Back to top
Strong opinions
Yellow background with smiling person in foreground holding a large cardboard arrow.

Strong opinions

UMass Poll tracks how we’ve changed our minds

Photos by

An American statesman once opined, “In this country, public sentiment is everything. With it, nothing can fail; against it, nothing can succeed.” That statesman was Abraham Lincoln—and his words ring truer today than when he spoke them 168 years ago. However, the science of determining public opinion is no small task, and the UMass Amherst Poll endeavors to do just that. For the past twelve years, a small group of scholars and students have taken the collective temperature of the commonwealth and the nation at large, reporting back our likes and dislikes, our shared values and polarizing beliefs.

In the world of polling, the UMass Poll is a relative newcomer, first established by UMass Political Science Professor Brian Schaffner in 2010. Eight years after its inception, Schaffner handed the reins to Provost Professor of Political Science Tatishe Nteta (Nah-teh-tah), who now serves as the poll’s director. Under Nteta’s leadership, the poll has differentiated itself from its peers by uncovering the hard-won “whys” of public opinion. From our feelings on the economy to progress in racial equity, the UMass Poll follows a lofty north star—reporting on the health of democracy itself.

Over the years, the poll has evolved to get at respondents’ true take on issues—and has tracked some surprising reversals on the most controversial of topics.

Infrequently Asked Questions

When Nteta first joined UMass Poll, running a public-opinion survey wasn’t foremost among his professional goals. “Absolutely not,” he recalls. “But it became clear to me that to fully explore the issues I was interested in—race, immigration, the changing demographic in our country—I needed quantitative methods in my tool kit.” In fact, many of Nteta’s colleagues view their polling work as in service to their academic research, rather than the focus of it. That alone has become a differentiator. “We’re professors first and pollsters second,” says Nteta. “We take the issues we research as faculty, discuss with our students and each other, and use those conversations to craft timely questions. But it’s all with the goal of understanding the civic health of the state and the country.”

Jesse Rhodes, one of UMass Poll’s three associate directors, agrees that the poll is an extension of his academic inquiry. As a political historian, Rhodes uses the poll to get at the underpinnings of contemporary public opinion and place them within a historical context. “For example, if the poll finds a correlation between people’s acceptance of violence [at the U.S. Capitol] on January 6 and their negative attitudes toward people of color, we need to place that in a long tradition of whites trying to maintain prerogatives through hostility toward people of color,” he explains. “Context is critical, and because all of us working on the poll have different backgrounds and research interests, it makes the poll work well.” Rhodes is joined by associate directors Ray La Raja and Alexander Theodoridis, both from the university’s political science department.

Graphic of 61 percent over a yellow marijuana leaf.

Suggested that marijuana legalization had been positive for the state.

Compared to polling organizations like Marist and Quinnipiac, the UMass Poll is comfortable being the turtle to their hare. Nteta and his team craft five to six polls a year, considerably less than the near weekly polls conducted by their peers. However, this suits the UMass team just fine. Though they include “horse-race” questions in every survey, gauging potential voters’ stance on upcoming elections, the team puts a premium on uncovering the social pressures and trends that underlie these results. Getting to that level of inquiry, however, requires thoughtful—and often experimental—methodology.

In recent years, the polling team has seen a tendency among respondents to answer questions in a way that supports their political identities, rather than reveal how they personally feel about the issue at hand. “We call this ‘expressive responding,’” explains Rhodes. “It’s a way for people to assert their identity—but we’re looking to get at their genuine beliefs.” To counteract this tendency, the pollsters often provide respondents with a list of belief statements and ask them, not which ones they agree with, but how many. “In that circumstance, we’ve minimized the incentive to engage in expressive responding,” explains Rhodes. “When you flat out ask people certain questions, you can run into methodological problems. Innovative and experimental questioning helps us get at how people feel in a more rigorous way.”

