Clinton and she won by two. The statewide polls were not and there were big differences between the results in many of the key states. We’re looking at that to try and see how do we do things differently, if we need to do things differently.

Brogan: Have you learned much yet in your inquiries since then?

Gerstein: We’re still waiting for the voter files to be updated to see who actually voted and to compare that to some of the projected turnout models that we have. That part of our analysis is incomplete but we have looked at why is it that everybody in this industry, Republicans and Democrats, thought that Clinton was going to win? One of our conclusions is that we all placed too much confidence in small leads. If we were in the same exact situation where the candidate wasn’t Donald Trump and where the race was within two or three points, we would not have had the kind of confidence that we had going into Election Day that we had this year. Why is that? I think that based on all the experience that we had, a candidate like Trump, who was caught on tape saying extremely inappropriate and offensive things about women, who attacked a judge on his Hispanic background and actually had a lot of false statements on it, someone who insults such broad segments of the electorate that we saw, that that generally doesn’t work. Our confidence in the small leads that Clinton had was a larger confidence than we would bring to a race that had [unintelligible].

Brogan: A more conventional—

Gerstein: I think we can over learn some of these things and say I don’t think that we want to take away from this election that everything Donald Trump did was successful. I think he actually pulled several inside straights along his way to win the Electoral College vote. I think we need to be careful not to give too much credit or too much learning to “anything can happen anytime.” Instead, I think that we do need to be a little more careful when we look at small leads.

Brogan: But I assume that one of the other issues that has to come into play here in one way or another is that, rightly or wrongly, a lot of people seem to have come away from the election saying that polls got it wrong. Are you having to deal with that skepticism at all from clients, from the public?

Gerstein: Some of our nonpolitical clients have raised that with us and it’s interesting because where the issue was in the political realm, most of the political clients, it’s not like they’re saying, “Well, we can’t pull anymore polls, they’re meaningless.” They recognize the value of polls, they recognize that they need it to achieve their goals, so it’s not coming from that quarter as much as it is from people who aren’t experienced in it as much. They say, “Yeah, everyone thought Clinton was going to win and polling must be wrong.” I’ll tell you one anecdote, and that is one of the differences between the national polling and the statewide polling, national being accurate and statewide being inaccurate, in many cases, was there were fewer statewide polls and not all of them went up to the last minute the way the national polls did. Things did change in the final few weeks of this election, especially after the Comey letter and we had a client that, two weeks out, it was a ballot initiative and two weeks out we were losing by four and the client wanted to continue polling through the election doing daily tracking.

We continued doing that and we saw erosion all the way up until Election Day to the point we were down by 14 points, we ended up losing by 17 points. Had our last poll been two weeks out when we were down four, the client would’ve said, “Oh my God, you guys don’t know what you’re doing. You said I was going to lose by four, we lost by 17.” Instead we kept polling up until the election and they said, “Well, it’s terrible that we lost but at least the polling was right.” Yes, that’s an example of folks who were involved in that, their takeaway from this isn’t “Oh, we’ve got to stop polling because it’s not going to work,” but people who don’t see that and people who are outside of the political process, we’ve had to explain a little bit and actually tell that story to help them understand.

Brogan: Does the current political climate make the work you do feel more urgent?

Gerstein: Absolutely. I’ve never been so focused on a midterm election so far out from when it’s going to take place. We are—

Brogan: 18, 19 months away from it now.

Gerstein: I’m recruiting top-quality candidates and developing the best strategy that we can. It feels more urgent than it has at any point in my time, in the work I’ve done in the United States and you can see it in the streets of American and the size of these rallies, whether it was the Women’s Marches across the country or the spontaneous reaction to the Muslim ban. There’s an energy, an urgency out there.

Brogan: We’re in an age where a lot of us worrying about the way things are in the world, spending all of our time on Twitter, whatever it is that we’re doing, have access to polling data, polls, in a way that we may not have in the past. Do you have any suggestions for those of us who don’t have a data background, who don’t have a polling background, maybe who don’t even have a politics background as we’re looking at evaluating these pieces of information to figure out what’s worth looking at, what matters, what doesn’t?

Gerstein: Sure. I think one of the things you can look at is the demographics in the survey. If the demographics are in line with what the country or what that particular state or that particular congressional district should be, then you’re in good shape. If a survey doesn’t present the demographics, that should be a red flag for you. Why aren’t they telling me what percentage of their sample is male and female and what percentage is white and African American and Hispanic? What percentage of it is broken down by age? If they’re hiding that, that’s a red flag. In addition to that, you can also look and see what the partisan breakdown is in the survey. What percentage of the respondents are Democrats versus Republican versus Independent? That will give you sense of whether something is misleading or not.

Brogan: Thank you so much for joining us today and sharing your work with us.

Gerstein: I’ve enjoyed it very much, thank you.

Brogan: Thanks for listening to this episode of Working. I’m Jacob Brogan. We’d love to hear your thoughts about the podcast. Our email address is working@slate.com and we do read those emails, we respond to them, and we learn a lot from them, as we’ve said before.