America in One Room

America in One Room was one of the most significant political experiments in U.S. history. For the first time ever, a scientifically accurate microcosm of the entire country was brought together to deliberate on and address the critical issues facing the nation.

A stratified random sample of 523 registered voters, recruited by NORC at the University of Chicago, was identified out of millions of Americans through a rigorous scientific process in order to represent the political, cultural, and demographic diversity of the American electorate.

Helena brought each American to a single location — a resort in Dallas, Texas. For four days, the delegates participated in an unprecedented deliberation on the critical issues facing the United States.

In a process called “deliberative democracy polling”, the 523 participants began by taking an in-depth questionnaire, stating their views on five key issue areas identified in earlier polling as most vital to the 2020 presidential election.

The group was then given a 60-page briefing document, prepared and vetted by a distinguished group of policy experts from across the political spectrum. The document was balanced with arguments for and against specific policy proposals across the five issue areas.

During the ensuing 4 days, the America in One Room group vigorously and respectfully deliberated on these polarizing issues with one another.

Their discussions took place in two formats: moderated, representative small group settings and evening plenary sessions, where the group convened as a single unit to ask live questions to 2020 presidential candidates and a bipartisan selection of experts and political figures ranging from the former White House Chief of Staff and former leaders of the National Security Council, Congressional Budget Office and White Economic Policy advisory.

At the end of the final day, the group took the original questionnaire again.

The outcome was a landmark data set that represented the “will of the people” — what Americans think about values, candidates, and policy issues when given the chance to think deeply, engage with different opinions, and deliberate in a fact-rich and respectful environment.

The results showed dramatic changes in opinion. The most polarizing policy proposals, whether from the left or the right, generally lost support, and a number of more centrist proposals moved to the foreground. Crucially, proposals that were further on the right typically lost support from Republicans and proposals that were further on the left typically lost support from Democrats.

Helena and Stanford published the project’s data publicly.

What ensued was a worldwide response that included a special report from the New York Times and global leaders including Presidents Barack Obama and Bill Clinton citing the project as a needed mechanism to mend structural problems in modern democracy.

A year later, on election day of 2020, the New York Times released a second report on America in One Room. Revisiting and re-polling the 523 person sample, researchers demonstrated that the project enacted enduring changes in the sample’s political values.

We believe that deliberative polling is a mechanism to address some of the structural problems facing not just the United States but democracies globally.

Helena is working to realize a future in which this process becomes more fundamental to how democratic nations consider and make decisions.

America in One Room was born from a mistake.

On December 3rd, 2017, Helena Member Nicolas Berggruen was hosting the ceremony for the annual Berggruen Prize for Philosophy and Culture at the New York Public Library. In attendance was Helena CEO Henry Elkus as well as James Fishkin, Professor of Communication and Professor of Political Science at Stanford University, the Director of Stanford’s Center for Deliberative Democracy, and a Helena Member (though not at the time).

The problem for Professor Fishkin was that his name was nowhere to be found on the guest list. Serendipitously, the person who was supposed to occupy the seat next to Henry never made it to the event, so Fishkin, having shown the apologetic staff his personal invitation from Berggruen himself, was given that chair, and he and Henry began talking.

In 1988, Fishkin originated the concept of what he called Deliberative Polling. The germ for it came from a book by then-Princeton professor Larry Bartels titled Presidential Primaries and the Dynamics of Public Choice. Fishkin was persuaded by Bartels assertion that, according to Fishkin, our primary system is “unrepresentative and unthoughtful. It’s arbitrary in sequence of states, and it’s a kind of sound bite campaign where nobody can really think in any depth.”

Fishkin, who has Ph.Ds in both Political Science and Philosophy, then tried to think of ways to fix it. His solution became the Deliberative Poll.

The methodology for a Deliberative Poll is as follows:

— A questionnaire is given to a randomly selected, representative sample of a particular electorate. The questionnaire can focus on any number of issues, and the electorate represented can be of any size. The key is that the sample represents the electorate accurately; the more demographics taken into account, the better. Theoretically, this baseline is indicative of the general opinions of the country as a whole.