If I have a normative commitment, it’s to democracy

Another method the UMass Poll employs is free association, asking respondents to describe a politician or political issue. “Rather than just ask, ‘Do you approve or disapprove of [Massachusetts Governor] Charlie Baker?’ we’ll ask, ‘What one word describes Charlie Baker?’” Nteta explains. The polling team then illustrates these responses in a word cloud, with the size of each word corresponding to their frequency in poll results. (During the pandemic, “competent” and “moderate” were leading adjectives used to describe the governor.) Nteta argues that by combining qualitative and quantitative methods, a more complete picture of public opinion emerges. “The overarching story we heard about Charlie Baker is that he’s a throwback to a New England Republican—socially liberal and fiscally conservative. He’s someone who voters feel really understands the state,” says Nteta. “We’re not just interested in whether Charlie Baker is up or down, but why. These methods help us get there.”

To get beyond respondents’ potential biases, Nteta and his team must first root out their own, ensuring that their questions don’t lead the respondent or betray an underlying agenda. Melinda Tarsi ’10MA, ’14PhD recalls a stringent vetting process when she worked on the poll as a doctoral student. She tells of long sessions parsing questions, debating with her colleagues how certain wording might show the writer’s personal preference or lead a respondent to a particular answer. “I learned to take into consideration the psychological aspects of taking a survey,” she says. “Even the ordering of questions can prime a respondent to think in a certain way.”

The lengths that Nteta and his team go to ensure that the UMass Poll maintains scholarly integrity reflect the kind of poll he and his colleagues wish to maintain. While many polls are designed to attract media attention, the objective of the UMass Poll is more aspirational. “If I have a normative commitment, it’s to democracy,” says Nteta. “There are fundamental questions facing our society that we need to debate: climate change, identity politics, and who we are as a nation. The UMass Poll is attempting to provide answers.”

The poll, administered digitally with the help of YouGov, a data-analytics firm, aims to report on the health of democracy in Massachusetts and the country at large. The story it has told in recent years is that of a nation struggling to reconcile its shared identity—and a commonwealth that might just hold the answers.

The State of the Union

In October 2011, the UMass Amherst Poll reported that 32% of respondents felt the country was headed in the right direction. In the same period, 43% of respondents felt the same about Massachusetts. Nine years later, nearly the same percentage of respondents felt positively about the national direction, but 7% more (50%) of respondents described the direction of the commonwealth as favorable. “This is a state that hasn’t embraced the polarized politics of the rest of the nation. In fact, recognizing the importance of various political viewpoints is important here,” said Nteta. “The story of Massachusetts has historically been about hope, and over the last two decades, the state has become one of the country’s beacons in that regard.”

Perhaps on no issue has the poll shown more divided public sentiment than that of race. In January 2022, the UMass Poll reported that, although three out of four Americans are in favor of teaching children about racial inequality, just over half believe that white Americans enjoy certain advantages because of their skin color. Forty-two% of respondents went so far as to say that racism is either “probably” or “definitely” not systemic in America.

However, in examining the numbers through a holistic lens, Rhodes finds reason for optimism. “When it comes to the high-profile cultural issues, there’s a lot more common ground than political elites demonstrate,” he says. “Most Americans support teaching the history of slavery; they don’t support the termination of abortion rights; they’re not opposed to participation of transgender people in sports. There’s a lot more room for compromise among the public than there is among elites.”

Graphic of 41 percent in black font overlaid on yellow dice.

Years after casinos were legalized, 41% of respondents felt the decision should be reversed.

Nteta also sees reason for hope in the data and historical context. “The nation has faced these challenges before, and I’m hopeful that we begin to recognize our shared threats and demand more from leadership,” he says. On this point, Nteta believes that the work he and his colleagues are doing will help others chart the way forward. “I see our poll as helping to improve civic debate, so that when people are making decisions, whether they’re private citizens or elected officials, they have access to good information,” he says.

Rhodes shares his colleague’s view that collecting reliable data on public sentiment is a critical component of a healthy democracy. However, he also emphasizes that this occasionally means reporting unpopular results. “We absolutely have a responsibility to report in an unbiased way, but that also means not pulling punches,” he says. “To censor your findings is itself a form of bias.” To that end, both Nteta and Rhodes underscore the need for unimpeachable polling methodologies, strong data, and a willingness to speak truth to power. As a torch bearer for that message, the UMass Poll has already succeeded with the next generation of political scientists.

