The Exchanges

Q And for a really late night here, John, um, let me ask you the, one of the most basic questions that I've had, which is what did we learn from? And tell me what exactly did people try to fix? Like, what was the thing that everybody got wrong and what changed?

A Well, the biggest thing that polls fixed was how they adjust their polls by education. What we see in polling, no matter how they're conducted, whether online, as we do at SurveyMonkey, or still on the telephone, which most media pollsters do, is you get people with more formal education, To answer those surveys in far greater proportions than you do people with lower education. So the biggest thing you got to do, and look, we always did it. So we weren't in the, among the state pollsters who kind of failed, I think, um, you know, kind of negligently to justify education at all. We always had to justify education, but we've, what we failed to notice in 2016 was there was an increasing gap between those with postgraduate degrees and those with BAs. They've always been both to pro-democratic group. But the gap in 2016 was abnormally large. A fix to our polls, which I just point out, weren't, you know, kind of, were actually standouts in the, in the upper Midwest in terms of showing it as a close race, not clear victories. We broke apart post grads and grads into two distinct categories, and that released kind of about a point and a half of unforced error Uh, in our polling for 2016. So that's fixed. We've used it to good effect. Again, this time around, we weren't showing what all the other national polls were showing. We've had this between a four and six point national lead. Ag…

AI assessment note: “the biggest thing that polls fixed was how they adjust their polls by education.”

Q question for John Cohen. John, um, let's go back to, um, sort of your understanding as you've been measuring different trends. Have you measured Um, people's sympathy towards lockdowns on a state by state level. And then second question is, have you measured people's sympathy to cancel culture at a state by state level? And by the way, you're on mute. So if you want to just take yourself off.

A Yeah, thank you. We have not done anything on cancel culture. We've done a lot on the coronavirus. We've been tracking that actually in three countries since mid February, and we have a state by state look. And what's interesting is we asked the question, like, is this primarily an economic issue or a primary health issue? And those two have been running neck and neck, but the health, you kind of, more people on average say it's a health issue than an economic one. Trump with Trump supporters overwhelmingly say the crisis is one that's financial, not health related. So there's always been that, but it's been like a 45, 55 gap there. So we've been measuring it state by state, but there's a solid core of people and it gets to David's point about why, what are they focused on? What is the, what is affecting them and their, you know, pocketbooks? It is the, you know, lockdowns and the kind of clamping down and what is this economic crisis, not a healthcare one, even though that's what we all say that they should follow.

AI assessment note: “We have not done anything on cancel culture. We've done a lot on the coronavirus.”

Q had effectively a repudiation of the establishment in 2016, and despite everything that's happened over the last four years, we may be on the brink of another repudiation again. If you, you know, where you're there to measure the pulse of what's going on, But less in sort of measurement speak and just more in just plain American English speak. John, what, like, what's going on if this happens again?

A See, you're, you're absolutely right. There's something major. I would also like to caveat it. We are looking once again at if Trump wins, it's because of the electoral college. Like, he is going to lose the popular vote. There's still far more Americans and American voters who voted today and, you know, kind of over the past several weeks who would prefer Joe Biden to be president. So again, we can't characterize with a broad brush the American voting population when this is about effectively, I hate to call it a quirk, but this is about our system of, of vote tallying and the president, you know, but to pull back your point about kind of, there is something major here. The fact that many people, you know, some of us might be friends with can't understand why this isn't a hundred to zero race. Fail to understand that the president's base isn't small. It is, you know, we've had a 44 to 46% approving of his job performance for many years now. Like, he has a completely durable, solid floor. He also has a high ceiling, right? So he was never going to win the popular vote this time around, but he had a chance at that electoral, you know, squeaking out another electoral college win because he's been so stable. You know, this is a president who, You know, kind of, oh, Trump now had an NC. Thank you for the chat window. So I think you're right that we need to understand more about wha…

AI assessment note: “we need to understand more about what is the componentry of that 45%”

Q be debated. Competence versus incompetence. I think what we can debate is he's a vessel and In that, I think that it's incredibly important what's happening irrespective of what happens today, because we were supposed to walk into a landslide. We're not, as you said, we're going to be in a nail biter. What is, what are, what are they using him as a vessel to communicate to everybody else?

A That's a really good question. I mean, some of That will depend on, you know, a closer analysis of the surveys. Ours, where we talk to more than a million voters, and the exit polls that are being conducted by two separate organizations today. But what's the, what are the storylines that come out of the election? You know, one of the things that's being reported early is there's a much tighter Hispanic vote in Florida than many early polls, you know, predicted. How will that play out as we start to get votes, you know, coming out in Texas? How will that be in Arizona? Is Arizona looking positively for Biden and Mark Kelly in the Senate, you know, in Arizona? Is it, is it really Hispanic votes that are driving some Trump strength in these states? Or you kind of, is it the obsession that the news media has had for the past four years around the kind of Trump middle-aged white male voter with less than college education who has been displaced by, you know, technological and industrial trends over the past 20 years? I think it's going to depend on what that voting coalition looks like for Trump, and it's more diverse than I think we've been focusing on for

AI assessment note: “some of That will depend on, you know, a closer analysis of the surveys”