Heather Cox Richardson:

From CAFE and the Vox Media Podcast Network, this is Now & Then. I’m Heather Cox Richardson.

Joanne Freeman:

I’m Joanne Freeman. Today, we’re going to be talking about a topic that probably is in the front of everyone’s mind at the moment because this episode, although we’re recording it a few days before the election, this episode will be dropping on the morning of the election. So what we want to talk about today is something that certainly has been talked about a lot in the weeks leading up to this election and that we’ll all be thinking about in one way or another on the day of the election, and that is polling. There have been certainly poll after poll after poll on all sides, reporting all kinds of things with everyone claiming, of course, that their poll is front and center, the most accurate.

What we want to talk about today is, first of all, why all the polling because actually, the foundation of the reason for it goes all the way back in American history. It has to do with the foundations of American democracy, and then we want to talk a little bit about how polling has worked over time, how it’s changed, and then of course, a little bit about what that can tell us about polling today and about how that may or may not reflect on the election that you will, of course, all be going to vote in after or just before listening to this episode.

Heather Cox Richardson:

One of the things that’s really important to remember about the polling that we see in the news today is that different polling organizations do their polls in very different ways, and that affects what they find in those polls. They use very different methodologies. Some use live interviewers. Some have panels that can opt into opinion polls. Some use online polls. There’s all sorts of different ways that people do their polls, which is one of the things that has an effect on how it comes out, but also on how accurate we consider that data.

It’s also important to remember that with the rise of technology, the barriers to becoming pollsters have dropped dramatically. While that can be a really good thing because it’s more democratic and people are perhaps listening to voices that they wouldn’t have otherwise, it also means that pretty much anybody who can get their hands on a few thousand bucks can become a pollster. What that’s done is it has created a number of polls that are a lot less valuable than the ones we used to have when there was a higher barrier to entry. So we have both good and bad coming from that.

Joanne Freeman:

I want to add to that though, Heather, a point that we’ve made on a number of episodes, and that is you just mentioned technology. The fact is, in so many ways, technology shapes democracy. But if democracy, in one way or another, is a conversation between us, the American public, people who supposedly have the power to give power and the people with power, anything that changes the way those people communicate with each other, us and them, is going to fundamentally shape democracy.

Heather Cox Richardson:

Well, now, I’m going to throw that back to you and suggest that you said when we were first talking about doing this episode that fundamentally, democracy is about translating the will of the people into government. It’s a real problem from the very beginning to figure out what the people want or what they’re thinking. So why don’t we start at the beginning?

Joanne Freeman:

Starting at the beginning. What’s interesting, and I’ve always found this really fascinating, is the founding generation understood and talked all the time about the fact that what differentiates a democratic republic from a monarchy is the central importance of public opinion. They talked about public opinion all the time. They talked about its importance. They talked about why they needed to know it. They talked about how if they didn’t know it, the government might not function the way it was supposed to. But even as they were saying again, and again, and again public opinion is central to the form of government we’ve instituted in the United States, they really weren’t sure who the public was or how to get their opinion. Right? Those are the things we take for granted now with polls, but put yourself back in 1792, 1793. How in the world can you tell what people are thinking?

Heather Cox Richardson:

There is an important aside there that is going to matter in the present, and that is the idea of who the public is. Of course, the framers and the people who are trying to figure out public opinion in the early republic were not interested particularly in what women thought about politics, I assume, and they sure were not interested in what Indigenous Americans or Black Americans thought.

Joanne Freeman:

Right.

Heather Cox Richardson:

That idea that we have people who are part of the American polity, but their opinions don’t need to translate into electoral politics is going to be really important because I’m betting that the people I just mentioned had really different opinions about the way the country should go than the people who were actually going to vote.

Joanne Freeman:

You think?

Heather Cox Richardson:

So, anyway, 1792.

Joanne Freeman:

Just one example of several of people in the founding generation wanting to know what the public thinks about something, and then trying to figure out in their weird early American ways what public opinion is. In this case, in 1792, apparently George Washington, President of the United States, wasn’t sure if he should run for a second term, and so he sent his personal secretary, Tobias Lear, out particularly into the North, into New England rather than the South where Washington is from to find out if public opinion wanted him to stick around for a second term.

