Myers Podcast: Explaining 1948 Presidential Election Coverage

For the 133rd episode of the Journalism History podcast, Historian Cayce Myers describes the tactics used by the press in explaining its errant coverage of the 1948 presidential election, drawing parallels and distinctions between the strategies used in 1948 and 2016. 

Transcript 

Cayce Myers: In ’48 and in 2016, there was a certain degree of confidence given to polls that really did not acknowledge their own limitation.

Ken Ward: Welcome to Journalism History, a podcast that rips out the pages of your history books to re-examine the stories you though you knew and the ones you were never told.

Teri Finneman: I’m Teri Finneman, and I research media coverage of women in politics.

Nick Hirshon: And I’m Nick Hirshon, [00:00:30] and I research the history of New York Sports.

Ken Ward: And I’m Ken Ward, and I research the journalism history of the Great Plains and Rocky Mountains.

And together, we’re professional media historians guiding you through our own drafts of history. Transcripts of the show are available online at journalism-history.org/podcast. This episode is sponsored by Taylor & Francis, the publisher of our academic journal Journalism History.

We’ve had some trouble with political reporting and polling these past 10 or so years in [00:01:00] the US, and the election night of 2016 is doubtless etched in the memories of many listeners of the show. But inaccurate polling and errant election reporting is far from a new problem. In today’s show, Dr. Cayce Myers, who’s director of graduate studies and professor of public relations in Virginia Tech’s School of Communication, walks us through the media’s handling of the infamous presidential election of 1948, when incumbent Harry Truman pulled off a surprise victory over Thomas Dewey despite media and poll predictions that the [00:01:30] election would go to Dewey. The resulting photograph of Truman holding up the front page of the Chicago Daily Tribune, crowned with the incorrect headline Dewey Defeats Truman, is burned into our collective consciousness, not entirely unlike that election night in 2016.

Myers, with co-author Karen Russell, studied the response of the media to their errant reporting on 1948 as well as its response to 2016. Today, Myers explains the importance of context to an election and why [00:02:00] reporters should never take for granted that the popular narrative around an election is correct.

Cayce, welcome to the show. So tell us, what is it that’s so dangerous about past polls? We’re gonna talk a lot about that today. Why is that so dangerous for journalism, for things more broadly? What’s the big deal?

Cayce Myers: Well, I think it creates an expectations violation for people. They rely on polls, they rely on public opinion to inform them of what the future’s gonna be. [00:02:30] And when that future doesn’t turn out to be as predicted, it creates a certain uncertainty with individuals. And I think that it ultimately challenges their faith in institutions or challenges their faith in the people that have told them information and in this case with [00:02:46] pollsters in the media.

Ken Ward: Okay, sure. What, what is the big danger toward media then, right? If I’m a journalist and I use bad reporting, or excuse me, bad polling in my reporting, what, what is the sort of danger posed [00:03:00] to my organization and its relationship with an audience?

Cayce Myers: Well, I think that the relationship of credibility and the kind of the honesty of what the media are saying. Now, we’re gonna talk today about some historic examples of poll inaccuracies, notably in 1948. But even within contemporary times, when we see polling that comes out that is saying one thing and then it turns out to be something else, the media do take a hit on that because that oftentimes [00:03:30] is, is stemming from the media themselves. Media conduct polls, and in fact, some media institutions have really created an identity around polling. So I think that in this climate, particularly in the 2020s, when you see a lot of decline in faith in traditional sources of information, a decline in faith in journalism, a decline of faith in media, when you have inaccurate polls, that sometimes results in a decline, a just [00:04:00] sort of evidence of, um, evidence of dishonesty, I think, by some viewers.

And I think it partially stems from not understanding how polls are done. And part of it stems from this expectation that we have that polls are gonna be predictive of the future and we’re gonna know the truth from them.

Ken Ward: Absolutely. So, um, let me… I wanna get to… You mentioned the 1948 election. I wanna get there, too. In your research on that and other elections, [00:04:30] you’ve used this theory. It’s called image repair theory, and we probably need to understand at least a little bit about that to really understand your findings. So in a nutshell, what is image repair theory?