Following her time at UMass, Tarsi went on to serve in several municipal positions in Massachusetts, as well as on statewide boards. In those settings, she found her experience working on the UMass Poll to be critical in building consensus. “Because of what I learned, I can provide decision makers with the data they need to make informed decisions for their town,” she says. “And I’m able to say, ‘Here are the survey results, and here’s how I did it. I’m passing this data to your citizens and appointed officials so you can make the best decisions possible.’ I love being able to provide that.”

Prepping the Pollsters

Beyond its status as an esteemed survey with national recognition, the UMass Poll also serves as a training ground for UMass students looking for real-world experience with quantitative data—as well as how to present results to the public. During his sophomore year, Anthony Rentsch ’18 got his feet wet proofreading survey questions and double-checking poll results, until then-director Brian Schaffner asked him to cowrite a blog post on early voting results in advance of the 2016 election. The blog post, entitled “Early Voting Predicts Who Wins,” was soon picked up by The Washington Post, giving Rentsch an early career win. “Especially as an undergraduate, having the opportunity to have my name in the Post was incredibly cool,” recalls Rentsch. “I left that one on my resume for a long time.”

Tarsi also points to her time with the poll as a definitive professional experience. As a political science professor at Bridgewater State University, she shares what she’s learned with the next generation of pollsters, pundits, and politicians. “I loved the war-room mentality we had at the UMass Poll, pitching survey questions to each other, reworking them until they were just right,” she recalls. “That’s something I carry forward with my students now, explaining that even where you place a comma, that level of detail makes a difference.”

Graphic of 53% in black font over purple outline of people.

Just over half believe that white Americans enjoy certain advantages because of their skin color.

Nteta and his team are keenly aware of the benefit in involving graduate and undergraduate students alike. He argues that the more diverse eyes he can put on both survey questions and results, the greater integrity the poll will have. Of course, the benefits run both ways. “We don’t just want to bring students in to talk about poll results, but hopefully to get publications out of their work here,” he says. “We recently wrote an article on [Washington,] D.C. statehood in which one of our graduate students played a key role in working on the data analysis and writing the paper. If we can get this thing published, that goes right on his CV for the job market.” Nteta and his team are also at work developing courses and learning opportunities that allow more undergraduate students to be hands-on with poll data.

Increasingly, the UMass Poll has shown up in media outlets across the state and country, with results cited by MSNBC, CNN, The Boston Globe, and even in a tweet from former President Donald Trump. For Nteta and the UMass polling team, notoriety brings increased responsibility—and they take their charge seriously. “As a poll, we can’t create the change, but we can damn well inform the conversations that are happening around the dinner table, on Beacon Hill, or in the Supreme Court,” says Nteta. “It’s our job to inform the public of what they currently think or believe, so people don’t feel alone in their support or opposition to an issue. That’s what this poll is all about.”

On the Other Hand

Despite its Puritanical origin, Massachusetts often stands at the leading edge of contemporary issues facing the nation. In recent years, the UMass poll has taken residents’ temperature on the hot-button questions of legalized marijuana and gambling. The story the survey has told says much about how public opinion can be bolstered—or lost.

In the case of marijuana legalization, residents’ support has only grown over time. Prior to the legalization of medical marijuana use in 2013, 68% of those polled voiced support. Four years later, 53% of respondents agreed that recreational marijuana should also be made legal for adults, a measure that the state legislature passed in 2016. In a November 2021 survey, 61% suggested that legalization had been positive for the state, with just 13% believing the impact to be negative. “Implementation has not been controversial,” acknowledged Nteta. “Many communities around the state are seeing positive financial benefits with few negative consequences—and that’s showing up in our data,” he says.

The story of legalized gambling in Massachusetts offers a very different case study in public opinion. Several years after the legalization of casinos in 2011, 41% of survey respondents felt the legislation should be reversed, with 48% believing gambling should remain legal. In the fall of 2021, just 34% supported expanding gambling to include sports betting, with 26% in opposition. But the real story lies between those numbers: In the same 2021 survey, 40% of respondents indicated that they neither supported nor opposed the legalization of sports betting. According to Nteta, their uncertainty indicates a communication failure on both sides of the issue. “When you see this level of ambivalence, it suggests that not enough information has been communicated to the public as to why they should support or oppose an issue,” he explains. “However, there’s reason for celebration on both sides, because those who are ambivalent are there for the taking. The question then becomes, who’s going to take them?”