What can we learn from the midterms?
Sign up for the On Point newsletter here.
We in the media, and political punditry, love a good narrative.
And for the past month, the chosen narrative was that a #Redwave would wash over America last night.
Voters had other ideas.
Democrats did indeed hold the line — though the GOP can still make gains in Congress.
Today, On Point: What are America’s voters telling their leaders and the media that neither was able to hear?
Tom Bonier, CEO of the Democratic political data services firm TargetSmart.
Lisa Desjardins, correspondent for PBS NewsHour. (@LisaDNews)
On the different way Americans are talking about democracy
Heather Cox Richardson: “You saw a shift in the American voters, and in the way Americans talked about our democracy, coming for a while now. People talking about democracy, about decency, about caring for each other. And they were talking in places that the pundits weren’t necessarily watching. Gen Z was huge and so were women. And they came to the fore yesterday and they will continue to come to the fore, I think.”
On young people showing up to the polls
Heather Cox Richardson: “What people missed was the huge numbers of students, especially, and young people, Gen Z, who were registering to vote for the first time. And they did not show up in the polls. But for the people watching registrations, the way the CEO of TargetSmart was, that’s why he came across my radar screen.
“He kept saying, I’ve never seen anything like this. And it was shocking to me to the degree to which he was off compared to all the other pundits. So I would read him and think, Is he right? And then I’d look at his numbers and I think, Gee, they sure look right. He did not miss it.”
On redistricting in Michigan
Stephen Henderson: “This is the first election under the new map, of course. Just like in every state, we went through a redistricting process last year like everyone else. Ours was different this time, though. Because in 2018 we passed a constitutional amendment that turned that process over to voters themselves. So we don’t have politicians in Lansing drawing maps in a dark room someplace that none of us see. It was all done out in the open, and people all over the state had a chance to weigh in on it. So it looks really different.
“We all believed that Democrats would have a better shot in at least one House at some point. But I don’t think many people thought that they would get that done in the first cycle. And I don’t know anybody who predicted that it would happen in both houses in the first cycle. You know, the other issue that I think was really animating voters here was the economy. And that cut against Republicans here, I think, because they didn’t present a plan to do things differently.
“I thought they did a wonderful job of pointing out how much people are paying at the pump, and at the grocery store and all of these other places that all of us are feeling that economic pain. But other than saying they would cut taxes, which it was not clear how that would help inflation a whole lot. They didn’t really outline an agenda that was something that people could respond to and connect with. And so voters decided to give Democrats another try.”
Should people exercise some caution in making presumptions about Gen Z?
Heather Cox Richardson: “I don’t think the game is over by any stretch of the imagination, but I do think one of the things that we really don’t talk about enough when we talk about demographics is gerrymandering and voter suppression. Because, again, you look at a place like Michigan, the point of that new map was not to put the Democrats in power, or necessarily to cut the Republicans out of power, but to make those districts competitive again. That actually wasn’t the point. That wasn’t why they started out.
“It turned out that it made all those districts competitive. And what you got was an election that was much more reflective of what people on the ground actually wanted to vote for. So, I think you can’t say, oh, what the Democrats wrote in that famous book didn’t pan out because they got it wrong. So much as they didn’t take a look at the extraordinary measures that the Republican Party would take after starting about 1986 to make sure that they could choose their voters. That’s put us in this cul-de-sac in a sense that is forcing some kind of a reckoning.”
What led those pollsters to project a red wave?
Tom Bonier: “It’s a little bit of everything. And I have to say, I share your discomfort with that. And we should note that the NPR Marist polls are still among some of the best polls out there. So we have those high-quality polls, but they’re being flooded, drowned out by more of these lower quality and questionable partisan polls. And to your point, yeah, sometimes it’s asking questions in a biased way to get a certain result that’s less transparent, because generally these pollsters aren’t sharing a lot of detail. But what we look at is the likely voter sample.
“Every poll is based on a prediction of turnout, which means the pollster has to do something that, frankly, is pretty difficult, especially in an election like this, where you have so many different unprecedented components like the overturn of Roe v. Wade. … And so these likely voter samples, what we do is you say we go under the hood and we look at who are they polling? And I’ll give you an example of this. There is a poll that came out in Pennsylvania right after that debate that got a lot of attention. I think you saw the Republican campaign, Dr. Oz was, you know, frankly at that point, pretty desperate to see the narrative change. He just wasn’t catching up in the polls. They needed to have a narrative that that debate changed everything.
“A Republican poll came out two days later, first warning sign. It was a poll that was conducted over 2 hours and one night. That’s a big no-no in polling. You have to diversify when you’re collecting the data, usually over 3 to 4 days. But the biggest problem when I looked at it was that they had basically eliminated young voters from their likely voter sample. Only 14% of the people they polled were under the age of 40.