Cayce Myers: Well, image repair theory is a public relations communications theory that looks at how an organization or an individual can repair their credibility after a crisis. And so in a nutshell, it is how do you restore a [00:05:00] good public image after a negative event? And this has been articulated first by the research of William Benoit, but also other people have used this theory in a variety of contexts. So let’s say we have a crisis, whatever the crisis may be, you know, organizational institution does X, resulting in negative public sentiment. What can they do to bring themselves back into good graces? Now, that is very contingent upon [00:05:30] what’s going on. So is it their fault? Is it not their fault? You know, how bad is the crisis, etc.? So when we were looking at this, uh, inaccurate polling from 1948, this piece that Karen Russell and I worked on, we used that theory to explain how does the media restore itself after a crisis?

And so a lot of times what you’ll see in that theory is that people will look at like businesses. They’ll say, [00:06:00] okay, you know, like Exxon Valdez oil spill, you know, what did they do to repair their image? Or what should they have done to repair their image? But it’s rarely applied to media. In this case where you have the media as providing information that turns out to be, let’s say makes false prediction, how does the media correct itself? And then in the theory, to get into a little bit of the weeds, but not too deep-

Ken Ward: Sure.

Cayce Myers: There’s different strategies. Like denial is a strategy. Evasion of responsibility [00:06:30] is a strategy. Uh, reducing the offensiveness of the event is a strategy. Corrective action is a strategy. Mortification, which is like apology, is a strategy. And all those strategies depend upon the circumstances that are surrounding the event itself.

Ken Ward: Well, so let’s talk about one election specifically. We’ve mentioned 1948. Great example and one that you’re familiar with from your research. So tell us, what did polling look like in the lead-up to that election? What was the state of political polling [00:07:00] pre- and during 1948?

Cayce Myers: Well, prior to 1948, going all the way back into the 1800s, 1840s, there were, there was an understanding of public opinion as a concept. I think that’s important for people to realize that, you know, prior to the 19th century, you know, the idea of having public opinion and surveying public opinion was not like it is today. I mean, it, that was [00:07:30] not a concept that people thought about necessarily. I mean, it was a different time. I mean, literacy was different. Media were different. The way that you could access individuals were different. But with the Industrial Revolution and the urbanization of Western Europe and America, you have an increased, I guess what you would call, like civic engagement and awareness. You have a higher degree of access to media. And so there is such a thing as public [00:08:00] opinion. So in the 1800s, people were aware that that existed and was an important component of life.

Now, the study of it started off really more qualitatively. And so we had a couple of people that were writing about public opinion in the 1890s, Gabriel Tarde and Gustave Le Bon, who were these, you know, French social theorists, who were looking at… They published a couple of books that were looking [00:08:30] at kinda how crowd behavior works and how individuals form opinions and things like that. But by the 1940s, you have a more robust polling in place, a more robust understanding of polls, an understanding of how to access individuals and get their public sentiment. Now, fast forward to 2023, you know, we’re in a much different place than they were then. Polling is such a [00:09:00] part of the conversation today, and access to polls are so prolific. So you go on something like realclearpolitics.com, the public has the ability to look at polls and aggregates of polls and, and really get into the data. That didn’t exist in the 40s.

So it’s comparatively today a much more crude, uh, much more crude process, but one that had accuracy and one that people did have confidence in. I will say just particularly [00:09:30] about 1948… And I think this is important to note… there were certain underlying assumptions about presidential politics, and one was that people more or less made their decision about who to vote for at the time of a party convention. And that was a theory, or a, a kind of a law, not a formalized theory, called Farley’s Law that was created by a guy named James Farley, who was FDR’s campaign manager in 1932 and ’36. And he believed [00:10:00] that, if you have a convention candidate and they are nominated, the people know whether or not they’re gonna vote for who they’re gonna vote for then. And so the whole presidential election process is kind of like theater. It’s not really gonna sway anybody one way or another, and everybody’s gonna kinda vote how they were always gonna vote. And you may have some people that are swayed one way or another, but it’s a small, small, small, small minority of people.

 And so that, [00:10:30] that concept, I would say was baked into some of the decision-making that the candidates made in the 1940s, particularly Dewey, and it was part of kind the understanding of what polls meant and, and how persuasive a campaign could be to change attitude. So-

Ken Ward: So you… Oh, I’m sorry. But you mentioned-

Cayce Myers: But that, that’s kind of in a nutshell where they were in 1948.