Probably what Lear did was wander around chatting with people because Thomas Jefferson has a conversation with Washington, and according to Jefferson who noted down memos about this conversation, he said that Washington, and this is now a quote from Jefferson, “Had desired Mr. Lear to find out from conversations without appearing to make the inquiry whether any other person would be desired by anybody.” Meaning, for president. “He had informed him. He judged from conversations that it was the universal desire that he,” Washington, “should continue.” But then, Lear went on to say, “This is only from the North, and it may be different from the South.” I’m sure that thrilled George Washington, but the point there is Washington says to Lear, “Go mingle with people in the North, in New England. See what they think. Don’t make it obvious that you’re asking. Put leading comments in your statements, and see what they say, and then report back to me.” That was typical.

Heather Cox Richardson:

“Yo, what do you think of George?”

Joanne Freeman:

I know. Exactly. “I’ll just stand here with a pen and paper.”

Heather Cox Richardson:

“I’m not a spy. Don’t mind me.”

Joanne Freeman:

A similar occasion. There’s an occasion in which a South Carolina congressman, William Loughton Smith, he wants to find out how Federalist, meaning Hamiltonian-Federalist, South Carolina is. The way that he does this is he goes into the southern backcountry, places where you might not know public opinion, and he knocks on farmhouse doors, and chats with people, and looks to see which newspaper is on the fireplace. At this point, you had a very Republican, Jeffersonian-Republican National Gazette and Federalist Gazette of the United States. So he’s looking to see, are they reading one or the other?

Apparently, if people were reading the Democratic-Republican newspaper, he exchanged it, told them what a horrible newspaper they’re reading, “What a rag. You should take this one. It’s a much better paper.” But even as he’s chatting with people, he’s trying to get a sense of how Federalist the backcountry is, and he’s trying to convert some of the people who he meets who don’t seem to be on the Federalist trail, again, by chatting with them, by conversation. Not seemingly by being upfront and open about what he’s doing, but just being someone who’s chatting, interested in politics of the backcountry.

Madison, James Madison does something similar in which he wants to know what people in Virginia think about Hamilton’s financial plan and the Federalist agenda. He sends people out to engage with public in Virginia to see what they think, to sit in taverns and overhear conversation, to just mingle around and get a sense of what’s out there. In this case, when word comes back that in Virginia, people were generally comfortable with the Federalist agenda, Madison says, “Essentially, wrong public.” He says that information was probably “collected from tainted sources, which ought to have suggested to a cautious and unbiased mind the danger of confiding in them. You talked to the wrong people, and they gave you the wrong opinion.”

So, think about this generally. What I’m describing here is national politicians, the president, leading congressman, the secretary of the treasury. They just want to know what the public thinks of what they’re doing. They know even more broadly that in a democratic republic, that’s supposed to lead the way, and the only way that they can figure out how to do that is to sit in taverns, look at newspaper titles, mingle with the public. Again, really informal, seemingly haphazard, and erratic, but in their mind, vitally important and worth doing nonetheless.

Heather Cox Richardson:

Well, and it is vitally worth doing in many ways if you think about the things that are not captured nowadays in the polls, but turn out to be enormously important, and things even in the past where people following politics have missed entire movements that you could have seen if you were watching things on the ground. The problem is who’s good at having their ear to the ground, and who is not? Because people like me, I couldn’t possibly go in there and figure that out.

Joanne Freeman:

Well, and also, some of what we’re talking about here are communities. They have their ear to the ground, and they’re wandering around in communities of people where probably there is more accuracy if you’re wandering around getting a sense of what people feel versus contacting people, mass numbers of people on phones, and anything else, which we’ll talk about shortly. I hadn’t thought about it in this way before, but it doesn’t surprise me that if you’re wandering around in a community getting a sense of what people feel, depending on the circumstances, that actually might be something that’s remarkably accurate.

Heather Cox Richardson:

Although in those days, they also had the advantage of knowing exactly how many people were going to vote in a town. There was less instability on that front than there is today because you simply counted as many people that were there on your side of the slate, even if they didn’t show up to vote in a lot of cases. So you could say, “Oh, we know we’re going to get 68 votes here because we’ve got 68 democrats in this town. Even if John Smith is laid up in bed with gout that day, he’s still going to be voting because we’re the ones in charge of the poll lists.”

Joanne Freeman:

Those are the days where, particularly in that way, these local elections can be won or lost by one vote, or two votes, or three votes. It’s not infrequent that it boils down to that when you’re talking about small populations of people, local areas voting and smaller numbers than we deal with today. So you have people wandering around in this early period trying to figure out what people think.

