WEDNESDAY, NOVEMBER 6, 2024
That is, with the Iowa Poll: In the past, Ann Selzer's Iowa Poll had frequently landed pretty much right on the money.
Every four years, results of the final survey would be released on the Sunday before the election. This year, that final poll showed Candidate Harris ahead of Candidate Trump by three points statewide, 47-44 percent.
We'll say this for the Iowa Poll—when it finally missed the mark, it held little back. Here's the current report on the topic in the Des Moines Register:
Pollster J. Ann Selzer: 'I’ll be reviewing data' after Iowa Poll misses big Trump win
Renowned Pollster J. Ann Selzer said Tuesday she would be reviewing her data to determine why a Des Moines Register/Mediacom Iowa Poll released just days before the election produced results so far out of line with former President Donald Trump's resounding victory.
Trump handily won Iowa for a third time, defeating Vice President Kamala Harris by 14 percentage points with more than 90% of the vote counted―a sharp contrast to Saturday's Iowa Poll that had Harris leading by 3 points.
"Tonight, I’m of course thinking about how we got where we are," Selzer, president of Selzer & Co., which conducts the Iowa Poll, said in a statement.
"The poll findings we produced for The Des Moines Register and Mediacom did not match what the Iowa electorate ultimately decided in the voting booth today. I’ll be reviewing data from multiple sources with hopes of learning why that happened. And, I welcome what that process might teach me."
As Kevin Drum noted last week, there are many ways a survey of this type can go wrong. So-called "margin of error" is only the start of the possibilities. And by the way, just a guess:
No journalist could possibly hope to explain the way that statistical artefact works. We couldn't exactly do so either, but at least we'd know not to pretend.
Simply put, everyone talks about margin of error, but no one does anything about it! In fact, basic "sampling error" comes into play if you're simply pulling red and blue ping-pong balls out of a big giant drum.
Sometimes, the sample you pull out of the drum will match the proportion of red and blue balls found inside the drum—but a fair amount of the time, the sample you pull out of the drum won't be a perfect match.
That's how it works with ping pong balls in a big giant drum. If you're polling a presidential campaign, other factors come into play, potentially messing things up.
Many people won't answer their phone when you try to reach them. Some people will answer their phone, but they won't answer your question.
Some people won't tell you the truth if they decide to answer your question. Some people may have changed their minds by the time they cast their votes.
The possible ways a poll can go wrong continue on from there.
On 24-hour "cable news," the pundits spend a lot of time, before an election, puzzling over the polls. At some point, this becomes an excellent way to kill giant amounts of time—a way to pretend you're presenting "news" as part of a process called "journalism."
This past Sunday, the Iowa Poll had Candidate Harris up by three points, with nine percent still floating around in the ether. (Three percent had said that they'd be voting for Candidate Kennedy Jr.)
For one brief shining moment, that's where matters allegedly stood. According to this AP post, here's where the statewide vote in Iowa stands with 98 percent reporting:
Statewide presidential vote, Iowa 2024
Trump: 55.9%
Harris: 42.7%
Kennedy Jr.: 0.8%
Three other hopefuls got handfuls of votes. Eventually, though, the day had to come:
Trump won by more than thirteen points! The Iowa Poll got it wrong.
This election may turn out to have had nothing to do with polling. Like the 2016 election, which included a great deal of outside help from Comey and Putin, there may be explanations related to meddling in other aspects of the campaign. I will wait to see whether that is true before believing that excellent pollsters somehow got it all wrong.
ReplyDeleteIt may boil down to misogyny and racism. It may be related to people's fears of vigilante violence from MAGAs. It may be that no one assumed that Harris could win, so they stayed home in droves. None of these explanations would show up in polling because those questions weren't asked of anyone. It is going to take a long time to discover what happened in this election.
Meanwhile, Trump is a disaster and I don't hear many people talking about what we do now. How do we mitigate the damage that Trump and his allies will surely do to all of us?
Where do women stand? Does it mean anything that Somerby has singled out this one female pollster when so many male pollsters got it just as wrong? I think it is typical of what Somerby does, and another reason why some voters couldn't imagine Harris as a viable presidential choice.
Note that Somerby is not telling us who he voted for. I'll bet, in the secrecy of the voting booth, Somerby voted for Trump, because in his heart Somerby is a bro and bitches be lyin', and Trump was unfairly convicted because Stormy is a grifter (according to Somerby). And if you cannot trust Somerby to tell the truth about his vote, you cannot trust anything else he says here, even about pollsters and Trump's margin of victory in a very strange election, manipulating partly by the NY Times and billionaire donors (on both sides), not to mention Russia. But Somerby wishes to pretend the polls were problematic and not the election they were measuring. And the con continues.
ReplyDelete"What was the matter with Iowa?"
Democrat pollsters are liars.
This has been another installment of simple answers to simple questions.
Somerby says that to be a liar, one has to know the truth and deliberately say something else. Did Selzer know Harris wasn't going to win Iowa? I doubt it. But isn't it a lot more fun to call people names?