Ken Ward: Yeah. No, that’s extremely helpful. Understanding the way that they were headed into that election [00:11:00] and that, that post-nomination period. So tell us a little bit about that election, ’cause I don’t want to take for granted that everybody knows who these characters are. Who are the main players in this election? And what did the coverage in 1948 look like in the lead-up to Election Day?

Cayce Myers: So Harry Truman was the incumbent president, who had assumed the presidency from Franklin Roosevelt. And he was running for re-election. Well, he was running for election. It was an unprecedented sort of election. If you look at election trends, [00:11:30] one party dominance in the White House for 12 years is highly unusual in the United States. And here we were going for – So from 1932 forward until 1948 had been a Democratic president. So he was… in that sense, it was kind of the historical narrative sense, what he was doing was counter, uh, him winning was sort of counter to what American elections had borne out previously. [00:12:00] He was suffering from some political defeats. In 1946, he was dealing with a Republican congress that had basically brought back into the national political scene more conservative Republican views, particularly around trade. They had had a very kind of an acrimonious political issue with the Taft-Hartley Act, [00:12:30] in which President Truman ended up kinda losing on that issue.

There was a sense that there was a Republican surge going through the country. This was post-World War II, post-FDR, and, and kind of the sense that there may be a time for change. In addition, internally within the Democratic Party, Truman was facing challenges from the right and from the left. [00:13:00] Uh, from the right, you had these states’ rights Democrats from the South who were segregationist Democrats challenging the president on civil rights issues and the role that civil rights played in the election. And that ended up creating a third-party candidacy for the states’ right Democratic Party, who ended up nominating Strom Thurmond as their, as their [00:13:30] nominee. And then you had left of – sort of left-wing part of the Democratic Party with Henry Wallace, who was challenging kind of Truman on some doctrinal liberal stances and, and kind of challenging him as, as a more of a moderate.

So any time you look at an election and you have internal party problems, that’s a sign of a weak incumbent. So Dewey on the Republican side was a person [00:14:00] who was a governor of New York, was very popular, had been elected by a large majority of the population in New York. And he was viewed more of a centrist. He had some challenges in 1948 in the Republican Party from more conservative forces, particularly Senator Taft from Ohio, but Dewey was somewhat of a known entity. He had made a [00:14:30] name for himself in the 1930s as a prosecutor. He was someone who was from a large state and who had won large in New York as a governor. And then he was running with a running mate from California, Earl Warren, who had won his governorship really through running kind of what would be considered almost like a centrist nonpartisan type campaign. So he had a wide range of appeal.

And so the thought [00:15:00] was, is that Truman’s problems were so bad that Dewey didn’t really need to run so aggressively against Truman. He just needed to run and that Truman would obviously be in a weakened condition and that Dewey could kind of run this above-board kind of campaign, kind of high-minded campaign, not really discussing Truman very much, not really discussing the Democrats very much, but talking about unity and, and [00:15:30] kinda some of these bigger picture issues and, and win handily. And that was reinforced by polls, and that was reinforced by the media.

Ken Ward: Interesting. And that… You… It does sound at least a little familiar, right?

Cayce Myers: It is very, very similar to 2016 in the sense of insurgent versus… Not to the compare the policies, but to compare the situation and, and kind of a person in a weakened condition, a lot of problems, [00:16:00] versus a very strong candidate and then kind of communicating and the communication style being different based on that perception of strength.

Ken Ward: And then somehow we still wind up with Truman holding up Dewey defeats Truman. Like we still have this image baked in our minds. The election didn’t go as-

Cayce Myers: It did not.

Ken Ward: Planned, right? Or at least as predicted. So why and how did the news media respond? Was it, was it about the polling? Or was it about something else?