Heather Cox Richardson:

Let’s hope it’s a little bit more directed than that, but okay.

Joanne Freeman:

I guess in my mind, it’s like William Loughton Smith, “Hey, there’s a farm now. Knock, knock, knock.” So, you’re right that it is a little bit more strategic than that, but still, in my mind, it’s these guys wandering around. It’s worth noting, and I find this really interesting, in 1824, the presidential election of 1824, some consider that the first time that in a presidential election, there was anything like polling that took place. In this case, 1824, what you have is the downfall of the Federalist versus Republican system. You have the Jeffersonian-Republican and now Democratic-Republican Party that’s just dividing up into factions, and you have a real problem with people predicting what way the election is going to go. Everything is scrambled, and this is the election where you end up basically having four candidates: Andrew Jackson, William Crawford, John Quincy Adams, and Henry Clay.

So people begin being really interested to see if there’s any way that they can tell who is going to win, and related to that, they obviously very much want to try and influence this election one way or another. So, a great example of this. There’s a newspaper that reports in 1824, “At a company muster held at Major William Watford’s Bertie County on the 17th July in the afternoon, it was proposed to take the sentiments of the company on the presidential question. When the vote stood as follows: Jackson, 102. Crawford, 30. Adams, one. Clay, Zero.”

Now, someone goes on to protest this and says, “In almost every captain’s company, the drums were beating and fifes whistling for the hero of New Orleans. The officers would treat their men, make them drunk, and then raise the war whoop for General Jackson. Then, the poor, staggering, drunken, diluted creatures would sally forth for the place pointed for them to vote. The result was always in favor of Jackson. I have conversed with some of them afterwards who told me they did not intend to vote that way at the proper election. They voted that way just to please their officers.”

So, here, you have that sort of thing happening all over the place and leading up to the 1824 election at tax gatherings, at July 4th celebrations. You have people leaving poll books out in public places for a few days at a time. If you happen to be there, you can sign in the book on who you think you’re going to support or not. Then, what would happen is partly, people are being informed by these kinds of give and takes, but equally important as the quote I just read a moment ago suggests, these reports are being published in the newspaper to shape the election, to shape public opinion. In the case of William Watford’s Bertie County, you see that, then you think, “Wow, Jackson is really in the lead,” and it makes things look very good for Jackson that he’s a really strong candidate. So, even at this early point 1824, they’re curious about what’s happening. They’re using polling in an informal way to try and chart what’s happening, but they’re very deliberately using these informal counts of votes to shape the election to come.

Heather Cox Richardson:

It’s called booming a candidate, making it look like there’s a huge rush behind one candidate or another.

Joanne Freeman:

Booming. There are so many good election words in earlier time periods. Booming is a wonderful one.

Heather Cox Richardson:

Yeah, yeah. Obviously, there is real room here in political polling for figuring out how to do things a little bit more scientifically, and that’s what we’re going to get in the 20th century. After World War I, when you start to see real attempts to organize, and to count, and to create statistics about things in America, there’s a man named George Gallup who enters the scene. He was born in Jefferson City, Iowa in 1901, and he attended the University of Iowa in Iowa City. Once he was there, he began managing The Daily Iowan college newspaper.

That sounds like I’m just giving piece by piece of a biography, but there’s a reason that that matters. In July of 1921, roaring ’20s, right, he published an editorial column called The Unattractive Women in which he wrote down a conversation that he claimed to have heard in a diner in Indiana. I’m kidding, but he claimed to have overheard between two college men who said that they needed to find wives who were not like the women on campus. They needed to find wives who were more than “a bone, a rag, and a hank of hair.”

Joanne Freeman:

Okay. You got to pause for a moment because we have to groan at that. When I was thinking about this and reading up for this episode, I read through that quote several times and every time went, “Ugh, come on.”

Heather Cox Richardson:

Well, it does suggest we should do an episode on what used to be considered insults, and that changed over time because in the 1920s, the ideal woman, of course, here is, he is suggesting, is not the flappers as they’re coming, the skinny women in their new flapper dresses.

Joanne Freeman:

A hank of hair with the short… the bobbed hair?