DeleteWhy would someone tell a pollster they were going to vote for Harris, and then vote for Trump instead? Perhaps those last-minute ads by Harris reminded women that their husbands would beat the shit out of them if they dared to vote for Harris? Or maybe their husbands just grabbed the mail in ballots and filled in both of them themselves? Or maybe they grabbed their wife's ballot and tore it up, resulting in unexpectedly decreased turnout.
Sometimes there are simple answers to simple questions.
As Somerby says, anything is possible. That's why we need investigation to know what happened. The purpose of research is to eliminate alternative explanations for what happened. Somerby thinks the polling was flawed. Is that likely? I think there are other explanations that need to be explored before concluding that experts don't know how to do their jobs.
DeleteSomerby has no faith in experts, so he leaps to his favorite, preferred conclusion, that pollsters don't know how to poll. Time will tell us what happened, but it won't be as easy as calling the pollsters liars when they don't predict an election result (especially when their job is to measure attitudes, not to predict).
https://www.forbes.com/sites/stephenpastis/2024/11/04/presidential-polls-might-be-saving-face-in-final-election-sprint-heres-why-they-might-be-wrong/
Delete"Somerby thinks the polling was flawed. Is that likely? I think there are other explanations that need to be explored before concluding that experts don't know how to do their jobs."
DeleteHold on there, expert-lover.
For starters, this is the third presidential election in a row where the polls have under-predicted Trump's performance.
But as a larger, logical matter: if a poll predicts result A and the actual result turns out to be far away from result A, what other explanation is there other than the poll was wrong?
Is it reality's job to conform to the experts' expert expectations?
"What was the matter with Iowa?"
ReplyDeleteSomerby is riffing on a book title by Thomas Frank, "What's the Matter with Kansas?". That book asked why so many voters in Kansas voted against their objective self-interest. That would be a very apt question for Somerby to have focused on this time, except he is more interested in chiding Selzer (because she suggested Harris might win).
As usual, Somerby borrows the phrase without mentioning where it came from, as if he thought it up himself. It was original when Frank said it, but is derivative when Somerby does, especially without relating anything in his essay back to the focus of the original book, an important examination of populism.
"Somerby borrows the phrase without mentioning where it came from"
DeleteHe might assume his audience is bright enough to know where he got the phrase and if not, knows it isn't such a big deal in any case.
But I guess with you he's asking more from the reader than you're able to give.
Here is Steve M at No More Mister Nice Blog, discussing the ultimate problem with the Harris campaign:
ReplyDelete"
Sixty countries have had female heads of state, but it's starting to seem as if America never will. It's not just the two Democratic women who have lost -- Republicans consistently reject female candidates in presidential primaries. Many of us think Sarah Palin is no longer taken seriously as a Republican leader because she's ignorant, because she talks nonsense, and because she's a tabloid-friendly drama addict -- but how does that make her different from Trump? Palin is Trump with a vagina. He's dominated the last decade of American politics. She's a has-been.
America is full of Christian conservatives who genuinely don't believe women should hold leadership positions. It's also full of young and not-so-young men who feel disgust when women seem powerful or step into what they regard as male spaces. Hire female leads for a remake of a beloved buddy movie and they wail, "You're ruining my childhood!" How do you get an Angela Merkel or a Jacinta Ardern past all that?
In the days before Barack Obama, I assumed that the first Black president and the first female president would be Republicans. Now I don't think I'll live to see a female president. There are too many trad Christians and too many whiny boy-men -- and they just elected the biggest whiny boy-man of them all."
Here is Trump's enemies list, compiled by Alternet and Politico:
ReplyDelete"Who’s on his list? According to Politico, nearly two dozen individuals, largely in the legal and political spectrum, along with dozens of intelligence specialists, and unnamed journalists:
President Joe Biden
Vice President Kamala Harris
Former President Barack Obama
Former Secretary of State Hillary Clinton
Former Speaker Nancy Pelosi
New York Attorney General Letitia James
Manhattan Justice Arthur Engoron
Former U.S. Rep. Liz Cheney
Special Counsel Jack Smith
Manhattan District Attorney Alvin Bragg
Former Chair of the Joint Chiefs of Staff Mark Milley
Former FBI Director James Comey
Hunter Biden and the rest of the Biden family
Former FBI special agent Peter Strzok
Former FBI attorney Lisa Page
Rep. Adam Schiff (Now Senator-elect Schiff)
Facebook founder Mark Zuckerberg
Former Manhattan Assistant District Attorney Mark Pomerantz
Former Trump attorney Michael Cohen
U.S. Capitol Police Lt. Michael Byrd
Rep. Jamaal Bowman
51 intelligence professionals who signed letter about Hunter Biden laptop
Members of the U.S. House Select Committee on the January 6 Attack
Unspecified people engaged in election fraud
POLITICO reporters, editors and publisher
It’s not just Politico.
Trump has often called the mainstream media, the “enemy of the people.”
In his victory speech early Wednesday morning, Trump referred to the press as, “the enemy camp,” according to The Guardian.
“Introducing his running mate, the Ohio senator JD Vance, Trump said: “I told JD to go into the enemy camp. He just goes: OK. Which one? CNN? MSNBC? He’s like the only guy who looks forward to going on, and then just absolutely obliterates them.”
Somerby, are you happy that you helped to elect a president who will use the power of the presidency to persecute political opponents (and people just doing their jobs too)?