“In Pennsylvania, it’s never been below a quarter of the electorate. So, yes, it produced the result that showed Dr. Oz winning by two points. But I looked at that poll and said, look, people are reading this wrong because the electorate is not going to be that old. And if he’s only winning by two points in a poll, that is assuming young people don’t come out at all. I feel really good about Pennsylvania. We’ve seen you know, obviously that’s a race that’s been called with John Fetterman, the winner.”
Pollsters are making decisions to not look at a certain set of potential voters. Do I have that right?
Tom Bonier: “Basically what they’re doing is overemphasizing or de-emphasizing. So generally, pollsters make a decision about what percent of their poll sample, because in the end it’s a sample. People ask and say, how can a poll be accurate if they only talked to 800 people. Statistically if you get a sample that’s representative of the electorate, 800 people is plenty. But it’s that representative of the electorate trick, and that’s where these polls just weren’t represented.
“So they’re making decisions to say, well, when we look at who were polling, we’re just going to have fewer young people in there, or we’ll have an electorate that is just more explicitly Republican. We saw a lot of that. Polls in Georgia that showed more momentum for Herschel Walker than really have materialized in the election results. And similarly, it was just polling an electorate that is much more Republican than we’ve seen in Georgia over the last few election cycles.”
On disinformation in polls
Heather Cox Richardson: “Polling has entered a phase in which it can be compared to disinformation coming from those who are trying to skew voters without actually having reality on their side. They’re using a narrative. And that’s an extension really of what we saw in 2016, 2018, 2020, the construction of a false narrative to affect the way people vote. But it’s going to be interesting now that the mainstream media has gotten so called out on it over this particular election, whether or not they’re going to address that going forward.”
On how the media handles election coverage
Lisa Desjardins: “I think the media needs to hear criticism. I think that makes us better and I think that’s important. But I really have to push back on a lot of what I just heard. I think that this idea, the media from both sides and in this conversation is a very easy, vague sort of bad guy to everything.
“And obviously, every different media outlet is different. You know, PBS NewsHour has a very specific way of handling polls. We really did not put poll results on air saying, we think this candidate will win or that candidate will win, that’s something we avoid. But I will say, I’ve looked at a lot of polls through this cycle. We spend a lot of time considering when they color our thinking.
“And I think the important ingredient you guys are missing in this conversation is that we spend a lot of time talking to Democrats and Republicans who are looking at these polls as well, not the pollsters, but people who are running campaign teams, whether it’s Senate, House or places like the DCCC, which runs the arm of the whole House national effort for Democrats or the same for the Republicans.
“And I have to tell you that through all of those calls. Democrats and Republicans both were telling us that they thought Republicans are going to have a big night. So I think this idea that it’s sort of a media concoction is off. You know, sure, the media plays a role, but we are talking to other people as well.
“This is something that a lot of people felt. Now, I think what’s happening here is a bigger factor in life of 2022, where there is an incredible amount of groupthink in our politics. And sometimes that becomes mob mentality, which is very dangerous. But I think that what was going on from my observation, the way I was conveying the election going into last night, the way I still see the election, is that Republicans were gaining momentum in the last few weeks.
“And to me, it was always a question of it was a horse race, where if you see there’s one horse out in front, which I thought was Democrats, for the most part, race by race is different. But in general, I saw Democrats as out front. But I saw Republicans as that horse that’s coming up fast. And it was always a question of is that horse from the outside going to have enough time, is it going to run out of gas before it’s able to pass the horse in front? And I think sometimes we just generalize as human beings, have a bias toward what’s changing. And I think the fact that Republicans were picking up some momentum was the change.
“And I think that colored a lot of thinking and made Republicans overly enthusiastic, made Democrats overly nervous. I mean, I’ll tell you, I’ve had some of the most sobering calls of my career with Democrats in the past few days. I had a Democrat whose job it is to look at national House races tell me that they thought they could lose 45 seats. That was two days ago.
“You know, so it isn’t just a media figment of imagination. This is something that is based on a lot of people doing the work. Now, maybe everyone clearly is looking at things not the right way, or maybe it’s just a case of, I think a lot of the House polling was right. And … others at NewsHour we’ve talked about this morning.
“These were close races. These were within the margin, three or four point races. And so really what that is, and my colleague John Yang put it well this morning on our editorial call. That’s a tie, you know, so just say that, you know, these were tie races and these races often break one way or the other, as my colleague Amy Walter has said. And in this case, they broke with the Democrats. And I think there’s a lot of interesting reasons for that. … I think concern for democracy really brought out a lot of Democrats in this election.”
This article was originally published on WBUR.org.
Copyright 2022 NPR. To see more, visit https://www.npr.org.