The Marquette Wire had the opportunity to sit down with John Johnson, a research fellow in the Lubar Center for Public Policy Research and Civic Education. Alongside Charles Franklin, professor of law and public policy and director of the Marquette Law School Poll, Johnson works on all of the polling projects done by the Marquette University Law School, including polling for the 2024 presidential election.
Johnson was recently named to the Milwaukee Business Journal’s 40 Under 40 list for his work in analyzing housing, demographics and political trends. The list recognizes rising stars in local professional settings for their current impact and the outlook for future influence.
This interview has been edited for clarity.
Your areas of interest are housing, demographics and political trends. What about those areas stands out to you and how did you become interested in looking at them?
The election stuff is what I’m formally trained in. I studied political science in grad school [and] was originally hired here to work on the poll, which I still do. So that’s always been a great interest of mine. Then, I got very interested in understanding the housing market in Milwaukee when I realized that the data we have in Milwaukee about homes and homeownership is quite good for American cities, but not a lot of people were using it.
There was a great deal of interest in housing issues, particularly following the publication of “Evicted” by Matt Desmond, but also from many organizations working on those issues and in Milwaukee. So, I did an early project with my colleague Mike Gousha looking at how the end of the residency requirement in Milwaukee had affected neighborhoods.
The residency requirement was a rule that if you worked for the city of Milwaukee, you had to live in city limits. This was struck down— a law was passed and there was some litigation ending that rule in the mid-2010s, and so we were looking to see what had happened in the neighborhoods where lots of city employees had left. Through that project, I realized that there were all kinds of interesting questions we could ask from some of that data, and so that was really the genesis of that project.
I would say, in contrast to some researchers [that] have one burning question that they spend their career trying to answer, I’m more motivated by answering the questions that other people come to me with, and housing stuff is of great interest to a lot of people in Milwaukee. Then, the demographic piece of it came out of that. You can’t understand demand for housing without understanding how household sizes and general population stuff has changed.
We just came off of an election season. How much information were you swimming in and what was that period of research like for you?
There were three really big projects with the election last year. The first was the poll. We do national polling now as well as state polling. We’re a very small operation— It’s really just Charles [Franklin] and I, so that was a lot.
At the beginning of the year, if you follow politics, you might recall that the state Supreme Court at Christmas a year ago threw out the existing state legislative maps and created this process by which new maps were chosen and convoluted. I won’t get into it now, but I did what I believe is the only real-time, independent analysis of the various maps that were before the court in the state. That drew a lot of interest and took a lot of time for about two months.
Then, around the election in November, I did a project where I collected all of the ward-level election results from every county on election night and over the following days and published those, which was a pretty large undertaking, but ended up creating a data set that wouldn’t exist otherwise of those granular election results. Actually, if you go on the New York Times website now, where they have a nationwide precinct election map, that Wisconsin data is mine to that project. That was a nice thing to be able to do.
Something else that we did, mostly last year, was publish this website where you can look up any rental property in the city of Milwaukee: MKEpropertyownership.com. This might be of interest to some of the students here if [they’re] moving off-campus, moving out of the housing that Marquette owns. But if you’re renting from a private landlord, that’s a good way to learn more about who that landlord is, what other properties they have, how frequently they receive code violations and file evictions and that kind of thing.
There was a study published in December gauging the perception of Donald Trump by asking participants open-ended questions about liking versus disliking him. In studies like that one, where does the research go once it’s completed? Does that information get funneled into another outlet or is it just an independent study?
We share that. We have a pretty large mailing list that we share our work with, largely of journalists— locally, nationally and some international. One of the big audiences that we see for the work that we do here are journalists and policymakers at other outlets, less sort of the public directly, although we do that too.
There were many, many, many editorials and think pieces being written about why people did or did not support Donald Trump. The great thing about polling is that we can just ask people, and then we can represent those people in a statistically proportionate way to the share of the population that they represent.