— A subset, itself also representative, is then recruited to be brought together in one location. The rest of the original sample is the control group.
Those participating in the event are given balanced, non-partisan briefing materials on whatever issue(s) they were polled on. These materials are comprehensive and vetted by experts of every ideology to ensure accuracy.

— The participants arrive at the event, where they are split into small discussion groups with trained moderators. In these groups the participants discuss the issue(s) among themselves and construct questions to be posed in a town-hall setting (called plenary sessions) to participating experts and policy-makers.

— The event concludes with a final questionnaire that is identical to the first. Theoretically, this final poll is indicative of what the country’s general opinions would be, were the entire electorate informed and engaged. It should be noted, too, that media coverage of the event is absolutely necessary, since the goal of a Deliberative Poll is to have a consequential impact on policy, and the only way to do that is through attention and visibility.

At the heart of deliberative polling is the belief that, with enough open and honest communication, it is possible for even the most ideologically disparate people to engage with each other respectfully and to find common ground.

Its scientifically-validated process gives participants the time, space, resources, and conditions to think deeply on issues. The result is perhaps the most nuanced and accurate view of what can be called “the will of the people.”

A few months after the Berggruen event, Fishkin emailed Helena a white paper he had written titled “Proposing a 2020 Election Reform: The National Caucus.” In that paper, he detailed his ambition to execute a Deliberative Poll in the leadup to the primary process of the 2020 presidential election, and was looking for Helena to put it together.

His proposal was for the largest Deliberative Poll in United States history, populated with a sample that would represent the entire United States electorate, that would engage with society’s most fraught, divisive issues.

For the event to be a success, the list of requirements was extensive.

For the science, Helena would need to find a delegation that was representative of the whole of the American electorate; to produce unbiased, curated briefing materials on the most complex issues in the public discourse; and to recruit a bipartisan advisory committee of scholars and policy experts to vet the materials and engage with the delegates in the plenary sessions. But none of that would have an impact if it wouldn’t get seen. Just as important were acquiring the media partners covering to cover the event and the messaging expertise to give the event traction and momentum leading up to it. And, of course, the capital to put the whole thing on.

But the potential benefits were immense. Scientifically, it would provide the most accurate, insightful data both on the country’s present opinion and on where that opinion might shift given more education and information—data that was particular valuable given that it was leading into an election year.

And by executing it on the scale and with the visibility Fishkin envisioned, it would be seen worldwide, offering a way to actively fight against political polarization and partisanship, to promote unity, civility, and thoughtful democratic discourse. A national Deliberative Poll could help inform politics on every level, from state and local elections to international governmental policy.

Helena agreed to make the event happen, and America in One Room—at the time called the National Caucus—commenced.

The first thing Helena did upon agreeing to pursue the National Caucus was set up meetings to secure the capital. Luckily, this was completed in just a matter of weeks, since what the project proposed — a scientifically-validated way to reinvigorate the primary process and promote cohesiveness and collaboration — was sorely needed and would provide significant, tangible benefit to the current political environment.

From there, the Helena team organized weekly calls to stay on top of the project. The calls started small but grew as different teams were brought in to handle different functions and lasted more than a year, from winter of 2018 through the event in September of 2019. Originally, there was Helena; the Stanford team, led by Fishkin and his colleague, Helena Member Larry Diamond, senior fellow at the Hoover Institution and at the Freeman Spogli Institute for International Studies (FSI) at Stanford University; and Dan Werner of By The People Productions, who had collaborated with Fishkin on past Deliberative Poll productions.

We thought finding the venue would be tricky.

It would need to comfortably host 500 people, with enough individual meeting rooms for all the small discussion groups, as well as a hall large enough for everyone to come together, and it would ideally be located in roughly the middle of the country, to remain as fair as possible for all delegates traveling. But it took little time to find the perfect location: the Gaylord Texan Resort and Convention Center in Grapevine, Texas.