Heather Cox Richardson:

Yeah, with the bobbed hair, that they want their women with some meat on their bones, fancier clothes, and long hair. Okay. This episode is done. Well, the reason that this column matters is because Gallup’s column infuriated obviously the female students, but also the professors, and it also led to a huge spike in readership of the newspaper because he claimed he was showing people what the buried popular sentiment on the campus was, that this was his story of, “I know something you don’t know.” After he graduated, Gallup doubled down on the idea that he could find a way to create finger on the pulse summaries of the way people in Iowa were thinking. So he went on to get a degree in journalism at Iowa State, and he began to research different ways to poll public opinion and to do so differently than they had in the past.

So while in the 19th century, politicians who were interested in learning how people thought about political issues went out, and did straw polls, and talked to people, and got as much of a broad sense as they possibly could, Gallup decided that he could learn how people thought by using carefully selected samples that were reflective of the larger population. So, rather than counting everybody, I think of it as coins in a giant jar that you could count as many of those coins as you could or you could just take handfuls. If you count handfuls of those coins, and then figure out how many handfuls there are in the jar, your chances of being accurate are much better than if you actually just started to guess and said, “We’ve got this many thousands of coins in the jar, and I’m going to guess what they all are.”

Joanne Freeman:

That’s a great example. I’m smiling or looking quizzical because I was like, “That’s a great way of putting it.”

Heather Cox Richardson:

That’s how I learned statistics actually was from an older man here in town, and there was a yearly contest in his house where everybody had thrown all the coins into a jar, and if you guessed it… I forget what the award was. He won it every single time by doing that. Everybody else tried everything else, and he would be like… He’d stir it up, and he’d take a handful, and it would be $1.89. He’d stir it up and take another handful, and it would be $1.94 and whatever.

Joanne Freeman:

Wow.

Heather Cox Richardson:

He wanted every time. Anyway.

Joanne Freeman:

Wow.

Heather Cox Richardson:

That’s a good way to think about it. So he decided that he was going to use his selective sampling, and a lot of other people thought that this was completely ridiculous, that it made far more sense to get as many people and ask as many people what they thought than it did to ask a very few people and say, “Well, I have corrected for sample size and for making sure my sample reflects the population.” Gallup went on to start the American Institute of Public Opinion in Princeton, New Jersey where he began publishing a syndicated weekly newspaper column revealing what his polls had discovered. In those polls, he talked about retail goods. He talked about the way people thought about civil rights. He talked about geopolitics. So the Washington Post started running these columns, and they were so excited that they hired a dirigible balloon to fly over Washington DC to illustrate the fact that they had this column from Gallup. I mean, you think about the era, the 1930s. It seems like people are going to get a sense of what Americans are really thinking about stuff. They’re going to create a…

Joanne Freeman:

It would be exciting, right? Maybe.

Heather Cox Richardson:

Yeah, and a ton. You got radio now. You’ve got barriers breaking down. You’ve got the idea that…

Joanne Freeman:

Science.

Heather Cox Richardson:

You’ve got science. Science is going to make everything uniform, and it’s going to make everything understandable, and we’re going to be able to figure out exactly how people think. So we can create a president or hire a president who reflects that. Like Gallup’s detractors, still insisted that he was wrong, that there were other ways to get better numbers, including by getting as many numbers as you possibly could from people in straw polls, by calling telephone numbers, for example, with these new telephones, or by checking out people’s driver’s license numbers. In 1936, when this fight was going on, Gallup said, “No, you’re wrong, because if you focus on telephones and on cars, you’re going to be cutting out lower income voters, and lower income voters are far more likely to vote for FDR, the Democrat, than they’re going to be voting for the Kansas governor, Alf Landon, a Republican.”

So his competitor, the largest competitor, The Literary Digest, said in their straw polling that Alf Landon was going to win by a very large amount, and Gallup said, “Absolutely not. You’ve got it backward.” Gallup used his methods, a much smaller sample, and said just the opposite of what The Literary Digest said. The Literary Digest said that Landon would win 56% to 44% of the popular vote, while the Gallup poll said the opposite, that Roosevelt would win by the same margin. In fact, both of them were wrong. Roosevelt outperformed both of them. He got 62.5% of the popular vote, but the point was Gallup was absolutely on the map.