My goal with that piece was to encourage people to really spend some time reading what people say in their own words about, in this case, Donald Trump. We’ll do that kind of open-ended question and answer format more over the coming year because I think the results are really interesting.
Now that we deliver more and more surveys through the internet, that gives us the ability to ask those open-ended questions in a way where it’s less practical to do it in a phone call. I really enjoy the open-ended stuff. I think they’re more interesting than some other kinds of polling, so I’m looking forward to doing more of that. It is pretty time consuming because we have to read all the responses carefully to make sure there’s nothing terribly offensive that we need to redact before we put it up for the public.
Local news stations repeatedly cited the Marquette Law School Poll to forecast results during the latest election season. How has the polling done here acquired such substantial credibility?
It’s something we’ve had to earn. I started on the poll in April of 2016. The poll was launched in 2012, and the poll correctly anticipated Scott Walker winning his recall in 2012 and then a few months later, Barack Obama winning re-election in Wisconsin. That was a good start, getting those two races correct and the other in opposite directions.
Over the course of poll’s history, it’s certainly had its share of misses, but less than most polls have. We’ve been pretty accurate. In FiveThirtyEight’s rankings before the most recent election (the new ones aren’t out yet), we were ranked third for accuracy in the country and first for transparency, which is something that we take a lot of pride in because we can actually control that. We try to get every race right, but that’s not entirely in your control. But you can control how much you explain about how you’re doing the polling.
We release an extensive methodology statement and then other documentation around each poll, complete results to the questions, those kinds of things. Then, Charles in particular, spends an enormous amount of time meeting with community groups and journalists to explain to them exactly what we’re doing.
I think that open-door policy helps a lot, too. If people have concerns, we’re happy to discuss the data and implementations with them to help make they’re reporting as [accurately] as possible. It also helps that Wisconsin is a pretty small state, and most states our size don’t have a dedicated polling operation, of course.
As someone from Illinois, what else do you think to be unique about Wisconsin?
The first presidential election between the Democrat and Republican [parties] was in 1856 and if you look over the history of the country since then, no state has been as closely divided in three consecutive presidential elections as Wisconsin in 2016, 2020 and 2024. Nowhere has ever been like this before. If you’re interested in American politics, this is the place to be.
Beyond politics, I love Wisconsin for a variety of reasons. I love the outdoors here, I think people are very friendly, and I think that the state’s mix of one big city in Milwaukee, but then a lot of small cities and a lot of, frankly, pretty thriving small towns in contrast to Illinois, is really cool.
Your research has appeared in publications such as The Atlantic, The Economist, The Milwaukee Journal Sentinel, The Financial Times and, as you said, The New York Times. Did you have a moment in this journey where you saw your work somewhere and thought, “I made it?”
I remember, when I was pretty early on in the job, the first time I helped a reporter from The Atlantic on a story. It wasn’t an important story or anything like that, but I remember that I was like, “Wow. This is the first time I’ve worked with a national journalist.” Now, that’s fairly routine. Now, I’ve worked with enough journalists that my opinions about news outlets are based on specific interactions with reporters.
I think really highly of The New York Times because the data journalists they employ there care so much about getting it right that they’re a real pleasure to work with. And there’s some other outlets that you work with sometimes where you can tell that they’re just looking for somebody to give them a quote that’s consistent with the narrative they’ve already decided to tell. That’s much less fun for me.
The Milwaukee Business Journal’s 40 Under 40 list selects honorees based on “the impact they’ve had in their jobs in the community and on Southeast Wisconsin overall.” Beyond being validated for the work being done, what does it mean to be specifically recognized for making an impact on the local community?
It was gratifying to know that people are reading and valuing what I do, because we’re not selling subscriptions or something like that. Sometimes when you write something, build a tool to explore some data, something like that, and you put it out in the world, you don’t always know if it’s finding the audience that you want it to. So this was a nice validation.
This story was written by Lance Schulteis. He can be reached at lance.schulteis@marquette.edu.