A sprawling property with 1,800 rooms, almost half a million square feet of meeting space, plenty of large halls, and located minutes from the Dallas-Fort Worth Airport, the Gaylord Texan Resort checked all the boxes. It also had every amenity—from 10 on-site restaurants (though food was provided) to a 4,000 sq. ft. fitness center with private jogging trail—to make the weekend as pleasant for the delegation as possible.

To handle recruiting the 500-person representative sample—the participants in the Deliberative Poll, who were called delegates—as well as the 800-person representative control group, Helena partnered with NORC at the University of Chicago, an objective, non-partisan research institution that is the most thorough and credible organization for census taking and tracking in the country.

NORC broke down the nation’s electorate with respect to age, gender, race and ethnicity, educational attainment, political party identification, geography, and other dimensions that corresponded with the American registered voter population benchmarks from the November 2018 Current Population Survey or the 2018 General Social Survey. Then, once the electorate had been broken down and the relative proportions for each category identified, NORC used stratified random sampling to randomly select the participants from the groups of people that fit each category. This way, NORC ensured that the participants and control groups were both perfect microcosms of the electorate as a whole. Along with hotel accommodations, delegates were given airfare, meals, and an honorarium for their time.

The team confirmed that the five issues the delegates would be deliberating on were the economy, the environment, foreign policy, healthcare, and immigration. In order to aid in their deliberations, each delegate was sent a briefing booklet with information on each of these five issues, as well as common points of contention, discussion, and arguments for and against. It was imperative that these documents be as robust, accurate, and balanced as possible, so Helena recruited leading political and academic minds on both sides of the aisle to ensure veracity and impartiality in all materials created and distributed.

This group was called the Advisory Council, and it was populated by 15 members from across political and ideological spectra.

In the lead-up to the event, the delegates and control group were both given the initial questionnaires, then the briefing materials were sent to the delegates for them to review beforehand.

For the findings and results of a Deliberative Poll to be consequential, they need to be seen and understood by the public. In the lead-up to the event, Helena worked to establish media partnerships and to develop a publicity strategy that would generate momentum before the event, faithfully record during it, and provide results and analysis post-event that were both accurate and widely disseminated.

On September 19, 2019, 523 delegates arrived at the Gaylord Texan Resort and Convention Center.

They received their schedule, another copy of their briefing materials, a badge with just their first name and hometown, and then got in line to have the portrait taken by The New York Times. There was a welcome ceremony — at which Fishkin, Diamond and Elkus spoke — dinner, and the proceedings began.

Over the next three days, the delegates’ schedules were packed with a combination of Small Group Sessions, Plenary Sessions with issue experts, and Plenary Sessions with candidates.

Small Group Sessions: There were seven small group sessions over the course of the weekend: one introductory meeting, one final meeting at which delegates shared their reflections and took their final surveys, and five discussion meetings, one for each issue. Each group was between 11 and 15 people, from all different backgrounds, and each had a trained moderator.

The issue meetings were 90 minutes long. They would open with a brief summary video on the issue to be discussed, and then discussion and deliberation would begin. Delegates had their briefing booklets, so arguments were informed and fact-rich. The small group sessions concluded with each group collectively crafting questions to ask the speakers in the plenary sessions, which incentivized cooperation, as well as a deep understanding of the issues in order to ask good questions.

Plenary Sessions: There were five plenary sessions with issue experts over the course of the event—one for every issue, just like the small group sessions. Each session was hosted by Ray Suarez, former host of Inside Story on Al Jazeera America and current John J. McCloy Visiting Professor of American Studies at Amherst College, and featured two (or three) experts on the given issue, each representing a different point of view.

The delegates had opportunities to interact directly with the speakers through the questions they formulated in their small group sessions. The experts ranged from former White House Chief of Staff Denis McDonough to Judy Feder, Professor of Public Policy at Georgetown University and former Dean of the Georgetown Public Policy Institute.

The final responsibility of the delegates before departing was to re-take the original survey, just as they had done months before.

There were 9 policy questions on immigration; 8 on the environment; 10 each on healthcare, the economy, and foreign policy; there was also a series of more ideological questions, such as whether taxes should be kept low, whether public officials care about what the public thinks, and whether their own political opinions were worth listening to.