Cayce Myers: I think that [00:16:30] there was in both the polling and the coverage. And, and you have to go back… And, and I know that’s why I mentioned that Farley’s Law… That was something that, to understand a historical era, you have to understand the mindset of that era. And so the mindset of politics in 1948 was that you had a, you know, almost, you know, since 1932 total [00:17:00] Democratic Party domination of the federal government. And so there is a certain fatigue with that. There is a 1946 kind of Republican insurgency in the midterms. So you look at that, and you say, well, they’re on the rise. Democrats are on the decline. And we have two strong candidates on the Republican ticket who are from large states who are kind of part of this, you know, kind of [00:17:30] new reality post-World War II. And then you have a Democratic candidate who has a third party running against him. And then… And I should just sort of mention this… the states’ rights Democrats, or the Dixiecrats, as they’re later called, their whole strategy was not to win the election. Their strategy was to prevent Truman from winning the election and assert their, you know, significance within the Democratic party. So the idea was to be a spoiler, not to be [00:18:00] a candidate that’s gonna actually win.

And so… And, and they did not end up accomplishing that, but they did come… had things kinda shaken out a little bit differently in some states, they very well could’ve done that. And so that was a strategy that they had. So, you know, if you’re looking at it from a perspective of that time, not knowing the results, as we do today, but at that time, it would seem that there’s a net advantage for Republicans. And in the polls, what you were seeing was a lot of people that were saying [00:18:30] they were gonna vote Republican, and you had a lot of people that were saying that they were undecided. And so in that mix of the polls at that time, it lent to this perception that how it’s not “Is Dewey gonna win?” but how big is he gonna win? And there was a tightening of the polls right before the election.

But again, as that famous headline from Chicago Daily Tribune that you mentioned that says “Dewey [00:19:00] Defeats Truman,” the media was prepared for a Dewey victory. They had already kind of made that decision to run some content that related to that the following day. So they were as shocked as anybody else by the outcome.

Ken Ward: And so what do they do then? Thinking about image repair theory or, or mustering other evidence, like what, what does the media do to explain its shortcomings in covering this election, or at least in getting the outcome incorrect?

Cayce Myers: I think [00:19:30] this is another thing that kinda highlights the differences of today versus the 1940s, and that is the media at that time were not so embroiled in a polarized political conversation. And so you had outlets that had identities, or political identities, but it wasn’t kind of, it wasn’t like what it is today. And so there was a certain degree of ownership of, of getting it wrong, but [00:20:00] in a humorous sort of way. And I don’t think that, uh, you know, just comparatively to today’s politics, that would not have been the natural – that was not the result in 2016.

Ken Ward: (Laughs)

Cayce Myers: Like a humorous, well, you know, they stomped us. Uh, can you believe it? You know, and then just sort of a good-natured kind of ownership of it. And that really was what it was in the 1940s. There was some blaming that went around to pollsters, you know, what did they do to get it wrong? [00:20:30] And you know, some of it had to do with who did they survey? And of course, that’s kind of the challenge now, like who did you survey? Truman was very popular with people in rural areas. And so one has to wonder if that population was not surveyed as well as it should’ve been. The other thing is that when they went back and did the kind of the post-mortem of the election, they found that, unlike what the Farley Law prediction was, is that people made up their mind in the last few days [00:21:00] of the election, that they were truly undecided and then broke toward Truman in a very significant way. And so the idea that, or the response to that was, you know, the polls got it wrong. Uh, there was some reasons for that, but it really was that it was a certain amount of unpredictability to it in that it was late decisions that were being made that broke toward the incumbent.

Ken Ward: [00:21:30] So how effective were responses in putting aside the potential problems caused by getting it wrong? Right? If that sort of relationship with the reader, in this case, the reader is what’s so important, were these responses effective at, in maintaining that audience and maintaining and protecting that relationship that had been built?

Cayce Myers: Well, I don’t think… So this is another thing kinda contemporary versus then. The audience didn’t go away. You know, I think maybe the big, [00:22:00] the big takeaway that may have changed from 1948 was the way that politicians used polls and the way that campaigns were run. You know, as I mentioned earlier, those polls really affected the way that candidates themselves managed their own election. So you know, Truman barnstormed in 1948 in this extremely aggressive campaign where he was out and about with people, particularly in rural areas and really [00:22:30] challenging the Republicans and really hammering them on a lot of issues and calling them out. And, and Dewey wasn’t doing that at all. Dewey was kind of running this kind of above-the-ground kind of campaign where he talked about very broad aspirational politics. And I think that was the result of those polls. And I think that that had to have influenced them, like why… don’t tread on your own success by kinda getting ensnared in sort of, uh, fight. [00:23:00] And, and ultimately what happened is that Truman’s strategy turned out to be better, and that’s probably the reason he won.