In 1940, he published his own manifesto called The Pulse of Democracy: The Public-Opinion Poll and How It Works, and what he argued here was that looking at polls enabled a democracy to do exactly what you were saying, Joanne, that it enabled politicians better to represent the people. “The practical value of the polls lies in the fact that they indicate the main trends of sentiment on issues about which elections often tell us nothing. The first stage of testing has demonstrated clearly that the polls can mirror the sentiment of large groups of individuals in concrete election situations. The second stage of practical application shows that the polls can also help to chart the main divisions of sentiment on issues, and so to make possible continuous measurements of public opinion.” Okay. So this sounds absolutely great. We’re all ready to go, except we have coming up here the election of 1948, and I’ve always loved what happens in the election of 1948 because, again, we should have this all settled. Right?

Joanne Freeman:

Science.

Heather Cox Richardson:

1940, we got this manifesto. Science. We’re going to know exactly what’s going to happen. The election of 1948 throws into this idea of scientific polling, the monkey wrench of actual sentiment. So what happens is, in 1948, the Republican governor of New York, Thomas E. Dewey, was widely expected to win the election over President Harry Truman, who was the incumbent. The pollsters almost all came out with numbers that said that Dewey was going to win. In September, the pollster for Fortune Magazine, a man named Elmo Roper told the New York Times that everybody had basically made up their mind. They made up their mind in favor of Dewey. He said, and this is a quote, “The man in the lead at the beginning of the campaign is the man who was the winner at the end of it. The winner, it appears, clenches his victory early in the race.”

Joanne Freeman:

Wow.

Heather Cox Richardson:

Part of that, I would argue, to go back to the earlier period, is that the newspapers were telling themselves the story they expected to hear. So newspapers in many of the major cities were leaning heavily toward the Republicans. Harry Truman was in real trouble in 1948 theoretically because he had started to advance desegregation and that they thought was unpopular enough that it was going to cost him the election. He however insisted on pulling out all the stops. He went on a 22,000-mile whistle-stop tour around the country. He gave speeches from the rear of the train, and crucially, he campaigned in Harlem. He doubled down on this idea that these were his policies and he was sticking with them, and rather than costing him support, it built his support. So, in fact, he ended up winning the election of 1948, leaving pollsters with complete egg on their faces.

Joanne Freeman:

One of the things that’s interesting about that is if you are in the business of tracking and making, in a sense, predictions, even though polls are supposed to be descriptive, in this case, you’re suggesting that they are forecasting what’s to come, you will see patterns that appear to be predictive. Right? So they see patterns, they’re used to collecting patterns, and they say, “Well, the pattern says this,” which makes perfect sense. But in this case, politics moves fast. This was why I laughed a moment ago when they said, “Well, if a man is leading at the beginning, he’ll be leading at the end.” Well, not in a democratic politics of the sort that we’re watching here. Truman took full advantage of that in many ways.

Heather Cox Richardson:

Well, and he was so interesting because he was breaking the pattern. He was saying, “I’m a Democrat. Nonetheless, I am going to start breaking down the barriers of segregation.” That was one of those things that pollsters looked at it and said, “This can’t possibly work. What on earth is he thinking?” Especially, and we have that very famous image from them, the Chicago Daily Tribune. The Chicago Daily Tribune had actually written the victory article about Dewey winning the election before the election was over, and it actually ran that article, did the printing of that newspaper with that very famous headline, “Dewey Defeats Truman,” and Truman held it over his head the next day laughing.

Joanne Freeman:

Big grin on his face, right?

Heather Cox Richardson:

Yeah.

Joanne Freeman:

It’s a famous photo. I think he’s standing in a car maybe holding up the newspaper, “Dewey Defeats Truman.” Just big grin.

Heather Cox Richardson:

Because he had won not by a little.

Joanne Freeman:

Right.

Heather Cox Richardson:

He had won by 49% to 45% in the popular vote, and by 303 to 189 in the electoral college. Now, other of those votes were peeled off by the Dixiecrats, led by Strom Thurmond, but the defeat of the pollsters paralleled that of the defeat of Dewey. They threw up their hands and said, “We don’t know what happened. We have absolutely no idea what happened.” What they discovered much later was that 14% of voters had actually made their final decisions in the last two weeks of the campaign, and they had broken 74% for Truman. So his gut sense was right. The scientific polls were wrong. His gut sense was right, and the errors of that campaign forced people like Gallup to rethink the way they did polling.