The results of America in One Room showed that American voters, regardless of political ideology or any of the other demographic breakdowns NORC used, are capable of keeping open minds and trying to find concordance on even the most divisive of issues. Across the board, delegates tended to soften their more extreme views, with the most polarizing proposals on both sides generally losing support, and the more moderate ones gaining it.

Immigration

The debate on immigration included several proposals that elicited a more welcoming position on both legal and illegal immigration. This change was mostly due to a softening of the opposition expressed by Republicans in their initial surveys.

The Economy

On certain economic issues, Democratic support dropped for potentially expensive government programs such as increasing the federal minimum wage, guaranteeing universal basic income and covering the cost of college tuition at public universities for all students who could not otherwise afford it. Meanwhile, support for expanding the Earned Income Tax Credit to more middle-class workers increased, especially among Republicans and Independents.

Health Care

Similar to what was observed in the economic proposals, support dropped among Democrats for potentially expensive healthcare programs such as Medicare for All, while support for some more centrist proposals–such as increasing the federal subsidies in the Affordable Care Act that help the poor, expanding these federal subsidies to the middle class, and ensuring that people should have reasonable access to health insurance regardless of pre-existing conditions–all rose across party lines. Support for repealing the Affordable Care Act (Obamacare), a conservative proposal, dropped among Republicans.

Foreign Policy

On foreign policy, Republicans increasingly supported a number of internationalist proposals, including the US rejoining the Trans-Pacific Partnership, reaffirming its commitment to defend any NATO ally attacked by a hostile force, and recommitting to the Iran Nuclear Agreement. Delegates overall also reaffirmed the importance of soft power by increasing support for using diplomacy and financial support to promote democracy and human rights.

The Environment

On the environment, delegates increased their majority support for rejoining the Paris Agreement, and even for exceeding the cutbacks in greenhouse gas emissions the Paris Agreement sets forth. In both cases, the increases were primarily from Republicans. Reducing greenhouse gas emissions using taxes and market incentives saw the strongest post-event support; however, the goal of requiring zero carbon emissions for cars, trucks, and buses lost support, especially among Democrats.

The results and story spread across the international press.

America in One Room was on the front page of the October 13th print edition of The New York Times, with a feature article accompanied by a spread of every portrait taken of the delegates. It is an arresting image; the journalist Emily Badger, the article’s author, describes the collection: “Together, their faces are a portrait of the American voter.”

It was featured in numerous publications both local and national, from the BBC, USA Today and the Christian Science Monitor to Vanity Fair, Salon, and FiveThirtyEight. It was the focus of multiple CNN segments, as well as an eight-part Snapchat video series. President Barack Obama tweeted a link to the Times piece with the caption “Here’s an interesting read—a reminder that behind every opinion lies a human being with real experiences and a story to tell. Sometimes we’ll agree and sometimes we won’t, but if we want our democracy work, listening to each other isn’t optional.”

What seemed to resonate even more strongly than the quantifiable significance of America in One Room was the emotional significance. Delegates returned home more inspired, more open-minded, and more motivated to actively participate in the democratic process, both local and national.

The data shows how people’s minds can be changed, but the bar graph does not encapsulate how meaningful it is to hear Beverly, a 74-year-old retiree from North Carolina call the experience “life-changing,” or Lauren, who works in healthcare administration in Florida, describing returning home with “a sense of confidence,” saying that the event “was a turning point in my adult life to feel like a more educated and engaged citizen. And that’s been very special for me.” America in One Room was proof that a country’s electorate can engage in thoughtful, civil, substantive discussions on any issue, no matter how polarizing. The indelible image of the event was one of empathy, its takeaway one of hope.

The America in One Room Project was not designed to be a one-off event; the weekend in Dallas was a way to spread Deliberative Polling, to legitimize and publicize it as a viable and worthwhile social technology. The event succeeded in this aim; it generated a swell of interest in deliberative democracy domestically and internationally. Prior to America in One Room, the Center for Deliberative Democracy at Stanford University had conducted one notable Deliberative Poll since November of 2017. Since A1R, it has conducted or announced nine of varying sizes. The two most significant of which were in Iceland and Chile.

