Pictured above: Janine Parry, who grew the Arkansas Poll into a nationally respected source, will retire following the spring semester. (Photos provided)
Dutiful, steadfast, diligent — that is how Janine Parry, professor of political science at the University of Arkansas in Fayetteville, describes herself. It is a description befitting the founding director of the Arkansas Poll, which has checked the pulse of Arkansas’ policy opinions annually for 25 years.
“It’s sort of a public service,” Parry said. “I’ve managed to get a little bit of research out of it, and it certainly informs my teaching, especially in the Arkansas politics class.”
The University of Arkansas position was Parry’s first tenured position, and she took it right after completing graduate school in her native home of Spokane, Wash. She replaced Diane Blair, who had retired, and met the university’s need for a specialist in state politics who could teach introductory courses on state and national government. As a bonus to Parry, Blair had taught a class on women in politics, one of Parry’s interests.
The Arkansas Poll was the brainchild of William Miller, who has since left the university where he served as founding director of the doctoral public policy program, which he was starting at the time of Parry’s arrival in 1998.
“He was interested in learning more about what people wanted than elections could provide,” she said. “I was a new faculty member with specialties in American state politics, including voting behavior and public opinion, so he thought it was something that should be done and, to be honest, that as the new person, I was the person who should do it.”
Collis Geren, who was then dean of the graduate school, funded the poll’s first two years, she said, and then former Sen. Tim Hutchinson helped the poll secure a Department of Education grant with the support of the outgoing Bill Clinton presidential administration. That grant turned into an endowment that now funds UA’s Diane D. Blair Center of Southern Politics and Society and, through that, the Arkansas Poll, Parry said.
The objective of the project was to provide a richer look at what people want from government than elections allow, she said, adding that although there is a lot of polling around elections, 99 percent of it uses a national sample, while 95 percent of public policy is made at the state and local level.
“People in Arkansas might want something quite different than the average person from Pennsylvania or Idaho, and yet there’s just not that much state-level public policy [polling] that’s happening,” she said. “There have been all kinds of initiatives over the decades to try to get more state-level polling projects going on, but they’re just really hard to fund and to keep running under traditional academic models.”
It can be hard to publish state-level research because national-level research is more appealing and familiar to political scientists, and that can be frustrating to someone who specializes in state politics.
“There’s so much more interesting variety happening at the state level, and all those policies are of so much greater consequence in terms of people’s daily lives, but it’s just a bias that political science has always had that somehow national-level data are better. Then it’s so hard to publish from because when you send it out to journals, those aren’t state-level journals; those are national and international journals. I have to answer this a lot — and I do other research too — but when I’m trying to publish from the Arkansas Poll, reviewers say all the time, ‘This is an Arkansas sample. Who cares? Arkansas isn’t representative of the country,’ and it’s like, ‘Well, it’s representative of a lot of the country, and you’ve been publishing data out of California for four years. How is it representative of the country?’ I’ve had to fight that battle so many times.”
She said she became interested in state politics due to passionate professors who taught what is often not considered to be a core course in political science curricula.
“The people that do this kind of research are engaged in politics in a meaningful way that I haven’t found for people in other specialties,” she said. “They know the local lawmakers, and they go to the meetings, and they observe. It’s kind of an older style of political science. The state politics crowd, I think, actually cares more, like really cares more, about the health of the republic. They want their studies to be used in a way that benefits people, not just to get into the highest-tier publications.”
The Arkansas Poll provides much-needed information about Arkansas policy views since state-level elections in Arkansas take place during non-presidential election years, she added.
“At best, 40 percent of the people are coming out and at least picking the red team or the blue team,” she said, “but that’s less than half of the people, and it’s not providing us with that much information, so we thought we would get in there and figure out a little bit more about what kinds of projects people supported, which kinds of regulation and how much, if any, and if they were satisfied with the representatives of the teams they had chosen, so that’s what we’ve been doing.”
The poll still uses “pretty traditional methods in a pretty traditional state,” she said. Although the amount of respondents reached by cellphone has increased, the poll has always been conducted by phone. About 800 respondents are included.
The poll costs about $50,000 each year, she said, adding that although the cost could be reduced to a few thousand dollars if the poll was conducted online or by text, the respondents would likely be younger and more educated than the average Arkansas voter.
The survey includes 60 or 70 questions, about a third of which are demographic, and takes about 20 minutes. About 75 percent of the questions are the same every year, she said, but the poll also facilitates outside research on topics such as taxation, education and other topics that may be only tangentially related to policy.
Parry said she has struggled with questions about abortion. After the overturning of Roe v. Wade in 2022, one would be hard-pressed to tighten Arkansas abortion law further, so it seems a bit nonsensical to ask a question like, “Do you think abortion policy should be the same, less strict or stricter?” she said. However, by changing the question, the poll would lose comparable data from previous years.
That comparative data is important to understanding matters such as political turnover, she added.
“Arkansas’ party transformation between 2010 and 2014 was bigger and faster than any state’s party transformation, at least going back to the Great Depression,” Parry said. “West Virginia comes really close around the same time and, of course, for basically the same reasons, but by the metrics that we’ve looked at, looking at all 50 states, Arkansas’ party flip was the most and the biggest, and it’s kind of cool that we’ve got public-opinion research to show it happening in real time, which most states don’t have.”
A snippet from the Arkansas Poll’s recent survey results.
One of the most pronounced takeaways from the 2023 poll was the comparatively low approval rating of Gov. Sarah Sanders, who ranked lower than average for all governors and had the lowest gubernatorial approval rating in Arkansas since her father, Mike Huckabee. However, Parry said, the low approval rating is less exaggerated when compared to other high-profile governors such as Florida Gov. Ron DeSantis, Texas Gov. Greg Abbott, California Gov. Gavin Newsom or Washington Gov. Jay Inslee.