Gallup later said, “We permitted the public to get the impression that polls had reached a stage of absolute perfection. As someone said, we led the people to believe that we could walk on water, but we were not wholly unaware of this fact.” In fact, he later went on to say, “Polling is merely an instrument for gauging public opinion. When a president or any other leader pays attention to poll results, he is, in effect, paying attention to the views of the people. Any other interpretation is nonsense, including, I suspect, the idea that polls are predictive.”

Joanne Freeman:

When a president or any other leader pays attention to poll results, he is, in effect, paying attention to the views of the people. So there, Gallup is saying polls show you public opinion. Obviously, what he’s not necessarily addressing there, it’s not part of what he’s discussing, is the fact that that same process, that same polling reported publicly could then shape the views of the people, can then shape the outcomes of elections. I’m sure people listening can think of many examples of cases in which… just recently even, cases in which polling seems to be going one way, and then people decide because of the polling, “It seems like the contest is over, so we won’t vote because the contest is over.” All kinds of ways in which the polls lead voters to either storm up and really vote because they’re going to jump on the bandwagon or they’re going to overcome bad odds, or they feel so discouraged by what seemed to be runaway polls that they don’t show up or vote at all.

Heather Cox Richardson:

Well, and the other thing about the ’48 event that always sticks with me is the degree to which the pollsters found what they wanted to see, even if they didn’t realize they were doing that, but they really called that particular election off by 10 points, and they did so thinking that they were doing it completely by the book, completely without bias, and the candidate was baked in. It was over. They had made this a scientific study, and this is what was going to happen. I always wonder when you see that, and you see ’48, how many things that affect public opinion pollsters by their very nature because of who they are, the same way you and I would be biased in certain ways, don’t see entire movements. They simply do not see something that’s happening. It’s not the water that they swim in and they don’t see it. I know they’re supposed to be correcting for that, but I always wonder if they are.

Joanne Freeman:

Right. Even though it’s science… You have to have the science voice. Even though it’s science, the fact of the matter is it’s human sentiment, human will, last-minute actions, people’s responses. It’s things that are hard to measure.

Heather Cox Richardson:

Well, and it’s also hard to measure the difference between what somebody thinks and says, and whether or not it translates into votes. That’s not simply whether or not somebody actually shows up to vote because sometimes people say they’re going to and they don’t, but also, in elections like now today, the rules for voting in a number of states have changed so dramatically that people expect to vote, and then discover that their votes are not going to be counted, and we’ve also got, in this particular election, huge numbers of new voters, and what that will mean is not yet clear.

Joanne Freeman:

What we also have, not for the first time, but given the urgency in many ways of this election, is polls after polls, after polls often saying almost seemingly opposite things. “The Republicans are in the lead. It’s going to be a big red wave.” “Looking good for the Democrats. They appear to be ahead.” So, we’re at the point. I mean, I don’t know about you, Heather, but I’m just basically not paying attention to them at all.

Heather Cox Richardson:

I’m actually not paying attention to them to say anything about the results of the upcoming election, but rather I think that current day political polling has entered a different phase. So we did the, “Let’s walk around, and chat with people in barrooms, and see what they think,” to, “Hey, we can do this all scientifically by using polls.” Now, in this era, from about 1980 on, 1970s on, we have the use of polls first to attract public opinion for politics, which really starts in about the Carter presidency, but then in an era in which information has been weaponized and disinformation has been weaponized to make people act in certain ways politically. I am looking at the polling the way it’s being done today not as an accurate reflection of what people are actually thinking, but rather as an attempt to manipulate the electorate into thinking that there is something going to happen in order to make it a self-fulfilling prophecy.

Joanne Freeman:

In that, we agree. I’m not looking at it in any predictive sense, but what is interesting right along the lines of what you just said is the way polls are being weaponized and also the way that technology is making that easy to do. Right? It’s no longer relying on print media. It’s no longer are people reading the newspaper at all. It’s no longer relying actually on the news on TV or radio. In one way or another, for most people, it’s some kind of social media platform, and right now, while Twitter still exists, you see on Twitter a great example of these polls bouncing back and forth, and people asserting them and oversaying that they’re good, they’re bad, but they’re being used almost as political electoral missiles that are being launched out into the public to sway opinion regardless of accuracy or not. They’re weapons. You said weaponized, and that’s precisely what they are. In that sense, this is something very different from science trying to help us somehow magically figure out what people are thinking. Now, it becomes not so much the gathering of information, but the deployment of information as a strategic electoral bomb. That’s what we’re seeing a lot of in this election.