In Iceland, the Poll was called Deliberative Poll-Public Consultation on the Revision of the Icelandic Constitution, and it was conducted by the Social Science Research Institute at the University of Iceland in collaboration with the Grant of Excellence project Democratic Constitution Making and the Center for Deliberative Democracy in November 2019. With revisions to the Icelandic Constitution impending, the event was focused on Iceland’s constitutional amendment process, with results brought to Prime Minister Katrín Jakobsdóttir and the Parliament to inform their subsequent decision-making.

There were six issues under consideration: the office of President, referendums, the Court of Impeachment and Parliament’s power of indictment, amendments to the Constitution and international cooperation. As befits a national Poll, the surveyed delegation was extensive; there were 233 people in the attending sample, with roughly ten times that number answering the questionnaire.

In Chile, in the wake of months of political demonstrations, plans were laid to conduct a 400-person Deliberative Poll, called Deliberación País, to address redrafting the current constitution, which was written in the transition period after Pinochet’s dictatorship. The Chilean senate committees will use the Deliberative Poll results, in conjunction with those from another poll conducted in December of 2019, to help form the terms of the new constitution.

The Times Followup

On the one-year anniversary of the event, the Center for Deliberative Democracy and NORC conducted a follow-up survey with the participants of America in One Room—both the delegates and the control group—to see what kind of lasting impact the event might have had (if any). The survey was also covered in a New York Times feature.

There were statistically-significant differences in opinion between the two groups, most notably in how they rated the government’s response to the pandemic (the delegates rated it worse), how they rated the government’s response to the public health effects (the delegates rated it worse, how they rated the government’s response to the economic effects (the delegates rated it worse), and how likely they were to vote for Trump (the delegates were much less likely).

In perhaps the most telling statistic, the delegates, too, were much more likely to pay close attention to the campaigns. The Times writes: “So many of these voters one year later said they were still clinging to the memory of sitting together, disagreeing but not disparaging one another.”

Fishkin is quoted as saying “for me, it’s proof of concept in that it shows a long-term deliberative effect is possible.”

The paper demonstrated that participants moved toward the center across 22 of 26 policy proposals identified as extremely partisan. In 19 of these, the results were statistically significant. In addition, America in One Room  had a considerable impact on the affective polarization of its participants. While partisans exhibited little change in assessments of their own party, their positive ratings of the opposing party increased dramatically. (13 points in the case of Democrats ratings of Republicans, and 14 points in the opposite scenario.) This effect was even more staggering among participants whose issue-positions were more extreme before deliberation.

In September of 2021, we conducted a follow up to America in One Room, this time focused entirely on climate change. A1R: Climate and Energy was convened online, aided by a Stanford-developed proprietary AI moderator. The event was the largest Deliberative Poll in history. 962 Americans, again recruited by NORC to represent a statistically accurate microcosm of the voting electorate, came together to discuss Net Zero policy pathways in multiple sessions over the course of 6 days.

As with the original A1R, participants received comprehensive, balanced briefing materials on the issues, vetted by a bipartisan Advisory Council that included scientists, policymakers, activists, and stakeholders within the private energy sector. They met in 104 small discussion groups where they debated dozens of policy proposals and formulated nuanced questions for a diverse panel of experts.

As world leaders gathered in Glasgow for the UN Climate Change Conference (COP 26), A1R: Climate and Energy produced a comprehensive picture of where Americans stand on climate change when given the opportunity to think deeply about the issues in a fact-rich and respectful environment. Participants post deliberation demonstrated strong bipartisan consensus in the direction of doing more to combat climate change — positions on nearly all substantive policy questions (66 of 72) shifted in support of a wide range of actions to substantially reduce greenhouse gas emissions. (No such shift was evident in a control group of 671 Americans who did not engage in the deliberative process.) The full results can be viewed here.

A1R: Climate and Energy was proof of concept that the deliberative effect can be scaled to many more Americans through repeated online convenings over time and testifies (against a backdrop of divisive digital interactions and information insecurity) to the power of AI to unite, rather than divide us.