Uh, in terms of confidence in the media, you know, media confidence and readership, there’s nothing to indicate from our study that, you know, media confidence was eroded. There may have been some erosion in confidence in the pollsters, but, but not in the media themselves. And it’s important that there was a distinction in the 1940s between the press and the pollsters themselves. You know, [00:23:30] the pollsters are a source. They are not synonymous with the press. That may be a little different today in that we do have press outlets that perform polling, and so they are synonymous with each other.

Ken Ward: So I don’t wanna get caught up in 2016, but I… I don’t think it would be right to completely ignore the connection between these two. And you’ve noticed… We’ve noted that some parts of the story at least overlap, right? So you’ve [00:24:00] pointed out some of the ways 2016 differed in terms of media response. Can you refine those for us here? What was different in the way the media responded to their poor coverage in 1948 or, you know, the outcome coming out differently in 1948 than they had predicted, versus that again happening in 2016? How did they respond differently-

Cayce Myers: So let me just, let me just give a little anecdote about ’48. And this, just imagine that one of the things they wanted to do in [00:24:30] 1948 was to throw a party and called the Crow Party where they were going to eat crow over being wrong. And they were going-

Ken Ward: (Laughs)

Cayce Myers: To invite Truman to it. And the Chicago Daily Tribune, right before Truman died… And, and they weren’t able to do it because he died before they did it… they were going to give him a bronze plaque supposedly of that infamous wrong headline.

Ken Ward: (Laughs)

Cayce Myers: As, as a sort of a memento of that occasion. So I mean, [00:25:00] that tells you… That – Yeah, you know, thinking of 2016 or 2020 or 2023, I mean, you can’t imag- I can’t personally imagine that being a result. So in 2016… And, and I should just say for the listeners… that was the other part of our study, like you know, ’48 was, was this bad poll, and then 2016 were kinda these bad polls. And then what’s the comparison? The media in 2016, one of the things they did is they took it very personally, [00:25:30] I think, that they had gotten it wrong. Uh, some of the blame went to the pollsters, and then some of the blame then went to those who were polled. And there was a question about whether people were truthful in how they answered polls and were you getting these false narratives created by untruthful answers or by bad sampling? And you know, and can we really trust polls? And I, and I say, [00:26:00] I should say this, too, in 2020, in that presidential campaign, there was also a lot of inaccurate polling. And there was a lot of discussion around that. Even though the end result prediction was correct, there was… It… The margins were not reflective of what the polling was gonna show.

They did not predict it be as close as it was. So you know, this is a constant narrative within American politics, is like, you know, as we get into polling, as the media relies on polling and, and what happens, [00:26:30] um, you know, how do they respond to something that’s inaccurate? I’ll tell you, just sort of anecdotally looking at it, you see today a real reluctance on the part of media outlets to call elections very quickly. And I think it’s because of this kind of… It’s rooted in some of these inaccuracies with polls where there’s been bad predictions made and then famously calls that have had to be retracted and, and things like that. So I think there’s a certain degree of hesitancy now amongst the [00:27:00] media because of this kind of thing, not just 2016, but just at large.

Ken Ward: So your case here I think presents an interesting opportunity. I’m always reluctant when I ask guests of the show, you know, based on your research into this thing that happened in 1880 or whatever, you know, what, what lessons should we learn about the present? Right? Like that’s always a little fraught with, with, uh, older examples, but you’ve got a historical example. It’s not … We’re getting close to 100 years here. We’re working our way toward it, but 1948, that’s [00:27:30] not so long ago. And then you’ve got this next touchpoint in 2016 that you’ve also already done this comparative analysis. So you’re kind of in a unique position among many of our guests. You can actually, I think, with some authority say like what are some lessons that we should be at least mindful of right now? Now, those could go in the direction of, you know, those of us who consume news media, those who are making it, news organizations responding when they get things wrong. I wanna leave it entirely open to you. What, for anybody, [00:28:00] are lessons that we could pull out of all of this to change the way we act today?