Heather Cox Richardson:

You’re hearing those complaints, especially from progressives, notably Simon Rosenberg, who has been following on Twitter the rise of what he says are ferocious Republican efforts to, as he said, flood the zone with their polls, game the averages, declare the election is tipping to them with the idea of depressing democratic voter turnout and what is turning out so far while we are taping this to be an election that has extraordinarily high turnout. So there certainly are people who are saying that there’s real reason to be very skeptical of polls nowadays because they’re not so much reflecting public opinion as trying to shape it.

Joanne Freeman:

The idea of a poll, regardless of the reality of the strategic deployment of polling, the idea remains. Right? Science is reflecting public opinion, and you and I, Heather, might say that personally, we’re not using these polls in any kind of a predictive way, but people tossing them out as electoral bombs are betting on the fact that people will understand and assume, “Oh, polls are actual information, and they’re predictive,” and so this matters. So, in a sense, it’s a legacy of Gallup, right? “Look what we can do,” in the banners in the sky, “Look what we can do. We can predict,” that in some ways, there’s a hangover of assumption about the accuracy and predictability of what’s coming out of polls.

Heather Cox Richardson:

So I’m going to push that a little bit further though in the… again, looking at polls not just as how do we tell about popular opinion, but what happens once you start to use information as a way to shape people’s expectations of outcomes and say what comes next if, in fact, you are willing to take the assumption that these polls are used, and it is an assumption. It’s worth saying that this is… I’m playing out this idea here, and we don’t know. We simply do not know how this election is going to turn out, but why would one want to build among the population an expectation that something is going to happen that isn’t?

That, to me, is a really important new phase of our politics, the idea that we’re setting you up to believe, for example, in a red wave, and what’s going to happen when you don’t get it? That’s one of the reasons I’m following polling as closely as I am is because to me, it looks like a continuation of the language that former President Trump used in 2020 to suggest he was going to win an election where his polling numbers sucked from the beginning. Those polling numbers never looked good, but he managed to convince a lot of people that he was going to win, and they decided to take matters into their own hands. So I’m wondering the degree to which the current polling is simply a continuation of that political strategy.

Joanne Freeman:

Well, it’s setting a stage so that people can say, “The polling had us way ahead, so there must be fraud somewhere.” Right? So there’s a case not just of tossing polls out into the world to drive people to the polls, to voting places, or drive them away from polling places. This is a setup for the response to the election. Right? That’s a different thing entirely. We’ve been talking about polling shaping the actual election. Now, we’re in territory where what really matters to some people is the response, and that’s because we are in a realm now where questioning the validity of elections is normalized by some. So setting that kind of stage that, “Oh, wow. All of the polling, the scientific polls suggest that the Republicans should win,” great stage set, great prop for then putting on a performance that says, “Oh, in that case, there must be fraud.”

Heather Cox Richardson:

Well, so we’ve gone from, “Let’s see what the people are thinking to create a good democracy,” to, “Let’s shape how the people think about the democratic system to overturn it,” and that is, I think, a really important story in what polls mean and how we should understand them. That being said, I don’t mean to be mean to pollsters here. There are very good pollsters out there. We know a lot of very good pollsters, but at the same time, it seems like a political moment right now more than a polling moment. I wonder, going forward in this election, how representative this polling has been because if we go back to the idea of your gut sense, and what you’re hearing, and all that, my take on how this election is playing out is really different than I’m reading in the mainstream media, and I’m wondering if I’m talking to the wrong guys the way they were in 1792, or if I’m actually talking to the right guys, and there’s going to be a different outcome than people expect the same way as happened in 1948.

Joanne Freeman:

Interestingly, I think where that leaves us now is as Americans and as voters, I think it does us well to not necessarily heed those polls and just all go and vote because that’s what you should be doing regardless of what the polling says. On the other hand, as people who are looking at the process, whether that’s as historians, or political scientists, or interested Americans of all kinds, it really, really needs to be said that we need to be watching the process in action of what’s being done with these polls, how they’re being used and deployed now, and what that adds up to after the election because we’re watching the evolution of a new way of engaging with our democratic electoral process as it’s being challenged, and the patterns that we see in this election will be very much worth considering when we move on to future contests.