“They have all struggled to keep their state-level approval rating above 50 percent because they’re kind of in that national fray,” Parry said. “They’ve departed from the gubernatorial norms of roads and industry — attracting new industry, building or maintaining roads and funding schools — which is kind of bipartisan most of the time, so for those who’ve moved away from that in order to make national bids, most likely, down the road, their approval ratings take a hit, so she’s just about average in that peer group.”
She added that the poll has been successful in providing insights regular election polling might not provide.
“I think the poll has facilitated a more meaningful conversation about public policy than otherwise would have been possible, so I’d give it pretty high marks on that,” she said. “You know, in election years, people concentrate on the horse-race contest and the people who like who’s ahead versus the people who don’t like who’s ahead, and sometimes I feel like I’m standing at the side of the road, trying to flag down all the things that people want and are saying, but I do think we’ve contributed to the conversation meaningfully.”
Furthermore, Parry said, the poll has historically been fairly accurate when it comes to predicting election results.
“We’ve got a good track record, which I keep track of because I don’t want to do it if it’s not accurate most of the time,” she said. “I certainly don’t want to push out information that isn’t valid. We’ve done well on that.”
However, research from the poll that could reveal how the public feels about policy is often lost amid partisan politics, she added.
“I wish people would consult [the poll] more with respect to public policy,” she said. “I wish there were more examples I could point to of it maybe raising an issue for people and leading to a specific bill being introduced or even an amendment being introduced and substantively influencing public policy that way. There aren’t that many examples of that because, I think, this is just a tough environment with hyper-polarization and an emphasis on the politics of outrage or the politics of performance instead of problem-solving.”
That said, the poll has influenced policy-making in the past. Parry pointed to a question about spanking in schools that was introduced by a member of the state board of education who hoped to test the waters when it came to banning the practice. When the question was introduced in 2009, about two-thirds of respondents said teachers should be permitted to spank children at school, she said, so the board member backed away from the topic. However, she added, the number has since dropped to half.
There are also instances she can recall of journalists using the poll to provide context when reporting on topics such as gun control or abortion legislation.
“Usually it feels pretty good, I guess, when I see people use [the poll] in a way that brings light instead of heat. It makes me feel like it’s a positive contribution,” she said. “There certainly are examples of it just contributing more heat than light, and that has been occasionally demoralizing, but I try not to dwell on that. It’s such as small part of my job at the university and such a small part of my life, and I try to direct and redirect people to use it in constructive ways, but it turns out people don’t always do what you want them to do.”
Parry received the UA’s Charles and Nadine Baum Faculty Teaching Award for Teaching Excellence in 2016.
During a time when a single political party may comprise 80 percent or more of a state’s legislature, data from the Arkansas Poll shows that most voters are not as deeply polarized as one might expect when it comes to public policy, she added.
“Most of us don’t really have these hard and fast extremist positions on the menu of public policies that our state legislators are hurriedly putting in place,” she said. “I think if we would look at those results, we might be able to kind of find each other again and maybe figure out what reforms could recenter or at least bring back from the fringes our candidates and our office holders and thus our public policy.”
In addition to directing the Arkansas Poll, Parry teaches four or five classes a year, mentors students and writes publications. She has co-authored research papers on direct democracy using data from the Arkansas Poll. She is currently working on a book that she hopes to finish before she retires at the end of the spring semester.
Parry said she began her academic career at a relatively young age, so she is planning to retire from university life earlier than most at 52. When her two twin children graduate from Fayetteville High School at the end of this school year, she and her husband will spend their summers working at national parks, she said. They will be based in Spokane, where she plans to work at city parks during the off-season, she added.
“That’s my goal,” she said. “No one will talk to me about approval ratings. I’ll just weed a public space in a light rain.”
She added that she also hopes to travel via the U.S. Fulbright Student Program.
The future of the poll has yet to be decided, Parry said, but she is working to leave as much information as possible for a new director.
“I hope there’s an interest in continuing it because it really is a unique thing about Arkansas in a good way,” she said, “but I totally understand if someone doesn’t want to take it up because it is quite time-consuming, and university life isn’t really set up to reward folks for the kind of investment it requires, so stay tuned.”
Parry said it is best if the director treats the poll as clinically as possible, works transparently and addresses any legitimate concerns about the poll. It took her 10 years to figure out how to glean from the poll publishable academic research that would garner recognition, she added.
“I would advise them, especially if they were as-yet untenured, to figure out earlier than I did how to make it a public service but also produce publishable research, and this next generation of scientists, I think they’ll be better at that naturally than I was out of the gate,” she said. “I think they have a stronger understanding of research expectations. It’s just a different era, and I think they’re more mature scholars in some ways.”
As for herself, Parry said she is ready to return home and spend more time outdoors than academia allows her. There are a lot of similarities between Little Rock and Spokane, she said, adding that the Arkansas Poll has taught her about the place where she spent her directorship.
“I feel like it has helped me understand a state that’s famously misunderstood,” she said. “I think she even doesn’t understand herself well, sometimes, that maybe she believes the bad press, sometimes, about herself, and really, Arkansans aren’t that different in so many ways than other states around the country. I mean, it’s a more rural state than other places. There aren’t a lot of big cities here. It’s older than a lot of other states. It’s whiter than a lot of other states. Those are the demographic realities, but it’s still a place full of lots of different kinds of people who have lots of different kinds of wishes and needs, and I feel like the poll has let me see those more clearly than I might have otherwise.”