Cayce Myers: Well, I think that the two case studies are similar, although, you know, the parties involved are radically different, the individuals involved are certainly radically different than they are, you know, compared to 1948 to 2016 or 2023. The politics are different. The media is different. It’s just a different dynamic. And so I think you’re absolutely correct. You know, you [00:28:30] don’t wanna superimpose something from the past on the present, but what I think is similar was in ’48 and in 2016, there was a certain degree of confidence given to polls and a presentation of polls that really did not acknowledge their own limitation. So margin of error is something that we hear about in the media, you know, plus or minus 3%, or whatever it may be, but [00:29:00] I think people don’t really process that necessarily. Then they say, oh, well, this says that he’s gonna win or that she’s gonna win, and that’s just the way that it is. And, and, you know, it is a science and an art, and it is something that is not 100% predictive.

And so I think that’s where kind of the – We started off the top of the show with, with, you know, kind of an expectations violation, like I expect it to be this, and then they’ve told me that it’s gonna be this, and then it isn’t this. I think that’s rooted in really understanding [00:29:30] what the source of polls are, or understanding the parameters of polls. And I think that was the challenge in ’48 and in the challenge of 2016. I think there’s also a lesson learned for journalism and journalism students who may be listening to the podcast and, and those in the media in that when you’re reporting, I think there’s oftentimes a tendency if you have certain facts to say, okay, these are the facts, and this is the story, but I think that this inaccurate polling [00:30:00] scenarios, or these case studies, show that maybe you have to question what the reality is, but… And, and the more that a particular viewpoint just seems entrenched, like this is the way that it’s gonna be, maybe you should really question that and say what are the underlying assumptions of that?

You know, the underlying assumption is that, you know, in 1948, was that people made their minds up really early about elections, and so, you know, you don’t really have to worry about what’s going on in the campaign ’cause they’ve already decided. [00:30:30] Well, baked into that is that I don’t need like current polls necessarily of the day before the election because the election was decided back in the convention era, or, or whenever that was, months prior. In 2016, you know, maybe there’s some, some polling inaccuracies about, you know, what are states that are not in play, and, and then maybe there’s an under sampling there or there’s not attention paid to that. So I think as a journalist, you know, journalists are good questioners. And when you’re doing journalism and [00:31:00] you see kind of these examples of where it gets wrong, a lot of times I think it’s because of not asking the right questions or challenging kind of the status quo understanding of an issue. And I think that’s what those two examples, from ’48 and 2016, are really insightful for contemporary people who are either working in journalism and consuming news.

Ken Ward: Well, Cayce, we’ve only got time for one last question. I absolutely wanna [00:31:30] hear your answer to it. It’s the one we ask all our guests. That question is why does journalism history matter?

Cayce Myers: Oh, it… So we, we need a whole ‘nother podcast, maybe two or three.

Ken Ward: (Laughs)

Cayce Myers: You know. So journalism history is unique in that it looks at the media to tell the story of the times. And that is, you know… And a lot of times when we look at history, we look [00:32:00] at an individual or we look at one particular source, like a diary or something like that, and we’re looking at it through the perspective of an individual. And that’s important, too, because that, that really provides a lot of the nuanced insights. But journalism history at large, looking at what the media says, it is so insightful for people to understand what we call zeitgeist of a time. You know, how did people think? [00:32:30] I’m a firm believer that perception is sort of the reality and that people’s perception of what’s going on, which is frequently informed by the media, is a reflection of what, you know, contemporary society is. What do they think? What’s important? What’s not important? What’s going on in terms of the larger social conversation? How do people see themselves situated in the larger society and country [00:33:00] and in the world? And journalism history does that. And I think that’s the true power of journalism history. To understand media history and journalism history is to really to understand ourselves as a society and as a community.

Ken Ward: Excellent. Well, Cayce, we’re out of time, but I just wanna thank you one more time for being on the show. I enjoyed this conversation. I think we learned a lot about two elections here. So thanks again.

Cayce Myers: Thank you. Appreciate it.

Ken Ward: Well, that’s it for this episode. Thanks for tuning in, and be sure to subscribe to our podcast. [00:33:30] You can also follow us on Twitter @JHistoryJournal. That’s all one word. Until next time, I’m your host, Ken Ward, signing off with the words of Edward R. Murrow, “Good night and good luck.”

Featured image: This newspaper is on display at the Harry S. Truman Presidential library in Missouri. Photo by Teri Finneman.