On the eve of the 2024 elections, a shock poll from Republican-red Iowa set the political world on fire.
The so-called “gold standard” Des Moines Register survey, conducted by veteran pollster Ann Selzer, found Vice President Kamala Harris leading Donald Trump 47 percent to 44 percent.
Trump carried the state by nearly 10 percentage points in 2020 and 2016.
Harris’s sudden rise in the state was likely fueled by support from women and especially female voters over the age of 65, who backed the Democrat by as much as 35 points.
The survey was seized upon by Harris surrogates and a left-wing media who suggested it was evidence of a “hidden Harris” voter that all other national and swing state polls had somehow missed.
Ultimately, of course, this “hidden Harris” voter never materialized – in Iowa or elsewhere.
Trump won the state by 13 percentage points, the largest margin ever since 1972, and a 16-point difference from Selzer’s prediction.
Days after the election, the Des Moines Register announced that they had launched a review of Selzer’s voting methods to determine what had gone so dramatically wrong.
Not only were the polls extremely bad, but Selzer’s erroneous numbers tarnished the reputation of the Register and the polling companies as a whole and may even have affected the outcome of the election.
The so-called “gold standard” Des Moines Register survey, conducted by veteran pollster Ann Selzer, found Vice President Kamala Harris leading Donald Trump 47 percent to 44 percent.
Trump won the state by 13 percentage points, the largest margin ever since 1972, and a 16-point difference from Selzer’s prediction.
Could her results, which were broadcast across the country in the final days of the campaign, have driven more Harris voters to the polls? Could they have had the opposite effect and motivated Trump’s supporters to turn out?
Both scenarios are conceivable. We will never know the full impact.
On Sunday, Selzer announced her pre-planned retirement from the Register to pursue “other ventures and opportunities.”
Hours later, President-elect Trump made his own statement.
‘A completely fake poll that caused great distrust and uncertainty at a very critical moment. (Selzer) knew exactly what she was doing,” Trump claimed on Truth Social, adding, “An investigation is absolutely necessary!”
Well, I did my own research on Selzer’s polling. And my conclusion is that this investigation may be the sloppiest work I have ever seen in my five decades in the polling industry.
First off, I will say that I don’t know Ann Selzer.
Our paths have never crossed. I only know her reputation as a trusted professional, dubbed the “Queen of Polls,” after she correctly predicted that a virtually unknown Senator would defeat Hillary Clinton in the 2008 Iowa Caucuses.
But after examining the methodology of her November survey, I am baffled. The glaring problem is not difficult to spot.
In the survey, Selzer asked 808 likely voters in Iowa, “Who did you vote for in 2020, Biden or Trump?”
Selzer must have expected to get results that resembled the actual vote in the 2020 general election — Trump won 53.1 percent to Biden’s 44.9 percent, an eight-point difference.
Those were indeed the results she achieved in the past.
In a February 2024 Register poll in Iowa, respondents said they voted for Trump over Biden in 2020, 45 percent to 38 percent — a seven-point margin.
In June 2024, the sample for Trump went from 49 percent to Biden’s 35 percent – a margin of 14 points.
But in September that gap narrowed to four points, and the numbers reversed in the November poll.
Then respondents said they voted for Biden over Trump, 41 percent to 40 percent. That’s a nine-point deviation from the actual 2020 election results in favor of Democrats.
Now, in some cases, poll responders have been known to lie about their past voting record and claim they supported the eventual winner. But the magnitude of the variance here is too large to ignore.
Selzer acknowledged this discrepancy last weekend when she published her analysis of the election blunder. Although she seemed to dismiss the finding, writing, “Maybe there is some merit in this,” she ultimately concluded, “I have found nothing to clarify (November’s) miss.”
‘A completely fake poll that caused great distrust and uncertainty at a very critical moment. (Selzer) knew exactly what she was doing,” Trump claimed on Truth Social, adding, “An investigation is absolutely necessary!”
As a polling analyst and former project director of President Ronald Reagan’s White House public opinion team, I know that a discrepancy of that magnitude should have sent a shiver down the proverbial spine of the Register team.
Because it indicates that Selzer may have surveyed a group of Iowans who do not accurately reflect the sentiment of the state’s overall electorate.
At this point, in my opinion, Selzer would have had two options: she could have sent her research team back into the field to collect a more representative sample, or she could not have released the poll.
This is the industry standard when data doesn’t smell nice. Selzer, however, did neither.
Instead, she “weighted” or artificially increased the influence of Trump 2020 voters in her poll.
“Weighting” is an accepted practice that allows a pollster to check for under- or over-representation by increasing or decreasing the relative percentage of a particular group of voters in a sample.
But applying “weighting” to a sample that is fundamentally flawed can only make the results worse.
In other words, it’s like putting lipstick on a pig. It will dress the animal up a bit, but it’s still livestock.
I can’t assume I know exactly what went wrong with Selzer’s poll, but the overrepresentation of Biden 2020 voters raises the possibility that she oversampled predominantly liberal regions of the state (such as the areas around the liberal Grinnell and Coe colleges).
If Selzer were to survey a disproportionate number of Trump 2020 voters who lived in Democratic enclaves, then it is reasonable to assume that these Iowans may not have been representative of Trump 2020 voters statewide.
This type of sample can never be ‘corrected’ by weighing.
Setting aside the curious potential internal features of Selzer’s polling methodology, the final results themselves should have been the biggest red-flashing warning sign.
Craig Keshishian was project director for President Reagan’s election team and later served in Reagan’s Presidential Speechwriting and Research Office
Iowa, aside from parts of Des Moines and the small college towns, is about as red as Judy Garland’s slippers in The Wizard of Oz. In fact, Iowa is so reliably red that most pollsters now don’t bother polling the state during a general election.
In addition, Selzer has had polling places in the state for 30 years. She should know the voters like the back of her mind, and honestly, Iowa isn’t the hardest place to conduct a poll. It is largely rural with only three major urban centers.
If there had been a huge wave of voters toward Harris, a native daughter like Selzer would surely have also recognized supporting anecdotal evidence, such as “Harris 2024” signs in front yards in Republican areas, or major GOP influencers throwing their support behind Trump pulled.
But that evidence was not present.
Selzer’s poll could have come as a shock to America. But it should not It was a shock to Selzer that her latest research was a failure.