Why Truman’s 1948 upset is no template for the 2024 U.S. presidential election, according to the expert who wrote the book on polling failure
Harry S. Truman’s stunning, come-from-behind victory in the 1948 presidential election has encouraged frissons of optimism for long-shot candidates ever since. Walter Mondale invoked the 1948 election in his race against President Ronald Reagan in 1984, insisting the polls signaling his overwhelming defeat weren’t picking up the popular support flowing to his candidacy. Mondale lost in a 49-state landslide. In 1996, Bob Dole channeled Truman’s pugnacity of 1948, declaring, “I’m going to win whether you like it or not.” The closing hours of Dole’s race against President Bill Clinton included a campaign stop at Truman’s hometown in Independence, Missouri. Dole lost by 8.5 percentage points.
President Joe Biden’s uncertain campaign for reelection this year has invited comparisons to Truman’s come-from-behind victory in 1948. A recent assessment in Politico, for example, said the blueprint Truman followed in his aggressive and strenuous campaign in 1948 offers Biden a playbook for victory. Even the leading newspaper in France, Le Monde, has invoked the 1948 template as a way of understanding this year’s U.S. presidential election.
While broad and superficial similarities may be detected between 2024 and 1948, the two cases are in fact quite dissimilar, especially in the conventional wisdom that pre-election polls generate and reinforce.
Polling-induced cockiness
As I note in Lost in a Gallup, my book about polling failure in presidential elections, pre-election polls are central to how journalists, and Americans at large, understand the dynamics of presidential campaigns.
For many months, opinion polls have signaled a tight race between Biden and former President Donald Trump. That wasn’t the case at all in 1948 when Republican Thomas E. Dewey maintained a clear polling lead throughout the campaign. The poll-driven narrative of 1948 anointed Dewey as the prohibitive favorite.
There were fewer national pollsters in 1948, but their track record in the previous three presidential campaigns (all won by Franklin D. Roosevelt) encouraged confidence that the results they reported were highly reliable.
Among the early pollsters was Elmo Roper, whose survey reports appeared in Fortune for many years beginning in 1935. In September 1948, Roper confidently declared on his CBS radio program that the “science” of opinion polling had “come of age” and had “proved its accuracy and worth.” He announced at that time he would release no further poll results, so certain he was that political campaigns made little difference and that Dewey would win the election by “a heavy margin.”
Other leading pollsters of the day, George Gallup and Archibald Crossley, likewise predicted Dewey’s victory. Both of them ended their polling in October, well before the 1948 election. Gallup was unequivocal in reporting his final pre-election poll that year, writing: “Dewey will win the Presidential election with a substantial majority of electoral votes.”
Gallup also said that “the whole world will be able to see down to the last percentage point how good we are” on Election Day 1948.
In their respective polls, Gallup and Crossley projected Dewey’s victory by 5 percentage points. Truman carried the popular vote by 4.5 points and won 303 electoral votes to Dewey’s 189.
Roper afterward conceded that pollsters “had gotten pretty smug, and I was one of the smuggest of the lot.”
Few similarities between the 2024 and 1948 campaigns
Pollster cockiness hardly has been a feature of the 2024 race–not after high-profile surprises in the most recent presidential elections. In an outcome that rivaled 1948 for shock value, Trump defeated Hillary Clinton in 2016, due in part to late-campaign preference shifts that polls mostly failed to detect in such key states as Michigan, Pennsylvania, and Wisconsin.
In 2020, several prominent national polls, including those conducted for CNN and jointly for NBC News and the Wall Street Journal, anticipated Biden’s winning by 10 points or more. Biden carried the popular vote by 4.5 points in what, overall, was the poorest polling performance in a presidential election since 1980.
As of Apr. 4, the RealClearPolitics polling average showed Trump with less than a percentage point lead in a two-way race against Biden–scarcely a margin to inspire 1948-style overconfidence.
Another notable difference lies in the sense of vigor projected by the Democratic candidates. Truman was 64 years old in 1948 and pursued a grueling campaign of a kind that is difficult to imagine the 81-year-old Biden even remotely contemplating. Truman logged thousands of miles by train in an arduous, cross-country whistlestop campaign, or what biographer David McCullough called “a fast-rolling political roadshow.”
“I know I can take it,” Truman said of the rigors of the campaign, adding in jest: “I’m only afraid I’ll kill some of my staff.”
Meanwhile, Dewey ran an above-the-fray campaign, a model Trump would never embrace. Dewey, unlike Trump, sought to minimize controversy and avoided specific policy pronouncements, once telling an aide “When you’re leading, don’t talk.”
Other notable differences between 1948 and 2024 are not difficult to identify.
The Democrats fractured three ways in 1948. The segregationist Dixiecrats nominated J. Strom Thurmond of South Carolina as its presidential candidate; he carried four Deep South states. The far-left Progressives selected former Vice President Henry Wallace of Iowa, who won not a single state. Truman led the mainstream Democrats and carried 28 states. (Dewey won 16 states.)
This year, by contrast, Democrats are largely, if not ardently, unified around Biden, although the independent candidacy of Robert F. Kennedy Jr. may attract votes otherwise destined for the president. As of early April, however, Kennedy’s campaign has qualified for the November ballot in just one state.
In U.S. presidential elections, past is seldom prologue. A polling-derived surprise akin to that of 1948 is highly unlikely this year.
W. Joseph Campbell, Ph.D., is a professor emeritus of communication at American University in Washington, D.C. He has written seven solo-authored books including, most recently, Lost in a Gallup: Polling Failure in U.S. Presidential Elections. You can follow him on X @wjosephcampbell.
More must-read commentary published by Fortune:
Glassdoor CEO: ‘Anonymous posts will always stay anonymous’
We analyzed 46 years of consumer sentiment data–and found that today’s ‘vibecession’ is just men starting to feel as bad about the economy as women historically have
90% of homebuyers have historically opted to work with a real estate agent or broker. Here’s why that’s unlikely to change, according to the National Association of Realtors
Intel CEO: ‘Our goal is to have at least 50% of the world’s advanced semiconductors produced in the U.S. and Europe by the end of the decade’
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.
This story was originally featured on Fortune.com
